This paper introduces a new method of chaining junipers. Elevated chaining is the felling of trees by striking the tree above ground level with an anchor chain pulled between two crawler tractors. We measured the force at different striking heights to fell individual trees for two different growth forms of junipers. Force was measured by a loadcell and data recorded in a laptop computer mounted on a crawler tractor. Striking trees at elevated height reduced draft by 67 to 84% compared to ground-level chaining, and effectiveness was similar. We have developed a roller-ball chain to accomplish elevated chaining and preliminary performance data is included.

Introduction

Junipers (Juniperus spp.) have encroached upon 21.7 million acres of rangeland in Texas and 3.5 million acres in Oklahoma and negatively influenced land use. Ansley et al. (1995) estimated that redberry junipers in northwest Texas have increased their distribution by 61% from 1948 to 1982. This rangeland is basic to the livestock, wildlife and recreation industries of both states. Herbaceous plant production has been doubled and tripled by juniper removal (Arnold 1964, Steuter and Wright 1983). Prescribed burning for control of juniper has been used (Wright and Bailey 1982), but mortality is low if trees are over 6-ft tall. Moderate to dense stands of juniper must be mechanically treated to release herbage production necessary for successful burns (Rasmussen et al. 1986).

Chaining is the felling of trees by an anchor chain pulled between large crawler tractors. In dense stands of large junipers on rocky sites, chaining can be very expensive because of low chaining productivity. A low-cost method of tree felling is needed as a pre-treatment for burning. We believe that an elevated chaining technique could be developed that would reduce draft, and subsequently costs, to fell juniper trees. The objectives of this research was 1) to determine the draft requirements to fell individual juniper trees and the felling efficacy at different striking heights, and 2) to determine the productivity of an elevated chaining technique.

MATERIALS AND PROCEEDURES

Study Sites

Sites with mature Ashe juniper (Juniperus ashei Buchholz) and scattered eastern redcedar (Juniperus virginiana L.) in south-central Oklahoma near Ardmore, and mature redberry juniper (Juniperus pinchotii Sudw.) in northwest central Texas near Vernon were selected for this study. Trees in Oklahoma were single stemmed with large branches, whereas the trees in Texas were multi-stemmed at ground level. Soil at the Oklahoma site was a Kiti-Rock Outcrop Complex. Soil moisture was 11% at 0 to 6 inches (in.) and 13% at 6 to 12 in. when the test was conducted in July 1992. At the Texas site, soil was a Knoco Complex, and soil moisture was 10% at 0 to 6 in. and 11% at 6 to 12 in. at the time of the test in December 1992. Soil conditions were favorable for commercial chaining at both locations.

Breaking Bar and Instrumentation

The horizontal breaking bar was a reinforced, 8 x 4-in. (depth x width) H-beam mounted on wheels so that above-ground striking height could be set at 2, 3, or 4 ft. Two 5/8-in. diameter wire cables attached at the outer ends of the 10-ft long bar were stretched around a selected tree and then attached to the tractor's instrumented drawbar for pulling. For ground-level measurements, an anchor chain was looped around a tree and both ends fastened to the instrumented drawbar. A chain was used because it better duplicated actual ground-level striking than the breaking bar would have on the uneven soil/rock surfaces. A 60,000 pounds of force (lbf) loadcell was attached to the tractor drawbar and interfaced with our tractor-mounted, laptop computer.

Treatments and Experimental Design

In Oklahoma, 10 Ashe juniper trees were selected in July 1992 for felling at each of 4 striking heights, ground level, 2, 3, and 4 ft. At the Texas site, 15 trees were selected in December 1992 for felling at striking heights of ground level, 2, and 3 ft. Woody plants were cleared around each tree to simplify sampling and movement of the tree-felling apparatus. Each tree was numbered and the crown height, crown width, number of basal stems, and stem diameter at ground level were recorded. In Oklahoma there were no differences in average Ashe juniper height, width or trunk diameter among striking heights. Tree height ranged from 16 to 22 ft, and trunk diameter ranged from 9 to 30 in. Similarly, in Texas there were no differences in average redberry juniper height, width or total stem basal area among treatments. Tree height ranged from 9 to 18 ft; total stem area ranged from 26 to 422 in.². Trees at each site were grouped by stem size and randomly selected within each group for each striking height. Lower branches were trimmed from each tree as a safety precaution for field personnel working around the trees. Additional experimental design information is available from Wiedemann and Cross (1996b)

Following cable attachment of the breaking bar around a tree and to the tractor, drawbar force was measured as the bar was pulled against the tree. In Oklahoma, tractors were unable to exert sufficient force at ground level to fell 2 trees. Forces to fell these 2 trees were based on the actual force applied by the tractors even though the trees were not felled. These 2 trees were then felled using the 2 ft striking height resulting in a data set of 12 trees at this height.

The extent of tree breakage or uprooting was recorded for each tree. Uprooting in this study meant that a tree was pulled down to a horizontal position with most of the root mass pulled away from the soil, even though some roots might still be attached to the soil allowing the tree to live. Because trees were being preconditioned for a later burn, complete severing of all roots was not necessary. Felling efficacy was rated satisfactory when a tree was uprooted, uprooted with all roots completely severed, or completely severed at ground level. A partly uprooted tree or a tree returned to an upright position with stems cracked was rated as unsatisfactory felling.

Statistical Analysis

Maximum force to fell each tree was determined from the loadcell measurements and averaged for each treatment. Treatments were compared statistically. The relationship between felling force and tree size was determined by linear regression. Statistics was applied to validate the accuracy of the regression. Efficacy of tree felling was judged satisfactory or unsatisfactory. Estimates of crawler tractor flywheel power for elevated chaining were based on drawbar pull to fell 2 trees simultaneously at a speed of 2 to 3 miles/hr using 2 tractors with direct drive transmissions operating at 50% of manufacturer's rated capacity. For further statistical information, see Wiedemann and Cross (1996b).

Preliminary Performance Study

Our concept to achieve elevated chaining was to attach a large, rotating ball midway in an anchor chain pulled between two large crawler tractors, thus the name "roller-ball chain." A 6-ft diameter ball would provide a 3-ft chaining height. The objective of this study was to compare the rates of tree felling for a 2- and 3-ft elevated chaining technique to ground-level chaining.

The selected site was near Admore, OK. on a loamy prairie limestone outcrop range site. Soil was classified as a Claremore-limestone outcrop complex with 1 to 5% slopes. This gently sloping soil is on smooth ridges and side slopes of uplands and is characterized by limestone outcropping over 20% of the area. The site was infested with a dense stand of mature Ashe juniper and eastern redcedar. Ninety percent of the trees were between 12- to 25-ft tall while 5% were 25- to 45-ft tall. The remaining trees were less than 12-ft tall. Soil moisture was 19.4% at 0 to 3 in. and 14.8% at 3 to 6 in.

Spherical balls, 4 and 6 ft in diameter, were fabricated from 1/2-in. steel plate. A 4-in. diameter axle through the center of the ball allowed the ball to rotate as the unit was pulled by chains attached to each end of the axle. Two D-8 Caterpillars pulled 180 ft of 2 1/16 in. diameter chain (27.3 lbs/link) for standard ground-level chaining, and then a ball was attached midway in the chain for elevated chaining. The three treatments, ground level, 4-ft and 6-ft roller balls, were replicated four times, and plot acreage varied between 5.2 and 7.1 ac. Chaining was conducted in October 1993.

RESULTS AND DISCUSSION

Oklahoma Experiments

Force to fell Ashe junipers was significantly greater at the ground-level striking height than all others striking heights, but the force at the 3-ft striking height was not statistical different from that at 2 or 4 ft (Table 1). The force at 2 ft was greater than the force at the 4-ft striking height. Felling efficacy at the 4-ft striking height was unsatisfactory, which was significantly different from the satisfactory ratings at lower striking heights (Table 2).

Table 1.

Experiment	Strike Ht(ft)	Force to fell trees1 (lbf)
Ashe juniper (Oklahoma)	0	33,600 ± 15,500 a2
	2	17,600 ± 9,300 b
	3	11,200 ± 5,800 bc
	4	8,600 ± 3500 c
Redberry juniper (Texas)	0	18,300 ± 5,700 a
	2	3,000 ± 1,300 b
	3	2,800 ± 1,300 b

¹ Mean ± a standard deviation ( ± SD)

² Means within an experiment followed by the same letter are not different at the 5% level.

Table 2.

Experiment	Strike Ht(ft)	Accept	Reject	2 df=1
Ashe juniper (Oklahoma)	4	3	7	--
	3	10	0	10.8**
	2	12	0	12.3**
	0	8	0	9.2**
Redberry juniper (Texas)	3	6	9	--
	2	15	0	12.9**
	0	14	1	9.6**

** P<0.01.

At the 4-ft striking height, 50% of trees were only partly uprooted and 20% (trunk dia. <14 in.) were returned to an upright position even when the main trunk was split (70% unsatis-factory). At the 3-ft striking height, 50% of the trees were uprooted while the other half were severed at ground level (100% satisfactory). At the 2-ft height, 75% of the trees were uprooted and 25% were severed at ground level (100% satisfactory). At ground level, all trees were uprooted and completely pulled out of the soil (100% satisfactory), except for 2 trees whichthe tractors were unable to exert sufficient force to fell.

Felling trees at the 3-ft striking height would reduce draft by 67% compared to ground-level chaining, and uprooting/breaking would be high. The 3-ft height would be preferred for tree basal diameters of 11 to over 30 in. because performance was unsatisfactory at the (Fig. 1) 4-ft striking height when tree basal diameter was smaller than 16 in. If trees were predominantly small (less than 18 in. basal diameter) and the ground was not very rocky, then the 2-ft height would be more effective than the 3-ft height.

Felling force was strongly influenced by stem basal diameter for all striking heights (Fig. 1, Table 3). Soil cavities following uprooting varied widely within similar tree sizes, with some soil cavities over a cubic yard in size. Predicted force to uproot a 22-in. diameter tree at the 3-ft striking height would be 14,270 lbf while felling the same tree at ground level would require nearly a fourfold increase in force to 53,100 lbf. With the potential for juniper trees having stems over 40-in. in basal diameters at this site, the need for elevated striking heights is apparent.

Table 3.

Striking Height(ft)	Regression Equation	Standard error of the estimate	r2
Oklahoma -- Ashe juniper Draft (Y) in lbf on basal stem diameter (X) in in.
0	Y = -14000 + 3050X	±10,270	0.60*
2	Y = -2980 + 1250X	±8,520	0.23
3	Y = 1510 + 580X	±5,140	0.29*
4	Y = -1650 + 630X	±2,560	0.52*
Texas -- redberry juniper Draft (Y) in lbf on basal stem area (X) in in.2.
0	Y = 10410 + 40X	±4,060	0.53*
2	Y = 740 + 11X	±800	0.64*
3	Y = 690 + 11X	±850	0.64*

* P0.05.

Considering the poor traction in rocky soil, we estimate crawler tractors with more than 200 hp would be necessary for elevated chaining (3-ft striking height) in trees with 24-in. basal diameter stems and at least 270 hp if trees are larger.

Texas Experiments

More force was required at the ground-level height than at the other heights, while there was no statistical difference in the force required to fell trees at the 2- and 3-ft striking heights (Table 1). However, at the 3-ft striking height tree felling efficacy was unsatisfactory which was significantly different than the satisfactory ratings at lower heights (Table 2). Trees with total stem basal area less than 60 in.² remained upright when struck at the 3-ft height, and 47% were only partly uprooted (60% unsatis-factory). At the 2-ft striking height, 67% were severed at ground level and the remainder were uprooted (100% satisfactory).Striking at ground level resulted in 80% of the trees being completely pulled out of the soil and the remaining trees being uprooted (93% satisfactory), except one small tree (stem area = 26 in.²) which was partly broken.

Felling trees at a striking height of 2 ft would reduce the required draft by 84% compared to ground-level striking. Elevating the striking height in multi-stemmed redberry juniper dramatically reduces required draft. However, trees need to be larger than 60 in.² in total stem area (e.g., several stems with diameters of 4 in. or larger) or taller than 10 ft for elevated striking to be effective.

Felling force was strongly influenced by tree size expressed as the total of stem basal area (Fig. 2, Table 3). Predicted felling force for a tree with a stem area of 350 in.² (e.g., a tree with 5 stems with diameters of 12, 11, 10, 8, and 4 in.) would be 4,590 lbf at the 2-ft striking height while ground-level striking would require 24,410 lbf, a fivefold increase. Crawler tractors with at least 140 hp would be required for elevated chaining (2-ft striking height) in multi-stemmed redberry junipers with stem basal areas of 400 in.² or less.

Preliminary Performance Study

Rate of chaining (acres/hour (ac/hr)) for ground level, 4-ft ball and 6-ft ball was 10.1 ± 1.3, 12.1 ± 3.4, and 12.4 ac/hr ± 2.7, respectively. Rates of chaining were not significantly different from one another; however, there was considerable difference in tractor operation. Tractor overheating (powershift transmissions) was a problem in both the 4-ft ball and ground-level treatments (downtime for cooling not included) because of the increased pulling load. These two treatments could not be maintained on a continuous basis. Moreover, ground-level chaining could not fell trees over 25-ft tall without extreme difficulty and excessive wear and tear on the tractors. Large trees in these plots were not chained.

Felling trees with the 6-ft ball was judged satisfactory as trees taller than 8 ft were all partly uprooted. Trees less than 8-ft tall were bruised, but left standing. Tree felling with the other two chaining methods was slightly better because some of the larger trees were totally uprooted. Roller-ball chaining with the 6-ft ball averaged $16.13/ac based on a contractor rate of $100/h/tractor to operate on the rocky site. A controlled burn was conducted 17 months post treatment (20 March 1995). Fourteen months following the control burned, juniper mortality averaged 98% for all treatments (Wiedemann and Cross 1996a).

Control of redberry juniper, a sprouting species, may not be as effective as the chain/burn treatment on Ashe juniper and eastern redcedar. Studies are underway to determine if partial uprooting by elevated chaining will allow a subsequent burn to kill the buds on the exposed root system of redberry juniper.

Conclusions

Felling Ashe juniper trees 9 to 30 in. in basal diameter (17- to 22-ft tall) at a striking height of 3 ft reduced the average drawbar force by 67% compared to ground-level striking while maintaining a satisfactory efficacy of tree felling. Tree felling at 4-ft striking heights was not satisfactory. Predicted force to fell a 22-in. diameter tree was 14,270 lbf at a 3-ft striking height compared to nearly a fourfold increase of 53,100 lbf for ground-level striking.

A striking height of 2 ft reduced average drawbar force by 84% when felling redberry junipers 26 to 422 in.² in total stem basal area (9- to 18-ft tall) compared to ground-level striking, and tree felling efficacy was satisfactory. Felling was not satisfactory at the 3-ft striking height. Predicted draft to fell a 350-in.² stem area tree was 4,590 lbf at a 2-ft striking height compared to a fivefold increase of 24,410 lbf at ground-level striking.

Elevating striking heights above ground level can be used effectively for reducing the force required to fell individual juniper trees. We believe that an anchor chain could be modified to achieve elevated striking heights by attaching a rotating steel ball in the center of the chain. This modification would substantially reduce the drawbar pulling forces required for juniper chaining, and subsequently reduce cost, when preconditioning junipers for burning.

In the preliminary performance study, elevated chaining with the 6-ft diameter ball average 12.4 ac/h and 12.1 ac/h with the 4-ft diameter ball compared to ground-level chaining at 10.1 ac/h. Values were not significantly different. However, continuous operation could not be maintained with the 4-ft ball chain or ground-level chain because of tractor overheating caused by the increased pulling requirement. Moreover, trees over 25-ft tall could not be toppled with the ground-level chain and were detoured. Chaining with the 6-ft ball partly uprooted all trees over 8-ft tall while trees less than 8-ft tall were bruised but remained standing. Tree felling efficacy was similar for the other two chaining methods. A controlled burn in mid-March, 17 months following chaining, achieved a 98% mortality of the juniper trees in all plots when evaluated 14 months following the burn. The roller-ball chain appears to be a practical method to precondition Ashe juniper and eastern redcedar for control by burning. Efficacy of the technique for controlling redberry juniper, a sprouting tree, is under study.

Acknowledgment

The author is grateful for the cooperation and funding provided in part by the E. Paul and Helen Buck Waggoner Foundation and the W. T. Waggoner Estate, Vernon, Tex.; Brush Control and Range Improvement Association, Albany, Tex.; Daube Cattle Co., Ardmore, Okla.; Fun Country Resource Conservation and Development, Pauls Valley, Okla.; Noble Foundation, Ardmore, Okla.; Butterly Estate, Davis, Okla.; Wright Dozer Service, Wynnewood, Okla.; and Standley Dozer Service, Pauls Valley, Okla. We wish to thank Jim Altom and Nelson Muller, NRCS, Pauls Valley and Sulphur, Okla. for their assistance; Aaron Campbell and Gerral Schulz (who fabricated equipment), Texas Agricultural Experiment Station, Vernon, Tex.; and Agricultural Engineering Dept., Texas A&M University.

Literature Cited

Arnold, J.F. 1964. Zonation of understory vegetation around a juniper tree. J. Range Manage. 17:41-42.

Ansley, R.J., W.E. Pinchak, and D.N. Ueckert. 1995. Changes in redberry juniper distribution in northwest Texas. Rangelands 17:49-53.

Rasmussen, G.A., G.R. McPherson, and H.A. Wright. 1986. Prescribed burning juniper communities in Texas. Range and Wildl. Manage. Note 10. Texas Tech. Univ., Lubbock, Tex.

Steuter, A.A. and H.A. Wright. 1983. Spring burning effects on redberry juniper-mixed grass habitats. J. Range Manage. 36:161-164.

Wiedemann, H.T. and B.T. Cross. 1996a. Draft requirements for tree felling by chaining. Paper No. 965003. ASAE, St. Joseph, Mich.

Wiedemann, H.T. and B.T. Cross. 1996b. Draft requirements to fell junipers. J. Range Manage. 49(2):174-178.

Wright, H.A., and A.W. Bailey. 1982. Fire Ecology: United States and Southern Canada. John Wiley & Sons, Inc., N. Y.