Selecting Cattle to Prevent Grazing Distribution Problems

Final Report for SW98-064

Project Type: Research and Education
Funds awarded in 1998: $115,598.00
Projected End Date: 12/31/2002
Matching Non-Federal Funds: $46,158.00
Region: Western
State: Montana
Principal Investigator:
Derek Bailey
Montana State University
Expand All

Project Information

Abstract:

The potential of using breed and individual animal selection to improve grazing distribution was demonstrated in foothill rangeland. Cattle breeds developed in mountainous terrain used rugged rangeland more uniformly than breeds developed in more gentle terrain. Some cows used steeper slopes and areas further from water (hill climbers), while others used gentler terrain near water (bottom dwellers) when compared together and in separate pastures. Hill climbers left more residual vegetation in riparian areas and other sensitive rangeland than bottom dwellers. Animal performance was not adversely related to terrain use, so selection for grazing distribution should not reduce overall herd production.

Project Objectives:
  1. Using simple and easily obtainable observations, identify a behavior(s) that can be used to predict the general grazing distribution patterns of individual cattle.

    Determine if removing cattle with undesirable grazing distribution patterns will result in a more uniform use of forage in foothills rangeland.

    Determine the relationships among individual grazing distribution patterns and livestock production traits such as calf weaning weight, pregnancy rate and mature cow weight.

    Disseminate the results of the project to livestock producers, other extension personnel, rangeland managers and the scientific community.

Introduction:

Livestock grazing distribution is a critical component of sustainable rangeland management. Many concerns with livestock on grazed watersheds are the result of uneven grazing distribution. Cattle often congregate in portions of extensive pastures and graze forage to excessive levels while other areas receive little use (Pinchak et al. 1991). In extensive pastures, cattle may graze areas near water more heavily than areas far from water (Valentine, 1947; Cook, 1966). Cattle often avoid steep slopes (over 20%) and graze on gentler slopes (Mueggler 1965). In foothill and mountain rangeland, vertical distance to water may be more important than horizontal distance to water (Roath and Krueger 1982). Cows avoid climbing to higher elevations. Cattle often prefer riparian areas and spend a disproportionate amount of time in these areas as compared to uplands (Smith et al., 1992). Concentrated grazing, especially in riparian zones, may reduce vegetative cover and streambank stability as well as increasing soil erosion (Blackburn, 1984; Kauffman et al., 1983). Increasing uniformity of grazing is major goal for ranchers and land managers (Del Curto et al. 1999). If cattle spend more time grazing upland slopes farther from water, condition and function of riparian areas can be improved and wildlife habitat can be managed more effectively. The key for sustainable livestock grazing is to modify grazing patterns and prevent animals from overusing preferred areas within pastures (Bailey et al. 1996, Bailey et al. 1998).

Most of the management approaches currently used to increase uniformity of grazing have been know for over 45 years (Skovlin, 1957). Water developments, herding, salting, and fencing have been used successfully to improve livestock grazing distribution on both private and public lands. However, water developments and fencing usually require large capital expenditures. In some cases, water developments, rangeland fertilization and fencing may not be practical or cost effective (Hooper et al. 1969). As a result, managers are often reluctant to develop water and build new fences. Less expensive solutions such as salting away from water are usually not effective enough to sufficiently alter cattle grazing patterns (Bailey and Welling 1999, Ganskopp 2001). New management techniques are needed

Selecting cattle with desirable grazing patterns and culling cattle with undesirable grazing patterns has been suggested as a tool for improving distribution (Roath and Krueger 1982, Howery et al. 1996, Bailey et al. 1998). Research conducted in southern Idaho by Howery et al. (1996) supports this approach. These researchers found that cattle maintained certain home ranges, some grazing primarily uplands and others grazing meadows and riparian areas. Thus, removing animals that concentrate in overutilized areas and selecting animals that travel further from water and up steeper slopes has the potential to improve livestock grazing distribution.

Prior to this project, the effectiveness of selection for undesirable grazing distribution patterns had not been evaluated. Mosely (1999) argued that cattle grazing distribution patterns were the result of social interactions among animals within a herd. Dominant cows would graze in preferred areas, while subdominant cows would graze in less preferred areas. Thus, if cows that spent more time in preferred areas were removed or culled, other cows would replace the culled animals and “fill the vacuum.” If Mosely’s hypothesis was correct, selection or culling would have little, if any, impact on overall grazing patterns. The overall goal of this project was to determine selection had the potential to effectively alter cattle grazing patterns in rugged rangeland.

Research

Materials and methods:

STUDY SITES. Research was conducted at the Northern Agricultural Research Center (part of the Montana Agricultural Experiment Station) and at the Ross Ranch (cooperating private ranch). Both sites are located in north central Montana. The main study sites were located about 5 km from each other in the Bear’s Paw Mountains (20 to 30 km south of Havre, MT). Topography at both sites included steep and gentle slopes. Vegetation was dominated by perennial cool-season grasses with a few areas of shrubs and trees. Dominant grasses included Kentucky bluegrass, bluebunch wheat grass, rough fescue, Idaho fescue, blue grama, needlegrasses and prairie junegrass.

Most of the research at Northern Agricultural Research Center was conducted at the Thackeray Ranch, which is located 20 km from the headquarters of the research center. Experiment station cows grazed at the Thackeray Ranch during the summer and fall and spent the winter and spring at the headquarters near Havre, MT. Most of the studies at the Thackeray Ranch were conducted in 4 pastures, Back East, Back West, Rakes East and Rakes West. Cattle were observed in similar pastures prior to entering the primary study. Cattle grazed in the Rakes East and West pastures during June and early July, and they grazed in the Back East and West pastures during late July and August. The Rakes East pasture enclosed 81 ha with slopes ranging 0.2 to 51.4 degrees (mean slope 12.4 degrees). Topographic relief in this pasture ranged from 1152 to 1285 m (mean elevation was 1203 m). The Rakes West pasture enclosed 78 ha with slopes ranging from 0.2 to 36.1 degrees (mean slope was 12.4 degrees). Topographic relief ranged from 1180 to 1293 m (mean elevation was 1231 m). The Back East pasture enclosed 175.7 ha with slopes ranging from 0.1 to 48.8 degrees (mean slope was 15.4 degrees). Topographic relief ranged from 1186 to 1398 m (mean elevation was 1260 m). The Back West pasture enclosed 161 ha with slopes ranging from 0.3 to 45.2 degrees (mean slope was 15.0 degrees). Topographic relief varied from 1222 to 1398 m (mean elevation was 1297 m).

Most of the studies conducted at the Ross Ranch were conducted in the Blackwood North and Blackwood South pastures. Cows grazed theses pastures during June and early July. The Blackwood North pasture enclosed 159 ha, and topographic relief in the pasture ranges from 1114 to 1290 m. The Blackwood South pasture enclosed 149 ha with topographic relief varying from 1088 to 1290 ha.

HORSEBACK CATTLE OBSERVATIONS. Much of the study was based on cattle locations recorded by horseback observers. Locations of cattle at the Thackeray Ranch were recorded during the summers of 1997 to 2001. At the Ross Ranch, cattle were observed from 1999 to 2001. The procedure used for recording cattle locations using horseback observers was identical in all years and study sites. Topographic maps of pastures at both study sites were subdivided into 1- to 7-ha units based on slope, elevation, aspect and distance to water. Observers were trained to recognize the boundaries of all subunits within each pasture. Two to 4 observers on horseback rode pasture during a 1 to 2.5 hour period during the early morning (0600 to 0900 h) and attempted to record the location and activity of every cow in the pasture. Observers rode close enough to each cow to observe her identification number from her ear tag and firebrand on the hip (Thackeray Ranch) or shoulder (Ross Ranch). The goal of these observations was to obtain a scan sample of individual cow locations of the entire herd. Ideally, scan samples should be instantaneous (Lehner 1979). However, individually identifying and observing 40 to 160 animals instantaneously in extensive foothill pastures was not feasible. Observers recorded about 87% of the animals in the herd during an observation period. Cows in each pasture were observed 2 to 4 times each week.

Average slope, elevation, and distance to water (horizontal and vertical) were determined from each pasture subunit. For each cow, all location data collected in a pasture were pooled and used to determine average slope, horizontal distance to water and vertical distance to water of observed cow locations. For a complete description of the procedures used to calculate these values see Bailey et al. (2001a).

TRACKING WITH GPS COLLARS. Some of cows at the Thackeray Ranch were tracked using Lotek GPS 2000 collars. These collars recorded cows locations with an accuracy within 7 m (Moen et al. 1997) using the Global Positioning System (GPS). In 1998, the locations of 9 selected cows were recorded every 10 minutes during the day and every 20 minutes at night for about 3 weeks in late summer. In 1999 and 2000, the locations randomly selected cows were recorded every 15 minutes. In 2001, the locations of randomly selected cows were recorded every 10 minutes. During 1999 to 2000 cows were tracked for 3 to 15 consecutive days. Collars were removed, batteries recharged and location data downloaded. Collars were then placed on another set of randomly selected cows.

CATTLE. Cows observed in 1997 (n= 183) and 1998 (n=160) varied from 3 to 9 years of age and were from Hereford and Tarentaise breeding. One 1/5 of the cows were Hereford, and 1/5 were Tarentaise. The herd also included crosses of Herefords and Tarentaise. One fifth of the cows were 1/2 Hereford and 1/2 Tarentaise (HT). There were also similar numbers of1/4 Hereford and 3/4 Tarentaise (1H3T) and 3/4 Hereford and 1/4 Tarentaise (3H1T). Except for animals that died or were sold (open or because of limited forage), these cows remained in the study during 1999 to 2001. Daughters of these cows that were sired Angus, Charolais, Piedmontese or Salers bulls were also incorporated into the study during 1999 to 2001. These cows were observed in 2 or more pastures and were at least 3 years of age before they were included in the Objective 2 study.

OBJECTIVE 1 – PREDICTING GRAZING PATTERNS OF INDIVIDUAL COWS. In late May during 1997 to 1995, cows at Northern Agricultural Research Center were trailed from headquarters to the Thackeray Ranch. The position of cows within the herd during trailing was observed and recorded each year. For example, the lead cow would receive a 1 while the last cow in a herd of 160 cows would be 160. Positions of individual animals were recorded at 2 to 6 locations during the 20 km trail drive. The average position of a cow within the herd was calculated each year. Using a statistical model that included year, cow age and cow identification, the least square mean of position within the herd during trailing was calculated for each cow calculated using all trailing data.

Horseback observations of cattle grazing locations collected during 1997 to 2001 were analyzed using a statistical model that included pasture, year, lactation status (dry or lactating), age and cow identification. Least square means for average slope, horizontal distance to water and vertical distance to water for observed cow locations were calculated for each cow. An index of overall terrain use was developed using the other indicators of terrain use. In addition, an average ratio of the 3 indicators of terrain use was developed by dividing individual animal mean by the mean of all animals and multiplying by 100. This resulted in values that were similar in magnitude for slope, vertical distance to water and horizontal distance to water. Ratios of these 3 indicators of terrain use were average for each cow (overall ratio).

Corresponding least square means for average position within the herd and terrain use (slope, horizontal distance to water, vertical distance to water and an index) were calculated for 475 cows. The relationship between average position within the herd during trailing and terrain use was evaluated using simple correlations of the least square means. In addition, the linear regression of terrain use indicators on position within the herd during trailing was calculated.

Tracking data from 1998 were evaluated to determine if any other behaviors (other than trailing) could be used to predict grazing patterns. Cows that were ranked in the top 10% and bottom 10% with respect of using steep slopes and traveling further vertically from water were identified from 1997 horseback observation data. Cows that preferred steeper slopes and traveled further vertically from water were classified as “hill climbers” while the cows that used gentler slopes near water classified as “bottom dwellers”. Four randomly selected cows within the top 10% (hill climbers) and 5 randomly selected cows within the bottom 10% (bottom dwellers) were tracked with GPS collars during late August and early September during 1998. The time that cows began traveling to water and the time that cows left water were calculated and compared. Using USGS digital elevation maps, the average slope and elevation of collared cows were calculated during the period they were tracked.

OBJECTIVE 2 – DETERMINE IF SELECTION CAN IMPROVE UNIFORMITY OF GRAZING. At the Thackeray Ranch, locations of cows were recorded during the summers of 1997 and 1998 by horseback observers. During each of these two years, cows were observed in 2 pastures. Cows were ranked based on average slope and vertical distance to water of observed cattle locations. An evaluation of existing data did not suggest that horizontal distance to water was important in pastures at the Thackeray Ranch. Horizontal distance to water was always less 1.5 km. Holechek (1988) suggested that grazing levels would not be reduced at distances less than 1.6 km. Cows were then assigned into 1 of 2 groups based on these rankings. The hill climber group consisted of cows (top 50%) that were observed on steeper slopes and at locations further vertically from water. The bottom dweller group consisted of cows that were observed on gentler slopes near water.

The hill climber and bottom dweller groups grazed in similar but separate pastures during 1999, 2000 and 2001. During June and early July, hill climbers and bottom dweller groups were randomly assigned to either the Rakes East or Rakes West pasture. During late July and August, hill climber and bottom dweller groups grazed either the Back East or Back West pastures. Random assignments of groups (treatments) to pastures were made annually. During the 3 years, hill climbers and bottom dwellers grazed all pastures one or two times.

Younger cows that had not been evaluated in 1997 or 1998 were observed in at least 2 pastures (4 to 6 weeks per pasture) before being ranked. These younger cows were ranked separately and half of each age group were assigned to the hill climber group and half to the bottom dweller group. After being observed, ranked and assigned to either hill climber or bottom dweller groups, younger cows were included in the study.

During 1999, 79 bottom dweller cows grazed Rakes West pasture and 75 hill climbers grazed Rakes East. In that same year, 110 bottom dweller cows and 119 hill climber cows grazed Back East and Back West pastures respectively. During 2000, 39 hill climbers and 40 bottom dwellers grazed Rakes East and Rakes West pastures, respectively. Later that summer, 109 bottom dwellers and 111 hill climbers grazed Back West and Back East pastures, respectively. During 2001, 29 bottom dwellers and 27 hill climbers grazed Rakes West and Rakes East pastures, respectively. Later in 2001, 99 bottom dwellers and 103 hill climbers grazed Back East and Back West pastures respectively.

Cows locations during the early morning were recorded by horseback observers 2 to 4 times per week. Hill climber and bottom dweller groups were observed simultaneously. Riders rode entered the pastures grazed by hill climbers and bottom dwellers at approximately the same time and recorded positions of cattle in both groups as quickly as possible.

Five to 7 randomly selected cows from each group were tracked using GPS collars. During 1999 and 2000, collars were placed on 3 sets of cows in each pasture (each set was tracked 3 to 10 days depending on battery life). During 2001, 2 sets of cows were tracked in each pasture (each set was tracked 10 to 15 days). The difference between 2001 and 1999 and 2000 was that longer lasting batteries were available for the collars in 2001.

Forage utilization was estimated from measured stubble heights in pasture subunits using the height-weight procedure described in Cook and Stubbendieck (1986). One or 2 transects were randomly placed within each subunit in each pasture. Larger subunits had 2 transects. Fifteen plants were measured at 2-m intervals along each transect. Height-weight curves were developed for major species (Kentucky bluegrass and rough fescue). Published relationships (US Forest Service 1980) were used for other grasses. Forage utilization was calculated from stubble heights using these height-weight curves.

Protocols at the Ross Ranch were identical to those used at the Thackeray Ranch except for the following differences. Cows were not tracked with GPS collars at the Ross Ranch. There were not enough collars to track cows at both sites. Cows at the Ross Ranch were first observed in 1999. Rankings were based on slope, vertical distance to water and horizontal distance water using 1999 data. Horizontal distance to water was used at the Ross Ranch because portions of the pasture were further than 1.6 km from water (Holeckek 1988). During 2000, 61 bottom dweller and 62 hill climber cows grazed the Blackwood South and Blackwood North pastures, respectively. During 2001, 65 bottom dwellers and 63 hill climbers grazed the Blackwood North and Blackwood South pastures respectively.

Originally, the statistical design for this study was a paired t-test where hill climbers and bottom dwellers were evaluated in sets of 2 similar pastures. However, we were unable to identify or develop sufficiently similar sets of pastures in the foothill rangeland of the Bear’s Paw Mountains. Thus, we changed the design so that treatment (bottom dwellers or hill climbers) was randomly assigned each year, and both treatments were assigned to a pasture at least once. Pastures were used as blocks in the statistical design. The pasture by treatment interaction was used as the error term to test for differences between hill climbers and bottom dwellers when evaluating horseback observation data and the GPS tracking collar data. For the horseback data, the F statistic has 1 and 5 degrees of freedom (n=16). For the GPS collar data, the F statistic has 1 and 3 degrees of freedom (n = 12). The statistical model for the horseback data consisted of pasture, treatment, and the pasture by treatment interaction. For the GPS collar data, the model included pasture, treatment, pasture by treatment interaction, period (first, middle or end of grazing in each pasture), cow age (3 to 4, 5 to 6, and 7+ years), treatment by period interaction, and treatment by cow age interaction.

Forage utilization and stubble heights in riparian areas and other critical areas were analyzed separately. These areas were identified before the study and were areas that typically received heavy use (greater than 60% utilization). Average forage utilization and stubble height were calculated annually in each pasture. The statistical model included pasture, treatment and the pasture by treatment interaction. The pasture by treatment interaction was used as the error term for testing for differences among treatments. Analyses using a paired t-test for this data resulted in similar results.

Forage utilization and stubble heights collected throughout the pastures were analyzed to determine if uniformity of grazing differed between treatments. A total of 1127-forage utilization measurements (transects) were evaluated. We hypothesized that forage utilization would be more strongly affected by slope, vertical distance to water, and horizontal distance to water in pastures grazed by bottom dweller than in pastures grazed by hill climbers. The statistical model included year and pasture as fixed effects to adjust forage utilization values collected annually in each pasture to a common value. Treatment was included as a fixed effect to determine if there were differences in the linear relationships (regression coefficients) between forage utilization and terrain slope, vertical distance to water, and horizontal distance to water. Continuous effects in the model included terrain slope, vertical distance to water, horizontal distance to water, interaction between treatment and terrain slope, interaction between treatment and vertical distance to water, and the interaction between treatment and the horizontal distance to water.

OBJECTIVE 3 – DETERMINE THE RELATIONSHIP BETWEEN TERRAIN USE AND PERFORMANCE. The relationship between terrain use and animal performance of cows observed in 1997 and 1998 was evaluated. Cow weight, body condition score (BCS, 1 = emaciated and 9 = obese) and pregnancy status were evaluated at weaning (Oct 1). Calves were weighed at weaning, and weights were adjusted to 205 d of age using the birth to weaning average daily gain. Terrain use was estimated in 2 pastures each year by horseback observers. The correlation between terrain use variables (slope, vertical distance to water, and horizontal distance to water) and animal performance variables were evaluated. Separate analyses were completed for 1997 and 1998. For attributes related to the cow (terrain use, cow weight and BCS), the model included cow breed, cow age, and lactation status (lactating or dry). Residuals from this model were saved for all dependent variables to calculate residual correlations between grazing locations and performance traits after the model adjusted for the fixed effects (cow breed, age and lactation status). A similar analysis was conducted for cow terrain use and calf performance. The model included cow breed, cow age, and the breed of the calf’s sire (Angus, Charolais, Piedmontese, and Salers). Evaluating the relationship between fall pregnancy status and terrain use was not possible using residual correlations because pregnancy status is categorical data (pregnant or nonpregnant). Terrain use of pregnant and non-pregnant cows was compared using a model that contained cow breed, cow age, lactation status and fall pregnancy status. Differences in observed grazing patterns between pregnant and nonpregnant cows would suggest a relationship between reproductive performance and terrain use.

Research results and discussion:

OBJECTIVE 1 – PREDICTING GRAZING PATTERNS OF INDIVIDUAL COWS. Least square means of the position of cattle within the herd during trailing and terrain use recorded by horseback observers were evaluated using all data from 1997 to 2001. Observations were available for 475 cows. As expected, indicators of terrain use were correlated. The correlations between the average slope use and horizontal and vertical distance to water of observed cow locations were 0.13 and 0.50, respectively. The correlation between horizontal and vertical distance to water of observed cow locations was 0.41. However, the correlations between the position within the herd during trailing and terrain use were extremely low. Values ranged from 0.06 to –0.06. Using linear regression, position within the herd during trailing could not predict average terrain use (P > 0.1). Although preliminary analyses from the 1997 study at Northern Agricultural Research Center (Bailey and Hoffman 1998) suggested that cattle that are in the front of the herd during trailing tended to use steeper slopes and higher elevations than cattle in the back of the herd, the relationship was relatively weak. Analyses of 5 years of data collected in this and earlier studies, do not support the use of trailing data to predicting individual grazing patterns of cows grazing foothill rangeland.

During late August and early September 1998, location data from the 9 cows (4 of the best hill climbers and 5 of the worst bottom dwellers) using GPS tracking collars were analyzed to evaluate how well early morning observation reflected where cattle grazed. Cows usually went to water between 9 and 11 am and left water at 5 to 7 pm. Bottom dwellers traveled to water about 1 hour before (P < 0.05) the hill climbers. After leaving water in the evening, cattle generally traveled to an area where they would stay until they walked to water again on the following morning. From the period 1 hour after leaving water until 1 hour before traveling back to water, 77% of the observations were within 300 meters of their location at 7 am (when horseback observers usually recorded locations). Thus, the 7 am observation was indicative where the cow generally grazed. Cows that are observed grazing on gentle slopes near water during early morning hours are probably cows that have the least desirable grazing patterns. In addition, cows that travel to water early in the day also may be those with undesirable grazing patterns. OBJECTIVE 2 – DETERMINE IF SELECTION CAN IMPROVE UNIFORMITY OF GRAZING. Selection of livestock can be implemented by selecting breeds and by selecting individual animals within a breed. Locations of cattle at the Thackeray Ranch recorded by horseback observers were used to compare terrain use of cows from Hereford and Tarentaise breeding (Bailey et al. 2001a). Tarentaise cows (19.9%) used steeper slopes (P<0.05) than Hereford cows (18.1%) in 1998 but not in 1997. Tarentaise and cows with primarily Tarentaise breeding (1H3T) were observed at further vertical distances to water (P < 0.05) in both 1997 and 1998. Tarentaise had traveled 9 to 10 m further vertically from water than Herefords. There were no differences in cow breed for the horizontal distance from water (P > 0.1).

A similar study (Bailey et al. 2001b) was conducted with daughters (n = 199) of these Hereford, Tarentaise and Hereford x Tarentaise crosses. Daughters were sired by Angus, Charolais, Piedmontese and Salers bulls. Horseback observations of cow locations showed that cows sired by Piedmontese bulls (52.3 m) traveled further vertically from water (P < 0.05) than cows sired by Angus bulls (50.2 m). Breed of sire was not important in slope use (P = 0.4) and horizontal distance to water (P = 0.16). Dam breed was not important for slope use (P = 0.8), horizontal distance traveled from water (P = 0.7), and vertical distance to water (P = 0.3). Results of comparisons of terrain use by Hereford and Tarentaise cows and by cows sired by Angus, Charolais, Piedmontese and Salers bulls suggest that cattle breeds developed in mountainous terrain will use rugged rangeland more uniformly than breeds developed in more gentle terrain The Tarentaise breed was developed in the French Alps while the Hereford breed was developed in more gentle terrain in England. The Piedmontese breed was developed in the foothills of the Italian Alps, while the Angus breed was developed near the coast of Scotland. The Salers breed was also developed in mountainous terrain in central France. Although the difference between cows sired by Angus and Salers bulls for vertical distance to water was not statistically significant, the mean for the Salers sire breed (52.3) was closer and not significantly different from the cows sired by Piedmontese bulls. The observed differences among breeds also suggest that cattle grazing patterns may be heritable. To determine if individual animal selection would be effective cows were observed and ranked based on previous grazing locations observed by horseback observers. Horseback observers continued to observe cows classified as hill climbers and bottom dwellers in separate pastures during 1999 to 2001. Hill climbers and bottom dwellers were observed on similar slopes (P = 0.3), similar horizontal distances to water (P = 0.16), and vertical distance to water (P = 0.13). Although the differences were not statistically significant, the least square means suggested that hill climbers (HC) were using steeper slopes (HC = 21.2% and BD = 20.4%), traveling further horizontally from water (HC = 610 m and BD = 561 m) and traveling further vertically from water (HC = 49.7 m and BD = 43.8). Because the analyses were based on differences between pastures and not differences between animals, it is not surprising that statistical differences between treatment means were more difficult to detect. In addition, the relative importance of the 3 indicators of terrain use appeared to vary among pastures. For example, horizontal distance to water was not as important in the Rakes East and Rakes West pastures because they were smaller than the other pastures. Thus, an overall ratio was developed to combine and pool the differences in observed terrain use. The overall ratio was higher (P = 0.04) for hill climbers (102.8) than for bottom dwellers (97.2). This overall ratio indicates that horseback observers found hill climbers on terrain that was steeper and further from water than bottom dwellers when the 3 indicators of terrain use were combined. Collared cows were tracked at the Thackeray Ranch. Hill climber cows (22.8%) used steeper slopes (P = 0.06) than bottom dweller cows (21.9%). Slope use observed from the collar studies were similar to those observed by horseback observers. Use of steeper slopes increased as grazing season in each pasture progressed. Slope use in first period in each pasture was less (P < 0.001) than in the middle or last period in each pasture. Older cows (5 + years) used steeper slopes than younger cows (3 and 4 years). Hill climber cows (1236 m) used higher elevations (P = 0.097) than bottom dweller cows (1232 m). The mean elevation that tracked cows used increased (P < 0.001) from 1231 m during the first period (3 to 14 days) of grazing in a pasture to 1238 m in the last period of grazing in a pasture. The mean elevation that older cows (5 + years) were observed on was at least 5.5 m higher (P < 0.001) than those of younger cows (3 to 4 years). Prior to the study, riparian areas and other sensitive areas with gentle slopes near water that typically received heavy use were identified. Forage utilization in these areas was less (P = 0.02) in pastures grazed by hill climbers (48%) than those grazed by bottom dwellers (61%). In addition, stubble heights in these sensitive areas were greater (P = 0.04) when grazed by hill climbers (13 cm or 5.1 in) than when grazed by bottom dwellers (8 cm or 3.3 in). These results demonstrate that selection for grazing distribution has the potential to improve conditions of sensitive areas that have been heavily grazed in the past. Forage utilization in pastures grazed hill climbers was affected less by slope (P = 0.02) and horizontal distance (P = 0.004) to water than pastures grazed by bottom dwellers. For every degree increase in slope, forage utilization declined by 0.29 percentage points more in pastures grazed by bottom dwellers than in pastures grazed by hill climbers. For every 100 m further horizontally from water, forage utilization was 0.9 percentage points less in pastures grazed by bottom dwellers than in pastures grazed by hill climbers. The relationship between vertical distance to water and forage utilization was similar (P = 0.2) in pastures grazed by hill climbers and bottom dwellers. These forage use measurements show that cows classified as hill climbers grazed rugged foothill rangeland more uniformly than those classified as bottom dwellers. Uniformity of grazing was affected by slope and horizontal distance to water to a greater degree in pastures grazed by bottom dwellers than in pastures grazed by hill climbers. OBJECTIVE 3 – DETERMINE THE RELATIONSHIP BETWEEN TERRAIN USE AND PERFORMANCE. During 1997 and 1998 at the Thackeray Ranch, average slope and distance to water of observed cattle locations were not consistently related (P > 0.1) to cow weight, hip height or body condition score. Residual correlations varied from –0.06 to 0.04. During 1997, cattle that were observed on steeper slopes and higher elevations calved earlier in the season (P < 0.05) and had correspondingly heavier calves at weaning than cows that were observed on gentler slopes and lower elevations. The correlation between calving date and slope and vertical distance to water were –0.18 and –0.19, respectively During 1998, there were no relationships (P > 0.2) between the terrain where cattle were observed and the birth date or weaning weight of their calves. Overall, pregnant and nonpregnant cows used similar terrain. In 1997, non-pregnant cows used steeper slopes than pregnant cow (P = 0.08), but horizontal and vertical distances traveled to water were similar for pregnant and nonpregnant cows (P > 0.2). In 1998, all 3 indicators of terrain use were similar for pregnant and non-pregnant cows (P > 0.2). Based on this data, cattle ranchers should not expect any decline in the performance or profitability of their herd if cattle that prefer gentler slopes and lower elevations were removed from the herd (Bailey et al. 2001a). This is critical information for producers who may consider the implementation of a selection program based on livestock grazing distribution.

LITERATURE CITED

Bailey, D. W., B. Dumont, and M. F. WallisDeVries. 1998. Utilization of heterogeneous grasslands by domestic herbivores: theory to management. Annales de Zootechnie 47:321-333.

Bailey, D. W., J. E. Gross, E. A. Laca, L. R. Rittenhouse, M. B. Coughenour, D. M. Swift, and P. L. Sims. 1996. Invited Synthesis Paper: Mechanisms that result in large herbivore grazing distribution patterns. J. Range Manage. 49:386-400.

Bailey, D. W., D. D. Kress, D. C. Anderson, D. L. Boss, and E. T. Miller. 2001a. Relationship between terrain use and performance of beef cows grazing foothill rangeland. J. Anim. Sci. 79: 1883-1891.

Bailey, D. W., D. D. Kress, D .C. Anderson, D. L. Boss, and K. C. Davis. 2001b. Evaluation of F1 crosses from Angus, Charolais, Salers, Piedmontese, Tarentaise and Hereford sires V: Grazing distribution patterns. Proc., West. Sec. Amer. Soc. Anim. Sci. 52: 110-113.

Bailey, D.W. and M. Hoffman. 1998. Position of cattle within the herd during a trail drive. Abstr., Soc. Range Manage. Ann. Mtg., Guadalajara, Jalisco Mexico.

Bailey, D. W. and G. R. Welling. 1999. Modification of cattle grazing distribution with dehydrated molasses supplement. J. Range Manage. 52: 575-582.

Blackburn, W. H. 1984. Impacts of grazing intensity and specialized grazing systems on watershed characteristics and responses. Pages 927-933 in Developing Strategies for Rangeland Management. Natural Resources Council / National Academy of Science, Westview Press, Boulder, CO.

Cook, C. 1966. Factors affecting utilization of mountain slopes by cattle. J. Range Manage. 19: 200-204.

Cook, C. W. and J. Stubbendieck. 1986. Range research: basic problems and techniques. Soc. Range Manage. Denver, CO.

DelCurto, T., M. Porath, M. McInnis, P. Momont, and C. Parsons. 1999. Management strategies for optimal beef cattle distribution and use of mountain riparian meadows, p. 119-129. In: K.L. Launchbaugh, K.D. Sanders, and J.C. Mosley (eds.), Grazing behavior of livestock and wildlife. Idaho Forest, Wildlife and Range Exp. Sta. Bull. #70. Moscow.

Ganskopp, D. 2001. Manipulating cattle distribution with salt and water in large arid-land pastures: a GPS/GIS assessment. Appl. Anim. Behav. Sci. 73:251-262.

Holechek, J. L. 1988. An approach for setting the stocking rate. Rangelands 10:10-14.

Hooper, J. F. J. P. Workman, J. B. Grumbles, and C. W. Cook. 1969. Improved livestock distribution with fertilizer: a preliminary economic evaluation. J. Range Manage. 22:108-110.

Howery L. D., F. D. Provenza, R. E. Banner, and C. B. Scott. 1996. Differences in home range and habitat use among individuals in a cattle herd. Applied Animal Behaviour Science 49: 305-320.

Howery, L.D., F.D. Provenza, and R.E., Banner. 1998. Social and environmental factors influence cattle distribution. Appl. Anim. Behav. Sci. 55:231-244.

Kauffman, J.B. and W.C. Krueger. 1984. Livestock impacts on riparian ecosystems and streamside implications. A review. J. Range Manage. 37:430-438.

Kauffman, J. B., W. C. Krueger, and M. Vavra. 1983. Impacts of cattle grazing streambanks in northeastern Oregon. J. Range Manage. 36: 683-685.

Lehner, P. N. 1979. Handbook of Ethological Methods. Garland STMP, New York.

Moen, R., J. Pastor, and Y. Cohen. 19997. Accuracy of GPS telemetry collar locations with differential correction. J. Wildl. Manage. 61:530-539.

Mosley, J. C. 1999. Influence of social dominance on habitat selection by free-ranging ungulates. Pages 109-118 in K. L. Launchbaugh, K. D. Sanders, and J. C. Mosley, (eds). Grazing Behavior of Livestock and Wildlife. Idaho Forest, Wildlife and Range Experiment Station Bulletin #70, University of Idaho, Moscow, ID.

Mueggler, W. F. 1965. Cattle distribution on steep slopes. J. Range Manage. 18:255-257.

Pinchak, W. E., M. A. Smith, R. H. Hart, and J. W. Waggoner, Jr. 1991. Beef cattle grazing distribution patterns of foothill rangeland. J. Range Manage. 44:267-275.

Roath, L. R. and W. C. Krueger. 1982. Cattle grazing and behavior on a forested range. J. Range Manage. 35: 332-338.

Skovlin, J. M. 1957. Range riding – the key to range management. J. Range Manage. 10: 269-271.

Smith, M. A., J. D. Rodgers, J. L. Dodd, and Q. D. Skinner. 1992. Declining forage availability effects on utilization and community selection by cattle. J. Range Manage. 45: 391-395.

U.S. Forest Service Rocky Mountain Forest and Range Experiment Station. 1980. Utilization gauge: an instrument for measuring the utilization of grasses. American Slide-Chart Corp., Wheaton, Il.

Valentine, K. A. 1947. Distance from water as a factor in grazing capacity of rangeland. J. Forest. 45: 749-754.

Research conclusions:

Rangeland livestock producers have a new tool to improve grazing distribution. Grazing patterns in extensive and rugged pastures can be improved by selecting breeds that were developed in mountainous terrain. Results from this research demonstrate that individual animal selection has the potential to improve grazing distribution patterns. Although more research is needed before individual animal selection can be widely applied, this research was the first to show producers how effective selection could be in increasing uniformity of grazing. Improving livestock distribution will help prevent degradation of riparian areas and fisheries. Riparian areas grazed by hill climbers met the 4 to 5 inch stubble height standard typically required on public rangeland while pastures grazed by bottom dwellers did not met this standard. Differences between the hill climber and bottom dweller treatments were not only different statistically; they were different based on public land management standards. Over the long term, improvements in grazing distribution from selection should also reduce erosion and minimize any adverse impacts on wildlife habitat that may result from concentrated livestock grazing in certain areas. Rangeland livestock production may also be enhanced. Many rangelands contain large acreages that receive little, or no, grazing use. By selecting animals that are more willing to utilize rugged topography, the amount of forage that could be harvested on many rangeland pastures could be increased by more than 30% with no adverse consequences to adjacent (less rugged) rangeland.

Participation Summary

Educational & Outreach Activities

Participation Summary:

Education/outreach description:

SCIENTIFIC JOURNAL PUBLICATIONS.

Bailey, D. W., D. D. Kress, D. C. Anderson, D. L. Boss, and E. T. Miller. 2001. Relationship between terrain use and performance of beef cows grazing foothill rangeland. J. Anim. Sci. 79: 1883-1891.

Bailey, D. W., M. R. Keil, and L. R. Rittenhouse. 200?. Observations of daily movement patterns of hill climbing and bottom dwelling cows. J. Range Manage. (accepted with revision).

PROCEEDINGS.

Bailey, D.W. 2002. New approaches to protect watersheds and riparian areas by manipulating livestock grazing behavior. Proc., Bienn. Mtg Watershed Manage. Counc. , Stevenson, WA.

Bailey, D.W. 2001. Evaluating new approaches to improve livestock grazing distribution using GPS and GIS technology. Proc. 1st National Conference on Grazing Lands, Dec. 5-8, 2000, Las Vegas, NV, p. 91-99.

Bailey, D.W., D.D. Kress, D.C. Anderson, Boss, D. L., and K.C. Davis. 2001. Evaluation of F1 crosses from Angus, Charolais, Salers, Piedmontese, Tarentaise and Hereford sires V: Grazing distribution patterns. Proc., West. Sec. Amer. Soc. Anim. Sci. 52:110-113.

Bailey, D.W. 1999. Influence of species, breed and type of animal on habitat selection. p. 101-108, In: K.L. Launchbaugh, K.D. Sanders and J.C. Mosely, (eds.), Grazing Behavior of Livestock and Wildlife. Idaho Forest, Wildlife and Range Exp. Sta. Bull. #70. Moscow, ID.

Howery, L.D., D.W. Bailey and E.A. Laca. 1999. Impact of spatial memory on habitat use. p. 91-100, In: K.L. Launchbaugh, K.D. Sanders and J.C. Mosely, (eds.), Grazing Behavior of Livestock and Wildlife. Idaho Forest, Wildlife and Range Exp. Sta. Bull. #70. Moscow, ID.

ABSTRACTS.

Bailey, D.W. 2002. Daily movement patterns of hill-climbing and bottom-dwelling cows. Abstr., Soc. Range Manage. Kansas City, MO.

Bailey, D.W. 2002. Understanding livestock behavior to protect riparian areas. Abstr., Soc. Range Manage. Kansas City, MO.

Bailey, D.W., D.D. Kress, D.C. Anderson, and D.L. Boss. 2001. Potential for improving livestock grazing distribution by selection. Abstr., Soc. Range Manage. Kailua-Kona, HI.

Bailey, D.W. and M. Hoffman. 1998. Position of cattle within the herd during a trail drive. Abstr., Soc. Range Manage. Ann. Mtg., Guadalajara, Jalisco Mexico.

PUBLICATIONS IN PROGRESS. Bailey was invited to present a paper at a livestock grazing distribution symposium during the 2003 Society for Range Management Annual Meeting in Casper, WY. A paper from this presentation will be published in the Journal of Range Management and is currently being prepared. The title of the paper and presentation is “Identification and Creation of Optimum Habitat Conditions for Livestock.” This paper will include results from this project. In addition, at least 4 other scientific journal articles are planned and being prepared from this project. One paper will summarize the overall distribution patterns of hill climber and bottom dweller cows when grazed separately (Objective 2). A second manuscript will investigate the short-term behaviors that occurred during the study by focusing on GPS collar tracking in more detail. A third paper will summarize the results of the trailing part of the study (Objective 1). A fourth manuscript will summarize the effects of sire breed on grazing distribution. Currently, this data has only been published as a proceeding. Other papers could discuss the effects of group and social interactions on grazing patterns. This project has generated an enormous amount of unique behavioral data that can be analyzed and evaluated to answer new and exciting questions for years to come.

OUTREACH. During 2002, results from this project were presented at 9 workshops producers, extension, and range-oriented BLM and Forest Service employees. Roughly 250 people attended these presentations and about 100 of the attendees were producers. Derek Bailey also gave 2 presentations to more non-traditional audiences at the Western Division American Fisheries conference in Spokane, WA (part of SRM riparian committee presentation) and the Watershed Management Council biennial congress in Stevenson, WA. About 30 people (fishery biologists, hydrologists, land planners, etc.) attended each of these well-received presentations. Presenting results from this project to non-traditional audiences has given some people in the environmental community new insights on the potential for sustainable livestock grazing on rangelands.

During 2001, preliminary results from this research project were presented to livestock producers, extension personnel, university students, rangeland managers and the scientific community at field days, producer meetings and a scientific conference. Derek W. Bailey gave a presentation summarizing this project at the 2001 annual meeting of the Society for Range Management in Kona, Hawaii during February 19 – 22, 2001. A second scientific presentation was given at the annual meeting of the Western Section of the American Society of Animal Science in Bozeman, MT on June 23, 2001. Bailey was invited to give presentations at cooperative extension agent training meetings in Davis, CA, Dupuyer, MT and Lewistown, MT. Bailey gave an invited presentation on this project at the Governor’s Winter Grazing Conference in Dillon, MT in January 2001 to over 100 producers. Results from this study were presented at the NARC annual field day in Havre, MT. Students and faculty from University of Wyoming and University of Nevada, Reno (30 students and 5 faculty) toured the NARC study site and spent a day discussing results of this study with Bailey. In addition, 6 meetings were held where results from this and other livestock grazing distribution research conducted at NARC were presented to producers from Montana, other western states and Canada.

During 2000, preliminary results from this research project were presented to livestock producers, extension personnel, rangeland managers and the scientific community at two field days, producer meetings and a national conference. Derek Bailey gave a presentation at the “First National Conference on Grazing Lands” in Las Vegas, Nevada during December 5 – 8, 2000. A paper summarized the presentation. Bailey was invited to give presentations at the Fields of Tomorrow field day held near Havre, MT. Results from this study were presented at the NARC annual field day in Havre, MT. In addition, 7 meetings were held where results from this and other livestock grazing distribution research conducted at NARC were presented to producers from Montana, other western states and Canada.

During 1999, preliminary results from this research project were presented to livestock producers, extension personnel, rangeland managers and the scientific community at two field days, a range tour, producer meetings and a scientific conference. Derek Bailey was invited to give a presentation at the “Grazing Behavior of Livestock and Wildlife” conference in Moscow, Idaho during March 23-24, 1999. A reviewed paper summarized the presentation and was included in the proceedings published by the Idaho Forest, Wildlife and Range Experiment Station. Bailey was also invited to give a presentation for Oregon State University at their Union Station Field Day. Results were presented at the NARC annual field day in Havre, MT. The International Mountain Section of the Society for Range Management toured the NARC study site in the Bear’s Paws Mountains. The public was invited to join this tour, and several livestock producers and local community members attended. In addition, 12 meetings were held where results from this and other livestock grazing distribution research conducted at NARC were presented to producers from Montana, other western states and Canada.

Summary of producer attendance during 1999 to 2002:

Workshops and producer meetings – 409 producers plus 150 non-producers at 32 workshops/meetings

Conferences – 115 producers and 345 non-producers at 8 conferences.

Field Days – 495 producers and 225 non-producers at 5 field days.

Range and Study Site tour – 5 producers and 25 non-producers.

Project Outcomes

Project outcomes:

We were able to demonstrate that selection has the potential to improve uniformity of grazing and prevent overuse of riparian and other sensitive rangeland. In some situations, incorporation of selection could potential extend grazing seasons on public land. Stubble heights standards were met in pastures grazed by hill climbers and not met in pastures grazed by bottom dwellers. On public lands, land managers would have required the bottom dweller cattle to be moved out the pasture earlier than the hill climber cattle. When cattle are moved out of pastures early, the corresponding grazing season is reduced, and permitees on public lands are often must seek out alternative private grazing lands or must buy supplemental feed. Such alternative feed purchases increase production costs.

Research from this project also demonstrated that selecting for grazing distribution should not adversely affect performance of the cowherd. Terrain use was not correlated to cow weight, cow height and BCS. Calving date and the weaning weight of a cow’s calf was also not related to terrain use. Reproductive performance of the cow also does not appear to be related to terrain use. Culling animals that prefer gentler terrain near water (bottom dwellers) and selecting that use steeper terrain and areas further from water should not adversely affect animal performance. This is an important finding because selection may not have been practical and cost effective if bottom dwellers performed better than hill climbers.

Specific economic analyses of these practices cannot be completed at this time because more information is needed. First, we must determine if grazing distribution is heritable and can be improved through sire selection. If it is heritable, progress can be much more rapid and costs to implement selection should be lower. We must determine if development of cattle that prefer rugged terrain (hill climbers) can be accomplished through sire selection, training or some combination, before costs for developing such a selection program can be estimated. Second, the costs and benefits of implementing any management to improve grazing distribution are site specific. Although this study demonstrated an increase in uniformity of grazing through selection, the value of such improvements will be site specific. Economic analyses of the costs and benefits of selection for improved distribution may require evaluation of case studies.

Farmer Adoption

A few ranchers have adopted the management techniques that were tested in this study. Bob Budd is a manager for a Wyoming cattle ranch owned by the Nature Conservancy. Mr. Budd says culling cattle that prefer gentle terrain and riparian areas minimizes adverse impacts of livestock grazing on the fisheries of their ranch. Mr. Budd wanted to know if our quantitative study would confirm his personal observations obtained on the ranch. It did. Based on results from this project, Larry Bitz (a rancher in Big Sandy, MT) is now keeping daughters of cows that use more rugged terrain (hill climbers) to improve uniformity of grazing in his BLM grazing allotment. Public land managers have been very interested in this research and are trying to determine ways to encourage producers to incorporate these findings. For example, Wes Yamamoto, a fisheries biologist with the US Forest Service in Tiller, OR, shared the results of this study and other grazing distribution conducted at NARC with livestock grazing permittees that he works with.

During workshops, producers including Leachman Cattle Co. (Billings, MT) asked how they could implement the results of this project into their operation. Our response was that producers could use cattle breeds that were developed in mountainous terrain when grazing rough topography. They could also cull extreme bottom dweller cows. However, genetic progress using culling is very slow and more progress can be made selecting sires. If early learning is important, cattle training and heifer development would be critical. Producers could develop, train and market heifers and cows specifically designed to graze rugged, extensive rangeland. Producers were also concerned if selecting cows for grazing distribution would adversely affect cow performance or weaning weights. This research showed that grazing distribution was not related to cow performance and weaning weights (Bailey et al. 2001a).

Recommendations:

Areas needing additional study

A remaining and critical question is the degree that genetic and environmental (early learning) factors affect grazing patterns, which must be quantified before any sire and family group selection or specific training programs can be implemented. This project demonstrated that selection has the potential to modify forage utilization patterns and increase uniformity of grazing. However, it did not identify how producers could best implement a selection program. As stated above, sire selection is much more effective than culling for modifying traits. We must establish if grazing distribution patterns are heritable before we can recommend sire or family selection programs. A few producers have been interested in developing programs to train cows where to graze. They were familiar with the Howery et al. (1998) study and were convinced that grazing patterns could be improved through early learning (as calves) and later training (after weaning). There is a potential niche market for heifers that were trained and/or bred to use in rugged extensive rangeland

In the future, we would like to incorporate this research with other techniques in an integrated program including strategic supplementation, salting and herding. More research is needed to develop and evaluate combinations of management practices designed to improve cattle grazing distribution. Practices may have very synergistic effects. Bob Budd, manager of the Red Canyon Ranch, Lander, WY, has found that the integration of selection and herding allowed them to improve the condition of their streams and riparian areas. The integration of animal selection along with other practices, such as off-site water development, herding and strategic supplementation, has the potential to significantly improve uniformity of grazing and give managers an alternative to livestock exclusion to protect sensitive watersheds.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.