Foraging behavior of two cattle breeds, a whole-year study: II. Spatial distribution by breed and season1 A. Dolev,* Z. Henkin,†‡ A. Brosh,† Y. Yehuda,* E. D. Ungar,‡ A. Shabtay,† and Y. Aharoni†2 *MIGAL– Galilee Technological Center,Qiryat Shemona, P.O. Box 90000, Rosh Pina 12100, Israel; †Beef Cattle Section, Agricultural Research Organization, Newe Ya’ar Research Center, P.O. Box 1021, Ramat Yishay 30095, Israel; and ‡Department of Natural Resources, Agricultural Research Organization– the Volcani Center, P.O. Box 6, Bet Dagan 50250, Israel
ABSTRACT: Spatial distributions of 22 mature largeframed Beefmaster × Simford (BS) cross cows and 16 mature small-framed Baladi (BA) cows were determined. Cows were allocated to the same paddock of a Mediterranean pasture and monitored during 6 consecutive seasons: spring (April 2006), summer (June 2006), autumn (September 2006), winter (February 2007), early spring (March 2007), and summer (June 2007). The locations of the cows were determined at 5-min intervals for 3 to 4 d during each season by using Global Positioning System (GPS) collars. The distances between consecutive locations and the average locations of each breed at each hour of the day in each season were calculated. The Lateral Foraging Index (LFI) was calculated as the ratio between the average distance per day travelled by the cows of each breed and the perimeter of the diurnal itinerary of that breed, which
was calculated from its hourly average locations. The 2 breeds maintained similar diurnal patterns of foraging and resting, characterized by morning and afternoon foraging, and resting at midday and during the night. In summer this pattern was more distinctive, with longer resting periods and more intense foraging periods than in winter, when the noon rest was shorter and only partial. The diurnal routes differed (P < 0.001) between the breeds in all seasons, in their locations, their travelling time, or both. The BA cows were more active than the BS cows in all seasons: they travelled longer distances (P < 0.001) and foraged for more hours during the day (P < 0.001). For both breeds the LFI was numerically greater in winter and spring (February, and March) than in summer (June and September), and it was numerically greater for BA than for BS cows in most seasons.
Key words: activity, Baladi cows, Global Positioning System, Mediterranean pasture, spatial distribution © 2014 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2014.92:758–766 doi:10.2527/jas2013-6996 introduction Spatial distribution, habitat selection and site preference of cattle in rangelands have been studied in the past (Smith et al., 1992; Bailey et al., 1996; Santos et al., 2003), and there are strong indications that in many foraging environments cattle do not uniformly exploit 1Contribution No. 634/13 from the Agricultural Research Organization, P.O. Box 6, Bet Dagan 50250, Israel. Supported by grant US-3694–05 from BARD, the United States–Israel Binational Agricultural Research and Development Fund. 2Corresponding author: [email protected] Received August 6, 2013. Accepted December 6, 2013.
the area available to them (Arnold, 1981; Gillen et al., 1984; Irving et al., 1995). Nevertheless, previous studies provided little information on the site selection of freeranging herds in Mediterranean grasslands, in general, and on possible differences between breeds, in particular. During the last 15 yr, Global Positioning System (GPS) devices were used by many workers (Bailey et al., 2004, 2006; Brosh et al., 2006; Hunt et al., 2007; Henkin et al., 2012) to monitor foraging activity and behavior of cows on pasture. The Mediterranean rangelands of Israel are predominantly foraged by large-breed cattle, but there are also lesser numbers of the smaller Baladi (BA) cow, which is an indigenous breed found throughout the Southern Mediterranean basin (Maule,
758 Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Spatial distribution of two cattle breeds
1990). In addition to its reputation for hardiness and disease resistance, it is believed that 2 features render the BA better adapted to harsh conditions than the heavier beef breeds: better of intake utilization efficiency and lower relative locomotion costs because of their small stature. The objectives of the present study were 1) to compare, on a whole-year basis, the spatial distribution and activity of 2 cattle breeds– BA and Beefmater × Simford (BS)– that differ greatly in body size and weight and 2) to study the diurnal patterns of foraging and the diurnal itineraries of the 2 breeds during the various seasons of the year. The raw GPS data used in the present study were those that were used by Aharoni et al. (2013), with the addition of 1 season (June 2007) that was not included in the earlier study because heat production values of the cows were not available. Materials and Methods Research protocols of this study were approved by the Agricultural Research Organization Institutional Animal Care and Use Committee. Experimental Site The study was conducted at the Karei Deshe experimental farm, located in eastern Galilee in north-eastern Israel (longitude 35°35¢ E; latitude 32°55¢ N; altitude 210 to 310 m above sea level; Gutman et al., 1990, 1999; Aharoni et al., 2009, 2013). The area is hilly, with slopes generally < 10%. The site has a Mediterranean climate, characterized by wet, mild winters and dry, hot summers. Mean annual rainfall is 570 mm, falling mostly from November to March. There are at least 6 mo (May through October) with little or no precipitation. Soils are shallow, brown, basaltic protogrumusols covering basalt rock, and about 30% of the area is covered by exposed basalt boulders (Gutman and Seligman, 1979), within a rich hemicryptophytic grassland (Zohary, 1973) dominated by Hordeum bulbosum L., Echinops spp., Bituminaria bituminosa L., and many annual species (Sternberg et al., 2000). Rainfall during the 2005 to 2006 season was below the yearly average, with only 450 mm, and an unusually long dry period from February to April, followed by a late fall of 30 mm during the second half of April. Rainfall during 2006 to 2007 was 519 mm, closer to the yearly average. Breeds, Stocking Rate Treatments, and Measurement Seasons The locations and spatial distribution of 2 cattle breeds (large-framed BS cross cows [BW 581 ± 76 kg] and small-framed native BA cows [BW 268 ± 60 kg]) were compared during 6 seasons. During 3 seasons (spring Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
759
[April], summer [June], and autumn [September]) in 2006 the comparison was conducted at 2 stocking rates: adjacent paddocks of 22 ha (Paddock 1) and 34 ha (Paddock 2) were assigned to high and low stocking rates of 1.8 and 0.9 cows/ha, respectively. Eight BS cows and 8 BA cows were assigned for measurement in each paddock, and additional cows of both breeds were introduced to the paddocks to reach the designated stocking rates and to serve as measured cows if necessary. The same cows were used in 2007, when the gates between paddocks 1 and 2 were opened to form 1 common paddock with a stocking rate of 1.5 cows/ha, and monitoring continued during 3 additional seasons: winter (February), early spring (March), and summer (June). Totals of 16 BA and 22 BS cows were monitored during the experiment, but the numbers of monitored cows in each season and paddock ranged from 3 to 11 and 2 to 13 for BA and BS cows, respectively. The BS and BA cows foraged together in the same paddock in all 9 season/paddock combinations, and 105,588 location records were collected during the entire experiment. No supplements were added to the paddocks during 5 of the 6 seasons of measurements; dry poultry litter was provided for ad libitum consumption during the September 2006 measurements. Cow Measurements The locations of the cows were determined with the aid of GPS collars. We used 2 alternative devices: 1) Lotek GPS collars of the 2200 and 3300 Series (Lotek Engineering Inc., Newmarket, Ont., Canada); and 2) Trilogical GPS collars (Trilogical, Rishon-LeZion, Israel), especially designed for the present study according to our specifications. In both collars the GPS antenna faced the sky from a horizontal orientation when the cow was upright. The accuracy of both devices was measured by leaving them for several hours on the ground and testing the distribution of the recorded locations. This procedure was repeated with 2 instruments of each device twice: once at the Newe Yaar Research Center and a second time at the experimental field in Karei Deshe. The standard deviation of these records was found to be 7 m for both devices. The GPS devices integrated and stored location data according to global coordinates (global network WGS84) at 5-min (Lotek) or 1-min (Trilogical) intervals; in the latter case we used only the last record of each 5-min group. After downloading from the GPS devices, the longitude and latitude coordinates were transformed to metric Israel Transverse Mercator (ITM) coordinates using ArcView 9.3 software, and then projected on the GIS map of the experimental area. In the first article in this series (Aharoni et al., 2013) the GPS data were integrated with measurements of heart rate (which were transformed to heat production) and activity to compute energy costs of the dif-
760
Dolev et al.
Table 1. List of cows monitored in the study, sorted by breed (BA = Baladi; BS = Beefmaster × Simford cross), together with their site of origin, birth yr, yr of transfer to Karei Deshe (KD yr), relationship, and season and paddock (P1, P2 or united– PU) of cow assignment. The number (n monitored) of cows from each breed in each season/paddock combination is summed. The number (n in paddock) of cows of each breed in the paddocks is presented as well Breed BA
BS
Cow
Origin
Birth yr
KD yr
260 295 484 701 702 703 704 705 706 707 708 709 710
Dalton farm Dalton farm Dalton farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm
1994 1991 2003 2002 2004 2004 2004 2004 2004 2001 2002 2002 2002
2002 2002 2005 2005 2005 2005 2005 2005 2005 2005 2005 2005 2005
711 723 726
Ragar farm Hula reserve Hula reserve
2002 1996 1997
2005 2005 2005
n n
Monitored In paddock
203 205 209 210 213 216 218 223 227 228 230 234 235 237 239 245 249 250 252 257 258 373
Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Manda farm
n n
Monitored In paddock
Apr-06 Relative
P1
P1
P2
Sep-06 P1
P2
Feb-07
Mar-07
Jun-07
PU
PU
PU
*
*
*
*
*
*
*
*
* *
*
* *
* Mother 702 Daughter 701
*
2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003
*
*
*
* * *
* * *
* *
* * *
*
3 10
3 5
3 9
4 6
6 8
* *
* *
*
* *
*
*
* *
*
*
*
* * *
*
10 13
8 13
* * * * *
* * * * * * * * 11 15 * * *
* *
* *
*
*
* * *
* *
* * * * *
*
*
*
*
* * * *
* * *
* *
*
* *
* *
*
4 32
8 35
* 3 10
ferent activities. We included in that analysis only cows that provided data from all 3 devices simultaneously for at least 36 h. For the sake of consistency the same cows were used in the present as in the previous analysis for the 5 seasons that were reported there. However, here we added a sixth season (June 2007), for which heat production records were not available and therefore not reported Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
*
*
3 9 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1995
P2
Jun-06
2 14
3 14
6 16
8 14
*
*
6 17
13 32
previously. Table 1 lists the monitored cows, their breed, origin, family relationships among them, and the seasons and paddocks to which they were assigned. The number of cows of each breed in each season/paddock combination is reported in Table 1.
Spatial distribution of two cattle breeds
761
Data Analysis The average location and its standard error of each breed at each hour of the day were calculated within each season/paddock combination and subjected to 2-way Analysis of Variance (ANOVA; unbalanced design) with cattle breed and hour of the day as independent variables, and location (on either the X or the Y axis) as the dependent variable. These estimates were corrected for cow age by including it as a covariate in the analysis. Distances walked by each cow were calculated from its successive locations, which were recorded at 5-min intervals. These distances were then analyzed within each season/paddock combination. The average 5-min distance and its SE were then multiplied by 12 or by 288 to obtain the distance per hour or per day, respectively. The perimeter of the diurnal itinerary of each breed in each season/paddock combination was calculated as the sum of the distances between average locations of successive hours of the day. This diurnal itinerary was calculated for a 24 h period of a notional day that averages the hourly locations of each breed across all the measurement days in the season. All the analyses were performed with the Genstat-7 software (VSN International, 2003). For each season/paddock combination we defined the Lateral Foraging Index (LFI) as the ratio between the daily distance walked by the cows and the daily perimeter of the diurnal itinerary of each breed. This index represents the ratio between the sum of 24 straight-line distances between successive hourly breed-average locations and the actual distance walked by the cows during those 24 h. RESULTS Diurnal Routes The average location of each breed at each hour of the day in each season/paddock combination is presented in Fig. 1. Diurnal itineraries of both breeds were shorter in winter (February; Fig. 1D) and early spring (March; Fig. 1E) than in other seasons, and shorter in paddock 1 than in paddock 2 in the 3 seasons in 2006, when the paddocks were foraged separately. Cows of both breeds took their noon rest at the southern edge of their diurnal itinerary and their night rest near to its northern edge. It should be noted that in both paddocks the water troughs were located at the southern edge, but cows exhibited the same pattern even in winter, when they did not use these troughs and took their noon rest far from them. The diurnal itineraries of both cattle breeds in all seasons were clockwise, i.e., their movement from the noon rest to the night rest took a westerly route, and their movement back from night rest to noon rest took an easterly route. There were significant differences between the diurnal itineraries of the respective cattle breeds in most of the season/paddock combinations. Table 2 presents the signifiDownloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Figure 1. Daily routes of Beefmaster × Simford cross (BS, solid line) and Baladi (BA, broken line) projected on orthophoto maps of the experimental paddocks (P1 on the right; P2 on the left) in 6 seasons during 2006 and 2007: A. April 2006; B. June 2006; C. September 2006; D. February 2007; E. March 2007; F. June 2007. Cows foraged on separate paddocks in 2006; the gates between the paddocks were opened, to form 1 paddock in the 2007 seasons. The average locations of each breed at 0500, 1100, 1700, and 2300 are indicated to define the direction of daily movement. (box with dot in center) water troughs, (box with ‘x’) gates between the paddocks.
cance of the effects of Breed, Hour, and Breed × Hour interaction for the X and Y axes. The hour effect was highly significant (P < 0.001) for both X and Y axes in all seasons; breed effect was highly significant (P < 0.001) for the X and Y axes in 7 and 8, respectively, of the 9 season/paddock combinations. The effect of the Breed × Hour interaction
P F
< .001 < .001 < .001 < .001 0.39 0.002 < .001 < .001 < .001 23 23 23 23 23 23 23 23 23
df P
< .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 93.7 98.1 48.1 719.3 357.3 2021.3 1922.7 1036.7 3045.4
F df
23 23 23 23 23 23 23 23 23 < .001 < .001 < .001 < .001 0.07 < .001 < .001 < .001 < .001
P F
118.4 198.7 173.4 68.8 3.4 14.2 241.8 456.2 529.6 1 1 1 1 1 1 1 1 1
df P
< .001 < .001 < .001 0.002 < .001 0.12 < .001 < .001 < .001 3.5 6.1 9.6 2 2.3 1.4 5.7 6.7 5.2
df
23 23 23 23 23 23 23 23 23
P F df
23 23 23 23 23 23 23 23 23 < .001 < .001 < .001 < .001 0.325 0.18 < .001 < .001 < .001
P F
1026.1 480.3 60.3 202.6 1 1.8 47.8 300.8 39.3 Jun-07 Sep-06
1+2 1+2 1 2 1 2 1+2 1 2 Jun-06
Paddock Season
Feb-07 Mar-07 Apr-06
df
155.6 161.7 34.7 81.5 551.2 134.6 531.1 683.2 301.7
< .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001
F
Movement Intensity and Distances
1 1 1 1 1 1 1 1 1
Breed × Hour Hour
Y axis
Breed Breed × hour Hour
X axis
was also very significant (P £ 0.002) in 8 of the 9 season/ paddock combinations for both X and Y axes.
2.4 3.1 3.1 6.9 1.1 2.1 19.5 6.8 26.9
Dolev et al.
Breed
Table 2. Age-covariated Effects of breed, hour of day, and the interaction Breed × Hour, on the hourly average location of cows in the various seasons. Parameters of the statistical analysis: df = degrees of freedom, F = F value, P = significance of the difference
762
Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
The velocity (distance per hour) of movement of each cattle breed in each season/paddock combination is presented in Fig. 2. The diurnal pattern of movement of both breeds changed from several peaks of movement in winter to 2 distinct peaks– one in early morning the other in late afternoon– separated by long periods of rest around midday and midnight, in summer. Although the diurnal patterns of the 2 breeds were similar within each season/paddock combination, some distinct differences between them were observed: for example, in June 2007 BA cows began both their morning and afternoon bouts of intense movement earlier than BS cows. Figure 3 presents the distances between average locations of the 2 respective cattle breeds in each season/paddock combination at each hour of the day. The diurnal patterns of these distances changed among seasons: distances were similar throughout the day in winter and early spring; they were larger in the daytime than at night in April; in autumn (September) the pattern reversed, so that nighttime distances were larger than daytime ones. Table 3 presents the daily distances walked by the two breeds, the perimeters of their daily itineraries in each season/paddock combination, and the LFI indexes, each calculated as the daily distance walked divided by the associated perimeter. For both cattle breeds this index tended to be greater in winter and spring, when the herbage was of high or variable quality, and smaller in the summer; across seasons, it tended to be greater for BA than for BS cows. Discussion Seasonal and Diurnal Spatial Distributions After midnight, cows of both breeds moved southwards from the northern ends of their diurnal itineraries, along an eastern path to their midday rest area. From there, after their midday rest, they moved northwards, along a western path, to their night rest area. Although this pattern was repeated in all seasons, we suggest that this clockwise circling was dictated by external factors related to landscape features, and not by internally driven tendencies of the cows; this pattern was not evident when the same BS cows foraged another paddock a year later (A. Brosh, A. Dolev, and Z. Henkin, unpublished data). The cows of the 2 breeds moved in separate groups in all season/paddock combinations (Fig. 1). The routes of the 2 breeds were closer to each other in summer (June and September) than in winter (February and March) but, even when they almost coincided spatially, e.g., in
Spatial distribution of two cattle breeds
763
Figure 2. The hourly average velocity (m/h) of Baladi (BA, dashed line) and Beefmaster × Simford cross (BS, solid line) cows in each season/paddock combination. Standard errors of the velocity are indicated by vertical bars. Seasons and paddocks: a. February 2007, united paddock; b. March 2007, united paddock; c. April 2006, paddock 1; d. April 2006, paddock 2; e. June 2006, paddock 1; f. June 2006, paddock 2; g. June 2007, united paddock; h. September 2006, paddock 1; i. September 2006, paddock 2.
P2 during June and September 2006, they were temporally separated, as indicated by the significant effect of the Breed × Hour interaction on the hourly average locations of the 2 breeds (Table 2). This finding implies that even when the 2 breeds took the same route, they did so at Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
different times. The differing selections of diurnal routes, especially in winter when the herbage was of high quality, might be the result of differing preferences of the 2 breeds for herbage species, of the readiness of the BA cows to spend more time on foraging and to walk longer distances
764
Dolev et al.
Figure 3. The distances between the average breed locations at different hours of the day in each Season × Paddock combination: a. February 2007, united paddock; b. March 2007, united paddock; c. April 2006, paddock 1; d. April 2006, paddock 2; e. June 2006, paddock 1; f. June 2006, paddock 2; g. June 2007, united paddock, h. September 2006, paddock 1; i. September 2006, paddock 2.
than the BS ones so as to select their food more carefully, or both (Aharoni et al., 2013). However, the finding that the 2 groups moved separately even when they selected the same routes suggests that the cows in each group preferred the proximity of cows of their own breed to that of the other breed, even though there were no family relationships among the cows in either group (Table 1). Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Diurnal Movement and Rest In all season/paddock combinations cows of both breeds tended to forage mainly during separate morning and afternoon bouts, and the seasonal changes in the diurnal patterns of movement and rest (Fig. 2) were similar between the 2 breeds, as determined by day length and am-
765
Spatial distribution of two cattle breeds
Table 3. Daily distance walked by the cows, perimeter of the diurnal itinerary, lateral daily index (LFI), and the ratio between LFI of BA and BS cows during each season Perimeter1
Daily distance
LFI2
LFI ratio
Season
Paddock
BA
BS
BA
BS
BA
BS
BA/BS
Feb-07 Mar-07 Apr-06
1+2 1+2 1 2 1 2 1+2 1 2
3,923 4,712 4,311 4,885 3,626 3,943 4,939 4,355 4,032
3,234 3,830 2,995 3,352 2,696 4,067 4,637 3,315 3,413
847 1,203 575 1,616 1,351 1,967 1,948 1,487 1,689
678 1,106 711 1,419 1,249 1,824 2,229 1,193 1,871
4.630 3.917 7.499 3.022 2.684 2.004 2.536 2.928 2.387
4.769 3.464 4.215 2.363 2.158 2.229 2.081 2.778 1.824
0.971 1.131 1.779 1.279 1.244 0.899 1.219 1.054 1.309
4,303 464
3,504 587
1,409 473
1,364 527
3.512 1.698
2.876 1.041
1.209 0.256
Jun-06 Jun-07 Sep-06
Mean SD
1 Sum of the distances between the average hourly breed locations in the diurnal itinerary. 2 The daily distance divided by the perimeter.
bient temperature: in winter and early spring (February and March) the morning and afternoon foraging bouts began at about 0500 and about 1400, respectively, but there was only partial resting during the intervening midday hours, i.e., at any given moment during the rest period, some cows of each breed probably were foraging. The night rest in the winter was also not strictly synchronized among animals; at any given hour some movement was recorded. During June (longest day) the morning foraging bout began at about 0300, i.e., about 1.5 h before sunrise, the afternoon bout at about 1600, and the cows of both breeds rested near the water troughs from 0700 to 1500. The night rest of both breeds occurred, away from the water troughs, between 2100 and 0300, with very little movement during this period. In conclusion: the bimodal movement pattern was much more distinctive in summer than in winter, and an intermediate pattern was recorded in April. Generally, although the diurnal patterns of the 2 breeds were similar within seasons, there were some differences, and in most season/paddock combinations the BA cows were more active than the BS cows: in paddock 1 in June, during both 2006 and 2007, BA cows began the morning foraging before BS cows– an observation that could account for the significant effect of the Breed × Hour interaction on the average hourly location of cows (Table 2), even though the diurnal itineraries of the 2 breeds in this season were almost identical (Fig. 1). The site that BA cows selected for the night rest was far (at least 100 m) from that of BS cows during 5 out of the 9 season/paddock combinations (Fig. 3). It is possible that cows of the 2 breeds selected different resting sites because of their differing body sizes or temperaments, but it is also possible that BA cows preferred to rest away from the bigger BS cows. For both breeds the midday rest location during all seasons except winter was dictated by their requireDownloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
ment for drinking water, therefore, both breeds took their midday rest at the same location, i.e., by the water troughs. Spatial Interrelationships between Breeds There is a lack of information regarding spatial relationships between bigger and smaller cows in grazing paddocks. Some relevant knowledge could be inferred from a study by Šárová et al. (2010), who examined the behavior of dominant and subordinate animals within a group of beef cows on pasture; they reported that dominant cows were closer to the front of the herd during both travelling and foraging, and that their pathway from one eating station to the next during foraging was shorter than that of subordinate cows. These findings suggest that subordinate cows preferred to move behind the dominant cows while travelling and foraging, and to get to the desired eating patches in the paddock in a more indirect way, to avoid confrontation with the stronger cows. Rook et al. (2004) suggested that grazing livestock could be considered to affect biodiversity in pastures by the creation and maintenance of sward structural heterogeneity, particularly as a result of dietary choice; they attributed this effect mainly to differences in body size. Aharoni et al. (2009) also suggested that BA cows were more selective in their foraging than BS cows when the herbage quality in the paddock was varied. If we assume, with regard to the present study, that the BS cows took a dominant position over the BA cows because of their greater weight, then it also could be suggested that the BA cows selected a movement strategy that minimized confrontation with the BS cows; by preferring different routes from the BS cows because of selectivity differences the BA cows avoided such confrontation. However, when the 2 breeds moved along
766
Dolev et al.
similar routes, the BA animals preferred to start their dawn and dusk foraging periods earlier than the bigger BS cows, rather than moving behind them, and thereby were able to maintain a safe distance between the groups, without compromising their direct access to desired eating patches. The advantage of this strategy is reflected also in the ratio between the respective breeds’ LFI (Table 3), which was close to 1 in some of the summer seasons, when the 2 breeds selected similar diurnal routes (Fig. 1). This ratio was also close to 1 in February, but in that season the 2 breeds selected different diurnal itineraries, therefore the movement of the BS cows did not threaten the BA cows. The BA cows’ strategy of keeping a safe distance from the BS cows can also be observed in the selection of noon and night resting sites in different seasons: in winter and early spring, when the cows did not need to drink frequently, the 2 breeds used different rest sites both at midday and at night, but in the summer BA cows preferred to rest near the water troughs at midday even though the BS cows rested at the same site, whereas for the night rest the BA cows selected different sites from the BS cows. If maintaining a safe distance from the BS cows was a substantial consideration in the BA cows’ strategy, then their LFI might be smaller during some seasons if BS cows were not present in the same paddock. Conclusions The small-framed BA cows exploited their relative advantage of lower specific costs of activity (Aharoni et al., 2013), especially when the weather was hot and herbage quality was low; they selected diurnal movement itineraries that were either different from those selected by the BS cows, or spatially similar but temporally separated. It is suggested that BA cows were more selective in seasons in which the quality of the herbage in the paddock was highly variable and, moreover, that the small-framed BA cows were more adaptable than the large-framed BS cows to hot weather conditions and poor-quality herbage. LITERATURE CITED Aharoni, Y., A. Dolev, Z. Henkin, Y. Yehuda, A. Ezra, E. D. Ungar, A. Shabtay, and A. Brosh. 2013. Foraging behavior of two cattle breeds, a whole-year study: I. Heat production, activity, and energy costs. J. Anim. Sci. 91:1381–1390. Aharoni, Y., Z. Henkin, A. Ezra, A. Dolev, A. Shabtay, A. Orlov, Y. Yehuda, and A. Brosh. 2009. Grazing behavior and energy costs of activity: A comparison between two types of cattle. J. Anim. Sci. 87:2719–2731. Arnold, G. W. 1981. Grazing behaviour. In: F.H.W. Morley, editor, Grazing Animals. Elsevier, Amsterdam, Oxford, New York. pp. 79–104.
Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Bailey, D. W., J. E. Gross, E. A. Laca, L. R. Rittenhouse, M. B. Coughenour, D. M. Swift, and P. L. Sims. 1996. Mechanisms that result in large herbivore grazing distribution patterns. J. Range Manage. 49:386–400. Bailey, D. W., M. R. Keil, and L. R. Rittenhouse. 2004. Research observation: Daily movement patterns of hill climbing and bottom dwelling cows. J. Range Manage. 57:20–28. Bailey, D. W., H. C. Van Wagoner, and R. Weinmeister. 2006. Individual animal selection has the potential to improve uniformity of grazing on foothill rangeland. Rangeland Ecol. Manag. 59:351–358. Brosh, A., Z. Henkin, E. D. Ungar, A. Dolev, A. Orlov, Y. Yehuda, and Y. Aharoni. 2006. Energy cost of cows’ grazing activity: Use of the heart rate method and the Global Positioning System for direct field estimation. J. Anim. Sci. 84:1951–1967. Gillen, R. L., W. C. Krueger, and R. F. Miller. 1984. Cattle distribution on mountain rangeland in northern Oregon. J. Range Manage. 37:549–553. Gutman, M., Z. Holzer, H. Baram, I. Noy-Meir, and N.G. Seligman. 1999. Heavy stocking and early-season deferment of grazing on Mediterranean-type grassland. J. Range Manage. 52:590–599. Gutman, M., Z. Holzer, N. G. Seligman, and I. Noy-Meir. 1990. Stocking density and production of a supplemented beef herd grazing yearlong on Mediterranean grassland. J. Range Manage. 43:535–539. Gutman, M., and N. G. Seligman. 1979. Grazing management of Mediterranean foothill range in the Upper Jordan River Valley. J. Range Manage. 32:86–92. Henkin, Z., E. D. Ungar, and A. Dolev. 2012. Foraging behaviour of beef cattle in the hilly terrain of a Mediterranean grassland. Rangeland J. 34:163–172. Hunt, L. P., S. Petty, R. Cowley, A. Fisher, A. J. Ash, and N. MacDonald. 2007. Factors affecting the management of cattle grazing distribution in northern Australia: Preliminary observations on the effect of paddock size and water points. Rangeland J. 29:169–179. Irving, B. D., P. L. Rutledge, A. W. Bailey, M. A. Neath, and D. S. Chanasyk. 1995. Grass utilization and grazing distribution within intensively managed fields in central Alberta. J. Range Manage. 48:358– 361. Maule, J. P. 1990. The Cattle of the Tropics. Centre for Tropical Veterinary Medicine. University of Edinburgh, Edinburgh, UK. Rook, A. J., B. Dumont, J. Isselstein, K. Osoro, M. F. WallisDeVries, G. Parente, and J. Mills. 2004. Matching type of livestock to desired biodiversity outcomes in pastures– a review. Biol. Conserv. 119:137–150. Santos, S. A., C. Costa, G. Silva e Souza, S. M. A. Crispim, R. A. Pearson, and R. Gutierrez. 2003. Foraging strategy of cattle in the Pantanal rangeland, Nhecolandia sub-region, Brazil. In: N. Allsopp, A. R. Palmer, S. J. Milton, G. I. H. Kerley, K. P. Kirkman, R. Hurt and C. Brown, editors, Proc. 7th International Rangeland Congress. 26th July-1st August 2003, Durban, South Africa. Document Transformation Technologies, Irene, South Africa. pp. 23–30. Šárová, R., M. Špinka, J. L. A. Panamá, and P. Šimeček. 2010. Graded leadership by dominant animals in a herd of female beef cattle on pasture. Anim. Behav. 79(5):1037–1045. Smith, M. A., J. D. Rodgers, J. L. Dodd, and Q. D. Skinner. 1992. Habitat selection by cattle along an ephemeral channel. J. Range Manage. 45:385–390. Sternberg, M., M. Gutman, A. Perevolotsky, E. D. Ungar, and J. Kigel. 2000. Vegetation response to grazing management in a Mediterranean herbaceous community: A functional group approach. J. Appl. Ecol. 37:224–237. V. S. N. International. 2003. Genstat for Windows. 7th ed. VSN International, Oxford, UK. Zohary, M. 1973. Geobotanical foundations of the Middle East. Gustav Fischer Verlag, Stuttgart, and Swets and Zeitlinger, Amsterdam, the Netherlands.
ABSTRACT: Spatial distributions of 22 mature largeframed Beefmaster × Simford (BS) cross cows and 16 mature small-framed Baladi (BA) cows were determined. Cows were allocated to the same paddock of a Mediterranean pasture and monitored during 6 consecutive seasons: spring (April 2006), summer (June 2006), autumn (September 2006), winter (February 2007), early spring (March 2007), and summer (June 2007). The locations of the cows were determined at 5-min intervals for 3 to 4 d during each season by using Global Positioning System (GPS) collars. The distances between consecutive locations and the average locations of each breed at each hour of the day in each season were calculated. The Lateral Foraging Index (LFI) was calculated as the ratio between the average distance per day travelled by the cows of each breed and the perimeter of the diurnal itinerary of that breed, which
was calculated from its hourly average locations. The 2 breeds maintained similar diurnal patterns of foraging and resting, characterized by morning and afternoon foraging, and resting at midday and during the night. In summer this pattern was more distinctive, with longer resting periods and more intense foraging periods than in winter, when the noon rest was shorter and only partial. The diurnal routes differed (P < 0.001) between the breeds in all seasons, in their locations, their travelling time, or both. The BA cows were more active than the BS cows in all seasons: they travelled longer distances (P < 0.001) and foraged for more hours during the day (P < 0.001). For both breeds the LFI was numerically greater in winter and spring (February, and March) than in summer (June and September), and it was numerically greater for BA than for BS cows in most seasons.
Key words: activity, Baladi cows, Global Positioning System, Mediterranean pasture, spatial distribution © 2014 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2014.92:758–766 doi:10.2527/jas2013-6996 introduction Spatial distribution, habitat selection and site preference of cattle in rangelands have been studied in the past (Smith et al., 1992; Bailey et al., 1996; Santos et al., 2003), and there are strong indications that in many foraging environments cattle do not uniformly exploit 1Contribution No. 634/13 from the Agricultural Research Organization, P.O. Box 6, Bet Dagan 50250, Israel. Supported by grant US-3694–05 from BARD, the United States–Israel Binational Agricultural Research and Development Fund. 2Corresponding author: [email protected] Received August 6, 2013. Accepted December 6, 2013.
the area available to them (Arnold, 1981; Gillen et al., 1984; Irving et al., 1995). Nevertheless, previous studies provided little information on the site selection of freeranging herds in Mediterranean grasslands, in general, and on possible differences between breeds, in particular. During the last 15 yr, Global Positioning System (GPS) devices were used by many workers (Bailey et al., 2004, 2006; Brosh et al., 2006; Hunt et al., 2007; Henkin et al., 2012) to monitor foraging activity and behavior of cows on pasture. The Mediterranean rangelands of Israel are predominantly foraged by large-breed cattle, but there are also lesser numbers of the smaller Baladi (BA) cow, which is an indigenous breed found throughout the Southern Mediterranean basin (Maule,
758 Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Spatial distribution of two cattle breeds
1990). In addition to its reputation for hardiness and disease resistance, it is believed that 2 features render the BA better adapted to harsh conditions than the heavier beef breeds: better of intake utilization efficiency and lower relative locomotion costs because of their small stature. The objectives of the present study were 1) to compare, on a whole-year basis, the spatial distribution and activity of 2 cattle breeds– BA and Beefmater × Simford (BS)– that differ greatly in body size and weight and 2) to study the diurnal patterns of foraging and the diurnal itineraries of the 2 breeds during the various seasons of the year. The raw GPS data used in the present study were those that were used by Aharoni et al. (2013), with the addition of 1 season (June 2007) that was not included in the earlier study because heat production values of the cows were not available. Materials and Methods Research protocols of this study were approved by the Agricultural Research Organization Institutional Animal Care and Use Committee. Experimental Site The study was conducted at the Karei Deshe experimental farm, located in eastern Galilee in north-eastern Israel (longitude 35°35¢ E; latitude 32°55¢ N; altitude 210 to 310 m above sea level; Gutman et al., 1990, 1999; Aharoni et al., 2009, 2013). The area is hilly, with slopes generally < 10%. The site has a Mediterranean climate, characterized by wet, mild winters and dry, hot summers. Mean annual rainfall is 570 mm, falling mostly from November to March. There are at least 6 mo (May through October) with little or no precipitation. Soils are shallow, brown, basaltic protogrumusols covering basalt rock, and about 30% of the area is covered by exposed basalt boulders (Gutman and Seligman, 1979), within a rich hemicryptophytic grassland (Zohary, 1973) dominated by Hordeum bulbosum L., Echinops spp., Bituminaria bituminosa L., and many annual species (Sternberg et al., 2000). Rainfall during the 2005 to 2006 season was below the yearly average, with only 450 mm, and an unusually long dry period from February to April, followed by a late fall of 30 mm during the second half of April. Rainfall during 2006 to 2007 was 519 mm, closer to the yearly average. Breeds, Stocking Rate Treatments, and Measurement Seasons The locations and spatial distribution of 2 cattle breeds (large-framed BS cross cows [BW 581 ± 76 kg] and small-framed native BA cows [BW 268 ± 60 kg]) were compared during 6 seasons. During 3 seasons (spring Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
759
[April], summer [June], and autumn [September]) in 2006 the comparison was conducted at 2 stocking rates: adjacent paddocks of 22 ha (Paddock 1) and 34 ha (Paddock 2) were assigned to high and low stocking rates of 1.8 and 0.9 cows/ha, respectively. Eight BS cows and 8 BA cows were assigned for measurement in each paddock, and additional cows of both breeds were introduced to the paddocks to reach the designated stocking rates and to serve as measured cows if necessary. The same cows were used in 2007, when the gates between paddocks 1 and 2 were opened to form 1 common paddock with a stocking rate of 1.5 cows/ha, and monitoring continued during 3 additional seasons: winter (February), early spring (March), and summer (June). Totals of 16 BA and 22 BS cows were monitored during the experiment, but the numbers of monitored cows in each season and paddock ranged from 3 to 11 and 2 to 13 for BA and BS cows, respectively. The BS and BA cows foraged together in the same paddock in all 9 season/paddock combinations, and 105,588 location records were collected during the entire experiment. No supplements were added to the paddocks during 5 of the 6 seasons of measurements; dry poultry litter was provided for ad libitum consumption during the September 2006 measurements. Cow Measurements The locations of the cows were determined with the aid of GPS collars. We used 2 alternative devices: 1) Lotek GPS collars of the 2200 and 3300 Series (Lotek Engineering Inc., Newmarket, Ont., Canada); and 2) Trilogical GPS collars (Trilogical, Rishon-LeZion, Israel), especially designed for the present study according to our specifications. In both collars the GPS antenna faced the sky from a horizontal orientation when the cow was upright. The accuracy of both devices was measured by leaving them for several hours on the ground and testing the distribution of the recorded locations. This procedure was repeated with 2 instruments of each device twice: once at the Newe Yaar Research Center and a second time at the experimental field in Karei Deshe. The standard deviation of these records was found to be 7 m for both devices. The GPS devices integrated and stored location data according to global coordinates (global network WGS84) at 5-min (Lotek) or 1-min (Trilogical) intervals; in the latter case we used only the last record of each 5-min group. After downloading from the GPS devices, the longitude and latitude coordinates were transformed to metric Israel Transverse Mercator (ITM) coordinates using ArcView 9.3 software, and then projected on the GIS map of the experimental area. In the first article in this series (Aharoni et al., 2013) the GPS data were integrated with measurements of heart rate (which were transformed to heat production) and activity to compute energy costs of the dif-
760
Dolev et al.
Table 1. List of cows monitored in the study, sorted by breed (BA = Baladi; BS = Beefmaster × Simford cross), together with their site of origin, birth yr, yr of transfer to Karei Deshe (KD yr), relationship, and season and paddock (P1, P2 or united– PU) of cow assignment. The number (n monitored) of cows from each breed in each season/paddock combination is summed. The number (n in paddock) of cows of each breed in the paddocks is presented as well Breed BA
BS
Cow
Origin
Birth yr
KD yr
260 295 484 701 702 703 704 705 706 707 708 709 710
Dalton farm Dalton farm Dalton farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm Ragar farm
1994 1991 2003 2002 2004 2004 2004 2004 2004 2001 2002 2002 2002
2002 2002 2005 2005 2005 2005 2005 2005 2005 2005 2005 2005 2005
711 723 726
Ragar farm Hula reserve Hula reserve
2002 1996 1997
2005 2005 2005
n n
Monitored In paddock
203 205 209 210 213 216 218 223 227 228 230 234 235 237 239 245 249 250 252 257 258 373
Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Ela farm Manda farm
n n
Monitored In paddock
Apr-06 Relative
P1
P1
P2
Sep-06 P1
P2
Feb-07
Mar-07
Jun-07
PU
PU
PU
*
*
*
*
*
*
*
*
* *
*
* *
* Mother 702 Daughter 701
*
2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003
*
*
*
* * *
* * *
* *
* * *
*
3 10
3 5
3 9
4 6
6 8
* *
* *
*
* *
*
*
* *
*
*
*
* * *
*
10 13
8 13
* * * * *
* * * * * * * * 11 15 * * *
* *
* *
*
*
* * *
* *
* * * * *
*
*
*
*
* * * *
* * *
* *
*
* *
* *
*
4 32
8 35
* 3 10
ferent activities. We included in that analysis only cows that provided data from all 3 devices simultaneously for at least 36 h. For the sake of consistency the same cows were used in the present as in the previous analysis for the 5 seasons that were reported there. However, here we added a sixth season (June 2007), for which heat production records were not available and therefore not reported Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
*
*
3 9 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1998 1995
P2
Jun-06
2 14
3 14
6 16
8 14
*
*
6 17
13 32
previously. Table 1 lists the monitored cows, their breed, origin, family relationships among them, and the seasons and paddocks to which they were assigned. The number of cows of each breed in each season/paddock combination is reported in Table 1.
Spatial distribution of two cattle breeds
761
Data Analysis The average location and its standard error of each breed at each hour of the day were calculated within each season/paddock combination and subjected to 2-way Analysis of Variance (ANOVA; unbalanced design) with cattle breed and hour of the day as independent variables, and location (on either the X or the Y axis) as the dependent variable. These estimates were corrected for cow age by including it as a covariate in the analysis. Distances walked by each cow were calculated from its successive locations, which were recorded at 5-min intervals. These distances were then analyzed within each season/paddock combination. The average 5-min distance and its SE were then multiplied by 12 or by 288 to obtain the distance per hour or per day, respectively. The perimeter of the diurnal itinerary of each breed in each season/paddock combination was calculated as the sum of the distances between average locations of successive hours of the day. This diurnal itinerary was calculated for a 24 h period of a notional day that averages the hourly locations of each breed across all the measurement days in the season. All the analyses were performed with the Genstat-7 software (VSN International, 2003). For each season/paddock combination we defined the Lateral Foraging Index (LFI) as the ratio between the daily distance walked by the cows and the daily perimeter of the diurnal itinerary of each breed. This index represents the ratio between the sum of 24 straight-line distances between successive hourly breed-average locations and the actual distance walked by the cows during those 24 h. RESULTS Diurnal Routes The average location of each breed at each hour of the day in each season/paddock combination is presented in Fig. 1. Diurnal itineraries of both breeds were shorter in winter (February; Fig. 1D) and early spring (March; Fig. 1E) than in other seasons, and shorter in paddock 1 than in paddock 2 in the 3 seasons in 2006, when the paddocks were foraged separately. Cows of both breeds took their noon rest at the southern edge of their diurnal itinerary and their night rest near to its northern edge. It should be noted that in both paddocks the water troughs were located at the southern edge, but cows exhibited the same pattern even in winter, when they did not use these troughs and took their noon rest far from them. The diurnal itineraries of both cattle breeds in all seasons were clockwise, i.e., their movement from the noon rest to the night rest took a westerly route, and their movement back from night rest to noon rest took an easterly route. There were significant differences between the diurnal itineraries of the respective cattle breeds in most of the season/paddock combinations. Table 2 presents the signifiDownloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Figure 1. Daily routes of Beefmaster × Simford cross (BS, solid line) and Baladi (BA, broken line) projected on orthophoto maps of the experimental paddocks (P1 on the right; P2 on the left) in 6 seasons during 2006 and 2007: A. April 2006; B. June 2006; C. September 2006; D. February 2007; E. March 2007; F. June 2007. Cows foraged on separate paddocks in 2006; the gates between the paddocks were opened, to form 1 paddock in the 2007 seasons. The average locations of each breed at 0500, 1100, 1700, and 2300 are indicated to define the direction of daily movement. (box with dot in center) water troughs, (box with ‘x’) gates between the paddocks.
cance of the effects of Breed, Hour, and Breed × Hour interaction for the X and Y axes. The hour effect was highly significant (P < 0.001) for both X and Y axes in all seasons; breed effect was highly significant (P < 0.001) for the X and Y axes in 7 and 8, respectively, of the 9 season/paddock combinations. The effect of the Breed × Hour interaction
P F
< .001 < .001 < .001 < .001 0.39 0.002 < .001 < .001 < .001 23 23 23 23 23 23 23 23 23
df P
< .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 93.7 98.1 48.1 719.3 357.3 2021.3 1922.7 1036.7 3045.4
F df
23 23 23 23 23 23 23 23 23 < .001 < .001 < .001 < .001 0.07 < .001 < .001 < .001 < .001
P F
118.4 198.7 173.4 68.8 3.4 14.2 241.8 456.2 529.6 1 1 1 1 1 1 1 1 1
df P
< .001 < .001 < .001 0.002 < .001 0.12 < .001 < .001 < .001 3.5 6.1 9.6 2 2.3 1.4 5.7 6.7 5.2
df
23 23 23 23 23 23 23 23 23
P F df
23 23 23 23 23 23 23 23 23 < .001 < .001 < .001 < .001 0.325 0.18 < .001 < .001 < .001
P F
1026.1 480.3 60.3 202.6 1 1.8 47.8 300.8 39.3 Jun-07 Sep-06
1+2 1+2 1 2 1 2 1+2 1 2 Jun-06
Paddock Season
Feb-07 Mar-07 Apr-06
df
155.6 161.7 34.7 81.5 551.2 134.6 531.1 683.2 301.7
< .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001 < .001
F
Movement Intensity and Distances
1 1 1 1 1 1 1 1 1
Breed × Hour Hour
Y axis
Breed Breed × hour Hour
X axis
was also very significant (P £ 0.002) in 8 of the 9 season/ paddock combinations for both X and Y axes.
2.4 3.1 3.1 6.9 1.1 2.1 19.5 6.8 26.9
Dolev et al.
Breed
Table 2. Age-covariated Effects of breed, hour of day, and the interaction Breed × Hour, on the hourly average location of cows in the various seasons. Parameters of the statistical analysis: df = degrees of freedom, F = F value, P = significance of the difference
762
Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
The velocity (distance per hour) of movement of each cattle breed in each season/paddock combination is presented in Fig. 2. The diurnal pattern of movement of both breeds changed from several peaks of movement in winter to 2 distinct peaks– one in early morning the other in late afternoon– separated by long periods of rest around midday and midnight, in summer. Although the diurnal patterns of the 2 breeds were similar within each season/paddock combination, some distinct differences between them were observed: for example, in June 2007 BA cows began both their morning and afternoon bouts of intense movement earlier than BS cows. Figure 3 presents the distances between average locations of the 2 respective cattle breeds in each season/paddock combination at each hour of the day. The diurnal patterns of these distances changed among seasons: distances were similar throughout the day in winter and early spring; they were larger in the daytime than at night in April; in autumn (September) the pattern reversed, so that nighttime distances were larger than daytime ones. Table 3 presents the daily distances walked by the two breeds, the perimeters of their daily itineraries in each season/paddock combination, and the LFI indexes, each calculated as the daily distance walked divided by the associated perimeter. For both cattle breeds this index tended to be greater in winter and spring, when the herbage was of high or variable quality, and smaller in the summer; across seasons, it tended to be greater for BA than for BS cows. Discussion Seasonal and Diurnal Spatial Distributions After midnight, cows of both breeds moved southwards from the northern ends of their diurnal itineraries, along an eastern path to their midday rest area. From there, after their midday rest, they moved northwards, along a western path, to their night rest area. Although this pattern was repeated in all seasons, we suggest that this clockwise circling was dictated by external factors related to landscape features, and not by internally driven tendencies of the cows; this pattern was not evident when the same BS cows foraged another paddock a year later (A. Brosh, A. Dolev, and Z. Henkin, unpublished data). The cows of the 2 breeds moved in separate groups in all season/paddock combinations (Fig. 1). The routes of the 2 breeds were closer to each other in summer (June and September) than in winter (February and March) but, even when they almost coincided spatially, e.g., in
Spatial distribution of two cattle breeds
763
Figure 2. The hourly average velocity (m/h) of Baladi (BA, dashed line) and Beefmaster × Simford cross (BS, solid line) cows in each season/paddock combination. Standard errors of the velocity are indicated by vertical bars. Seasons and paddocks: a. February 2007, united paddock; b. March 2007, united paddock; c. April 2006, paddock 1; d. April 2006, paddock 2; e. June 2006, paddock 1; f. June 2006, paddock 2; g. June 2007, united paddock; h. September 2006, paddock 1; i. September 2006, paddock 2.
P2 during June and September 2006, they were temporally separated, as indicated by the significant effect of the Breed × Hour interaction on the hourly average locations of the 2 breeds (Table 2). This finding implies that even when the 2 breeds took the same route, they did so at Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
different times. The differing selections of diurnal routes, especially in winter when the herbage was of high quality, might be the result of differing preferences of the 2 breeds for herbage species, of the readiness of the BA cows to spend more time on foraging and to walk longer distances
764
Dolev et al.
Figure 3. The distances between the average breed locations at different hours of the day in each Season × Paddock combination: a. February 2007, united paddock; b. March 2007, united paddock; c. April 2006, paddock 1; d. April 2006, paddock 2; e. June 2006, paddock 1; f. June 2006, paddock 2; g. June 2007, united paddock, h. September 2006, paddock 1; i. September 2006, paddock 2.
than the BS ones so as to select their food more carefully, or both (Aharoni et al., 2013). However, the finding that the 2 groups moved separately even when they selected the same routes suggests that the cows in each group preferred the proximity of cows of their own breed to that of the other breed, even though there were no family relationships among the cows in either group (Table 1). Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Diurnal Movement and Rest In all season/paddock combinations cows of both breeds tended to forage mainly during separate morning and afternoon bouts, and the seasonal changes in the diurnal patterns of movement and rest (Fig. 2) were similar between the 2 breeds, as determined by day length and am-
765
Spatial distribution of two cattle breeds
Table 3. Daily distance walked by the cows, perimeter of the diurnal itinerary, lateral daily index (LFI), and the ratio between LFI of BA and BS cows during each season Perimeter1
Daily distance
LFI2
LFI ratio
Season
Paddock
BA
BS
BA
BS
BA
BS
BA/BS
Feb-07 Mar-07 Apr-06
1+2 1+2 1 2 1 2 1+2 1 2
3,923 4,712 4,311 4,885 3,626 3,943 4,939 4,355 4,032
3,234 3,830 2,995 3,352 2,696 4,067 4,637 3,315 3,413
847 1,203 575 1,616 1,351 1,967 1,948 1,487 1,689
678 1,106 711 1,419 1,249 1,824 2,229 1,193 1,871
4.630 3.917 7.499 3.022 2.684 2.004 2.536 2.928 2.387
4.769 3.464 4.215 2.363 2.158 2.229 2.081 2.778 1.824
0.971 1.131 1.779 1.279 1.244 0.899 1.219 1.054 1.309
4,303 464
3,504 587
1,409 473
1,364 527
3.512 1.698
2.876 1.041
1.209 0.256
Jun-06 Jun-07 Sep-06
Mean SD
1 Sum of the distances between the average hourly breed locations in the diurnal itinerary. 2 The daily distance divided by the perimeter.
bient temperature: in winter and early spring (February and March) the morning and afternoon foraging bouts began at about 0500 and about 1400, respectively, but there was only partial resting during the intervening midday hours, i.e., at any given moment during the rest period, some cows of each breed probably were foraging. The night rest in the winter was also not strictly synchronized among animals; at any given hour some movement was recorded. During June (longest day) the morning foraging bout began at about 0300, i.e., about 1.5 h before sunrise, the afternoon bout at about 1600, and the cows of both breeds rested near the water troughs from 0700 to 1500. The night rest of both breeds occurred, away from the water troughs, between 2100 and 0300, with very little movement during this period. In conclusion: the bimodal movement pattern was much more distinctive in summer than in winter, and an intermediate pattern was recorded in April. Generally, although the diurnal patterns of the 2 breeds were similar within seasons, there were some differences, and in most season/paddock combinations the BA cows were more active than the BS cows: in paddock 1 in June, during both 2006 and 2007, BA cows began the morning foraging before BS cows– an observation that could account for the significant effect of the Breed × Hour interaction on the average hourly location of cows (Table 2), even though the diurnal itineraries of the 2 breeds in this season were almost identical (Fig. 1). The site that BA cows selected for the night rest was far (at least 100 m) from that of BS cows during 5 out of the 9 season/paddock combinations (Fig. 3). It is possible that cows of the 2 breeds selected different resting sites because of their differing body sizes or temperaments, but it is also possible that BA cows preferred to rest away from the bigger BS cows. For both breeds the midday rest location during all seasons except winter was dictated by their requireDownloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
ment for drinking water, therefore, both breeds took their midday rest at the same location, i.e., by the water troughs. Spatial Interrelationships between Breeds There is a lack of information regarding spatial relationships between bigger and smaller cows in grazing paddocks. Some relevant knowledge could be inferred from a study by Šárová et al. (2010), who examined the behavior of dominant and subordinate animals within a group of beef cows on pasture; they reported that dominant cows were closer to the front of the herd during both travelling and foraging, and that their pathway from one eating station to the next during foraging was shorter than that of subordinate cows. These findings suggest that subordinate cows preferred to move behind the dominant cows while travelling and foraging, and to get to the desired eating patches in the paddock in a more indirect way, to avoid confrontation with the stronger cows. Rook et al. (2004) suggested that grazing livestock could be considered to affect biodiversity in pastures by the creation and maintenance of sward structural heterogeneity, particularly as a result of dietary choice; they attributed this effect mainly to differences in body size. Aharoni et al. (2009) also suggested that BA cows were more selective in their foraging than BS cows when the herbage quality in the paddock was varied. If we assume, with regard to the present study, that the BS cows took a dominant position over the BA cows because of their greater weight, then it also could be suggested that the BA cows selected a movement strategy that minimized confrontation with the BS cows; by preferring different routes from the BS cows because of selectivity differences the BA cows avoided such confrontation. However, when the 2 breeds moved along
766
Dolev et al.
similar routes, the BA animals preferred to start their dawn and dusk foraging periods earlier than the bigger BS cows, rather than moving behind them, and thereby were able to maintain a safe distance between the groups, without compromising their direct access to desired eating patches. The advantage of this strategy is reflected also in the ratio between the respective breeds’ LFI (Table 3), which was close to 1 in some of the summer seasons, when the 2 breeds selected similar diurnal routes (Fig. 1). This ratio was also close to 1 in February, but in that season the 2 breeds selected different diurnal itineraries, therefore the movement of the BS cows did not threaten the BA cows. The BA cows’ strategy of keeping a safe distance from the BS cows can also be observed in the selection of noon and night resting sites in different seasons: in winter and early spring, when the cows did not need to drink frequently, the 2 breeds used different rest sites both at midday and at night, but in the summer BA cows preferred to rest near the water troughs at midday even though the BS cows rested at the same site, whereas for the night rest the BA cows selected different sites from the BS cows. If maintaining a safe distance from the BS cows was a substantial consideration in the BA cows’ strategy, then their LFI might be smaller during some seasons if BS cows were not present in the same paddock. Conclusions The small-framed BA cows exploited their relative advantage of lower specific costs of activity (Aharoni et al., 2013), especially when the weather was hot and herbage quality was low; they selected diurnal movement itineraries that were either different from those selected by the BS cows, or spatially similar but temporally separated. It is suggested that BA cows were more selective in seasons in which the quality of the herbage in the paddock was highly variable and, moreover, that the small-framed BA cows were more adaptable than the large-framed BS cows to hot weather conditions and poor-quality herbage. LITERATURE CITED Aharoni, Y., A. Dolev, Z. Henkin, Y. Yehuda, A. Ezra, E. D. Ungar, A. Shabtay, and A. Brosh. 2013. Foraging behavior of two cattle breeds, a whole-year study: I. Heat production, activity, and energy costs. J. Anim. Sci. 91:1381–1390. Aharoni, Y., Z. Henkin, A. Ezra, A. Dolev, A. Shabtay, A. Orlov, Y. Yehuda, and A. Brosh. 2009. Grazing behavior and energy costs of activity: A comparison between two types of cattle. J. Anim. Sci. 87:2719–2731. Arnold, G. W. 1981. Grazing behaviour. In: F.H.W. Morley, editor, Grazing Animals. Elsevier, Amsterdam, Oxford, New York. pp. 79–104.
Downloaded from https://academic.oup.com/jas/article-abstract/92/2/758/4702579 by Universite Laval user on 23 April 2018
Bailey, D. W., J. E. Gross, E. A. Laca, L. R. Rittenhouse, M. B. Coughenour, D. M. Swift, and P. L. Sims. 1996. Mechanisms that result in large herbivore grazing distribution patterns. J. Range Manage. 49:386–400. Bailey, D. W., M. R. Keil, and L. R. Rittenhouse. 2004. Research observation: Daily movement patterns of hill climbing and bottom dwelling cows. J. Range Manage. 57:20–28. Bailey, D. W., H. C. Van Wagoner, and R. Weinmeister. 2006. Individual animal selection has the potential to improve uniformity of grazing on foothill rangeland. Rangeland Ecol. Manag. 59:351–358. Brosh, A., Z. Henkin, E. D. Ungar, A. Dolev, A. Orlov, Y. Yehuda, and Y. Aharoni. 2006. Energy cost of cows’ grazing activity: Use of the heart rate method and the Global Positioning System for direct field estimation. J. Anim. Sci. 84:1951–1967. Gillen, R. L., W. C. Krueger, and R. F. Miller. 1984. Cattle distribution on mountain rangeland in northern Oregon. J. Range Manage. 37:549–553. Gutman, M., Z. Holzer, H. Baram, I. Noy-Meir, and N.G. Seligman. 1999. Heavy stocking and early-season deferment of grazing on Mediterranean-type grassland. J. Range Manage. 52:590–599. Gutman, M., Z. Holzer, N. G. Seligman, and I. Noy-Meir. 1990. Stocking density and production of a supplemented beef herd grazing yearlong on Mediterranean grassland. J. Range Manage. 43:535–539. Gutman, M., and N. G. Seligman. 1979. Grazing management of Mediterranean foothill range in the Upper Jordan River Valley. J. Range Manage. 32:86–92. Henkin, Z., E. D. Ungar, and A. Dolev. 2012. Foraging behaviour of beef cattle in the hilly terrain of a Mediterranean grassland. Rangeland J. 34:163–172. Hunt, L. P., S. Petty, R. Cowley, A. Fisher, A. J. Ash, and N. MacDonald. 2007. Factors affecting the management of cattle grazing distribution in northern Australia: Preliminary observations on the effect of paddock size and water points. Rangeland J. 29:169–179. Irving, B. D., P. L. Rutledge, A. W. Bailey, M. A. Neath, and D. S. Chanasyk. 1995. Grass utilization and grazing distribution within intensively managed fields in central Alberta. J. Range Manage. 48:358– 361. Maule, J. P. 1990. The Cattle of the Tropics. Centre for Tropical Veterinary Medicine. University of Edinburgh, Edinburgh, UK. Rook, A. J., B. Dumont, J. Isselstein, K. Osoro, M. F. WallisDeVries, G. Parente, and J. Mills. 2004. Matching type of livestock to desired biodiversity outcomes in pastures– a review. Biol. Conserv. 119:137–150. Santos, S. A., C. Costa, G. Silva e Souza, S. M. A. Crispim, R. A. Pearson, and R. Gutierrez. 2003. Foraging strategy of cattle in the Pantanal rangeland, Nhecolandia sub-region, Brazil. In: N. Allsopp, A. R. Palmer, S. J. Milton, G. I. H. Kerley, K. P. Kirkman, R. Hurt and C. Brown, editors, Proc. 7th International Rangeland Congress. 26th July-1st August 2003, Durban, South Africa. Document Transformation Technologies, Irene, South Africa. pp. 23–30. Šárová, R., M. Špinka, J. L. A. Panamá, and P. Šimeček. 2010. Graded leadership by dominant animals in a herd of female beef cattle on pasture. Anim. Behav. 79(5):1037–1045. Smith, M. A., J. D. Rodgers, J. L. Dodd, and Q. D. Skinner. 1992. Habitat selection by cattle along an ephemeral channel. J. Range Manage. 45:385–390. Sternberg, M., M. Gutman, A. Perevolotsky, E. D. Ungar, and J. Kigel. 2000. Vegetation response to grazing management in a Mediterranean herbaceous community: A functional group approach. J. Appl. Ecol. 37:224–237. V. S. N. International. 2003. Genstat for Windows. 7th ed. VSN International, Oxford, UK. Zohary, M. 1973. Geobotanical foundations of the Middle East. Gustav Fischer Verlag, Stuttgart, and Swets and Zeitlinger, Amsterdam, the Netherlands.
