Dense stands of redberry juniper (Juniperus pinchotii Sudw.) and Ashe (blueberry) juniper (J. ashei Buchholz) severely reduce forage production, interfere with handling and movement of livestock, degrade wildlife habitat, and they waste our State's water resources. The juniper problem on Texas rangelands is viewed by many of us involved in rangeland research, extension, education, and management as the most critical and insidious problem we face as we enter the 21st century. The rapid rate at which juniper is increasing in numbers and canopy cover and invading Texas grasslands dictates the immediate use of effective, ecologically safe, and economically sustainable juniper management systems on the 20 million acres of Texas rangeland currently infested. Juniper cannot be eradicated, but it can be managed effectively and economically with the knowledge and technology currently available. Most land owners and managers will likely have to "set aside" their old "mind sets" (paradigms) about brush control, and to begin thinking as "ecologists" to successfully manage their juniper in the long term. The screw worm problem is being managed, not with heavy machinery or millions of tons of insecticide, but rather with a little technology and a lot of knowledge of the pest's biology, ecology, and behavior. The major difference between the screw worm and juniper solutions is that individual landowners, and not the federal government, will have to design and implement the juniper program.

Principles of Juniper Management

Effective, ecologically safe, and economically sustainable juniper management systems must be based upon all knowledge available on the biology and ecology of junipers. We must also rely upon knowledge gained from past successes and failures in managing junipers and other noxious species and fully utilize the available knowledge on livestock grazing management. Some of the principles critical in planning and implementing effective juniper management systems include:

1) Consider your future goals and objectives for your juniper-infested rangeland resource. Is your only interest in growing more forage for livestock? Are you also interested in aesthetics, wildlife, recreation, and in increasing the real estate value of your property? Keep in mind that some juniper, even some strategically located dense stands, may be valuable, depending upon the multiple land use options that you, your heirs, or potential land buyers may be interested in. It may be in your best interest in the long term to create a mosaic of grasslands interspersed with thinned juniper savannahs and juniper woodlands. Excessive brush control can be detrimental to wildlife populations, as well as to aesthetic, recreational, and real estate values of the land. Carefully planned and selective juniper control programs can optimize the value of the land for multiple considerations;

2) Good grazing management is a "primary tool" for managing junipers and other noxious plants ecologically. Proper stocking rate coupled with periodic pasture rest during the growing season maximizes the vigor and cover of desirable grasses and forbs and the ability of these desirable plants to out compete emerging juniper seedlings. Proper grazing management minimizes the abundance of "safe sites" (bare ground) in which juniper seedlings can emerge and establish. Good grazing management puts pressure on the "weakest link" in juniper's life cycle, its establishment phase. It shifts the competitive advantage to the desirable forage plants. Good grazing management is also essential for the effective use of another "ecological tool", fire, because it facilitates the accumulation of an adequate amount of grass (fine fuel) and adequate fine fuel continuity to support intense fires;

3) Utilize prescribed fire effectively and safely, and on an 8- to 10-yr cycle. Intense fire was a critical "ecological factor" that kept junipers out of Texas grasslands prior to settlement by Europeans. The low per-acre cost for burning renders fire more economically sustainable than all other juniper management strategies, with the exception of good grazing management and goating. Prescribed fire must be correctly timed, relative to the juniper's life cycle, so that it impacts the vulnerable growth stage;

4) Utilize biological control to the fullest extent possible. Goating is another effective ecological tool for controlling or suppressing juniper seedlings, saplings and regrowth. Maintenance control of juniper with goats can be a profitable ranch enterprise, hence it is an "economically sustainable" element for juniper management systems. Defoliation of juniper by goats decreases its vigor, forces it to utilize food reserves stored in its stems, crowns, and roots, and decreases its use of soil water. This shifts the competitive advantage to the desirable forage and browse plants. Goating must be carefully managed to prevent the destruction of desirable browse plants;

5) Single-treatment approaches have rarely resulted in economically acceptable, long-term solutions to brush problems. The initial juniper management plan must incorporate follow-up, maintenance control practices that are synergistic, or at least complimentary, to the initial treatment. Properly selected and sequenced follow-up control practices that cost relatively little are essential to extend the life of expensive initial treatments for a sufficient time to recover the cost of the initial treatment;

6) Utilize the "weak link" principle to the fullest extent possible. Put the pressure on juniper in its most vulnerable life stage, i.e. while it is a seedling or sapling and (in the case of redberry juniper) before the bud zone is covered with soil;

7) Minimize your capital outlay and maximize your choice of treatment alternatives by controlling junipers in the seedling or sapling life stages rather than waiting until they are full-grown, mature plants. It's much easier and less expensive to kill seedlings and saplings compared to mature junipers. The conventional approach has been to ignore brush infestations until they are mature and dense because the small brush plants have little effect on forage production or livestock carrying capacity. This philosophy evolved back when the costs for broadcast, conventional control practices were relatively low. This situation has changed dramatically. Preventative juniper control is the "common sense" approach, just like preventative health maintenance is the common sense approach we should take to avoid serious diseases and expensive medical treatment;

8) Minimize the necessity for future juniper control by managing the "seed bank". Kill junipers before they become reproductively mature. If your junipers are already mature and capital is limiting, then use selective control methods to kill the female junipers first. Control the male plants in subsequent years. It may be desirable to leave male junipers in drainages and other selected areas for wildlife cover or as windbreaks/thermal cover for livestock.

Mechanical Methods

Chaining, tree dozing or grubbing, bulldozing and root plowing have been the traditional methods used for controlling junipers (Scifres 1980). Use of mechanical methods seemed to decline during the 1980's as costs of equipment and fuel escalated and with changes in tax laws. Recent engineering innovations, however, have rejuvenated interest in mechanical methods. Front-end loaders with smaller track cleats or rubber tires and with hydrostatic steering (each track driven by a separate hydraulic circuit) provide greatly improved maneuverability and minimize turf damage during juniper dozing. Hydraulically assisted grubbers reduce the horsepower requirements for tree dozing and various types of new transmissions increase the speed and efficiency of grubbing (Wiedemann and Cross 1981). Foam-filled tires allow rubber-tired equipment to be used for grubbing without the "down time" for repairing flats (Wiedemann and Cross 1982). Small-to-medium size junipers can now be controlled economically with grubbing implements mounted on the 3-point hitch of low-horsepower farm tractors (McFarland and Ueckert 1982) or on the front of small front-end loaders. A new innovation in chaining technology, the "roller-ball chain" reduces the horsepower requirements for chaining redberry juniper by 80% (Wiedemann and Cross 1996; Wiedemann 1997).

Maintenance (follow-up) control is always necessary within a few years after these mechanical methods are used because of resprouting and very rapid growth of junipers not uprooted mechanically and because of rapid growth of small junipers missed by these treatments and those that establish subsequent to the treatments. Killing the large junipers releases much more water, sunlight, and nutrients for the juniper seedlings as well as the understory grasses and forbs, and the growth response of small junipers is usually more dramatic than that of the desirable forage plants. Densities of redberry juniper seedlings averaged 508/acre 3 years after the mature trees had been tree dozed in a study near Sonora, Texas (Ueckert and Whisenant 1982). Densities of seedlings less than 6 in. tall on 3 sites near San Angelo ranged from 260 to 1,469/acre (Dye et al. 1995). Similar seedling densities in most mature stands of Ashe and redberry juniper dictate planning for follow-up control practices. The planning should be done prior to installation of the initial treatment. Several highly effective and proven maintenance control treatments are available, including prescribed fire, goating, individual plant herbicide treatments, hand cutting, hand grubbing, and mechanical grubbing. The two "ecological" methods, prescribed fire and goating, should be utilized first following the initial mechanical treatment, if at all possible, to reduce juniper seedling abundance to the greatest extent possible at least cost. The more labor-intensive and energy-consumptive methods may then be economically feasible for controlling the juniper seedlings and saplings that survive fire and/or goating. These maintenance control practices will be discussed in detail later in this paper.

Hydraulic shears are a fairly recent innov-ation for mechanical control of junipers and other woody plants. These powerful shears can be mounted on front-end loaders or to the 3-point hitch of farm tractors. They can be quite effective for killing Ashe juniper since it is not a crown sprouter. Redberry juniper, however, is a crown sprouter and removing the top of mature plants with hydraulic shears results in prolific resprouting, i.e. only temporary "suppression" of this species. Redberry junipers can be killed after the tops are removed with shears, chain saws, or axes if the stumps are sprayed immediately after cutting, to the point of runoff, with 2 to 4% picloram (Tordon 22K) in a water carrier + ¼% surfactant. This treatment has resulted in 78 to 100% rootkill of mature redberry junipers (D.N. Ueckert, unpublished data).

Hand grubbing and hand cutting are additional "mechanical" methods known to be effective for juniper control. Ashe junipers of any size can be killed by cutting at or slightly above the soil surface with an ax, chain saw, or hand-pruning shears. Redberry juniper seedlings and saplings can also be killed if cut at the soil surface, as long as the "bud zone" (a swollen "bulb" on the basal stem) is still above the soil surface. After the bud zone is covered by soil and litter, redberry junipers can be killed by hand grubbing as long as the stems are severed below the bud zone (Ueckert and Whisenant 1982; McGinty and Ueckert 1996). Seedlings and saplings up to about 28 in. tall can be hand grubbed fairly easily and inexpensively when soils are relatively moist and not too rocky. We found that the cost varied from $3.65/acre (2.15¢/juniper) to hand grub 170 junipers/acre to $9.00/acre (1.36¢/juniper) to grub 660/acre in a trial near San Angelo (D.N. Ueckert, unpublished data). Labor for this work was valued at $5.00/hr and the junipers varied from about 2 to 28 in. tall. Maintenance control of juniper by hand grubbing or cutting should not be necessary more frequently than on a 6- to 8-yr cycle. Based on our average cost ($6.18/acre to grub 434 juniper/acre), the pro-rated annual cost would be about $1.03/acre/yr on the 6-yr cycle or $0.77/acre/yr on the 8-yr cycle. This work is less stressful and most effective (small seedlings more easily seen) during late autumn through early spring when air temperatures are cool and the grasses are dormant. Capital outlay per worker is minimal (± $30) and no special skills or license are necessary.

Herbicides

Herbicidal control of juniper in Texas is currently limited to individual plant treatments. Redberry and Ashe junipers can both be controlled with high-volume foliar sprays (leaf sprays) containing 1% picloram (Tordon 22K) or soil treatments of undiluted hexazinone (Velpar L), applied at 2 ml/3 ft of juniper height or diameter, beneath the juniper canopy (Welch 1995; McGinty and Ueckert 1996, 1997). Ashe juniper can also be controlled with soil treatments of undiluted picloram applied at 4 ml/3 ft of canopy height or diameter, but this treatment is not recommended for controlling redberry juniper. Recent research in the western Edwards Plateau showed that 10 ml of undiluted picloram/3 ft of canopy height killed 82% of the redberry junipers less than 6 ft tall and 65% of those over 6 ft tall (D.N. Ueckert, unpublished data).

The individual plant treatments listed above are most applicable for controlling relatively low densities (fewer than 200 plants/acre) of juniper seedlings or juvenile plants no taller than about 3 ft. Costs of labor and herbicide escalate rapidly in dense or mature junipers. Easy-to-follow instructions for do-it-yourself juniper control are available in the Brush Busters "How to Master Cedar" leaflet (McGinty and Ueckert 1996).

Aerial sprays of herbicides used on rangelands for honey mesquite, pricklypear, and herbaceous weed control sometime defoliate, but rarely kill redberry or Ashe junipers. We recently found that redberry juniper saplings were not controlled by broadcast sprays of picloram (Tordon 22K) applied at rates as high as 1 lb/acre in water + surfactant, water + penetrants, or in diesel fuel-water emulsion carriers (D.N. Ueckert, unpublished data).

Fire

The use of fire for juniper control is actually not a recent discovery. Landowners in Blanco County, Texas had an effective and widely used hand-cut/fire system for Ashe juniper control in the 1930's. About 63,000 acres were cut and burned during 1938 alone in that single county (Jenkins 1938). The spectacular grass response on 65-acres that had been hand cut in the early summer of 1930 then accidentally burned on an August afternoon was the "great stimulus in Blanco County that started the ranchers to cutting and burning their cedar" according to an article in The Cattleman in 1939 (Jenkins 1939). The reported benefits included: livestock carrying capacities increased from 1 animal unit/20-30 acres to 1 animal unit/5-6 acres; springs started flowing much higher volumes, and some springs began flowing that hadn't flowed in 45 yr; sheep losses from "blow fly" damage decreased 300-400%; reduced stress on livestock, horses and mules from "blood sucker flies" (horse flies and deer flies); and an abundance of bobwhite quail in areas where they had not been present for 40-50 yr. The Cattleman magazine article touted the A.A.A. program that subsidized ranchers for cutting and burning their juniper as "the best program the Government has ever put on and [it] means more to the country than any other move yet attempted".

Prescribed burning is rarely an applicable practice on rangeland supporting high densities of mature junipers because competition from the junipers prevents production of enough grass (fine fuel) to carry a fire of sufficient intensity to kill woody plants. However, prescribed fire can be extremely useful for controlling the initial invasion of juniper seedlings and saplings into grasslands. It can also be effectively used as a "follow-up" treatment after chaining or tree dozing on rangelands where a high proportion of the junipers have reached maturity. The initial mechanical treatment is used to release the grasses from juniper competition so that a sufficient fine fuel load can be produced to carry an intense fire. The Link Ranch, near Strawn, Texas, has been effectively using a roller chopper to knock down Ashe juniper for grass release to allow them to use prescribed fire.

Properly sequenced fire prolongs the life of the expensive mechanical treatments, eliminates large amounts of downed woody debris, kills many of the smaller plants missed by the mechanical treatment or that establish subsequent to the mechanical treatment, and suppresses resprouting trees (Rasmussen et al. 1986). The first fire should be installed 3 to 5 years following chaining, tree dozing, or roller chopping, and subsequent repeat burns should be installed on an 8- to 10-year cycle. It is of utmost importance to burn when the juniper seedlings and saplings are vulnerable, i.e. before Ashe junipers seedlings exceed 4 ft in height and before the bud zone of redberry junipers seedlings become covered by soil. The burning cycle for lowland sites will be shorter than for upland sites because junipers grow faster and the bud zones become buried sooner on lowland sites compared to upland sites (Steuter and Britton 1983).

Fine fuel (dry grass) loads of 2,000 lb/acre are recommended for successful burning in juniper communities. Successful burns can be conducted in areas dominated by sod-forming grass (such as buffalograss) with only 1,000 lb/acre of fine fuel, but in areas dominated by bunchgrasses (such as little bluestem) at least 2,000 lb/acre is required (Wink and Wright 1973; Rasmussen et al. 1986).

Burning in areas with live, chained, or dozed junipers is hazardous because of the volatility of juniper, the length of time partially buried juniper wood may continue to burn, and the fact that firebrands from burning piles of juniper can ignite spot fires up to 1,000 ft away. The precautions and prescription burning techniques outlined in Texas Tech Management Note 10 "Prescribed Burning Juniper Communities in Texas" (Rasmussen et al. 1986) should be followed to the letter when burning juniper-infested rangeland.

Ashe juniper is more easily controlled with fire compared to redberry juniper because it is a non-sprouting species. Ashe junipers less than 4 ft tall can be killed if the fine fuel loads are at the recommended levels, and larger trees can also be killed with heavier fine fuel loads (Wink and Wright 1973). Secondary brush species, such as flameleaf sumac, shin oak, and live oak, may become problems after Ashe juniper is controlled by burning (Rasmussen and Wright 1989). About 70% of the redberry junipers whose bud zones are above the soil surface may be killed by burning with the recommended fine fuel load, but only about 3% of those with buried bud zones are likely to be killed (Steuter and Britton 1983). Redberry junipers on shallow, rocky soils may have exposed bud zones until they are 10 to 15 yr old, but bud zones of those on deep soils may be buried in less than 10 yr.

Production of most warm-season grasses in juniper communities is increased or not affected if winter burns are followed by growing seasons with normal or above-normal precipitation. In contrast, production will decline if the winter burn is followed by a dry spring and summer (Wink and Wright 1973; Steuter and Wright 1983).

Soil erosion may be a serious problem where fire is used on sites with moderate to steep slopes. Adverse effects relative to soil loss, runoff, and water quality lasted 9 to 15 months after burning on moderate slopes and for 15 to 30 months or longer after burning on steep slopes in an Ashe juniper community (Wright et al. 1976).

There is little doubt that summer fires would be more effective than winter fires for controlling juniper. However, the probability of damaging the perennial grasses and the acute hazards associated with burning are also greater with summer burning. The novice is best advised not to try summer burning at all! Those with experience in installing winter prescribed fires are advised not to try summer burning until researchers report their final evaluations of many summer burning experiments.

Fire + Herbicides

Sequential applications of fire and herbicides have proven highly effective for control of other low-value range plants, such as pricklypear, and the partial susceptibility of juniper to both fire and herbicides suggests that fire and herbicide might be synergistic for juniper control if applied sequentially at the proper interval. Preliminary experiments have suggested there is a synergism between fire and picloram for redberry juniper control. Fire intensities that simulated the time-temperature curves generated with 1,500 lb/acre of fine fuel applied when soils were moist near Mertzon, Texas killed 10% of the redberry juniper, but broadcast sprays of picloram applied at 0.25, 0.5 and 1.0 lb/acre when regrowth from these top-killed plants was 2 to 4 in. tall increased juniper mortality to 30, 44 and 45%, respectively. The same fire intensity applied near San Angelo when soils were dry killed 40% of the junipers, and broadcast sprays of picloram at 0.25, 0.5 and 1.0 lb/acre applied when regrowth was 2 to 4 in. tall increased redberry juniper mortality to 64, 79, and 95%, respectively (D.N. Ueckert, unpublished data). These results indicate synergism since picloram sprays alone rarely kill redberry junipers.

Researchers at Uvalde are evaluating the feasibility of using herbicides such as picloram (Tordon 22K) and paraquat (Gramoxone) prior to prescribed burning in Ashe juniper communities. The objective of this approach is to generate Ashe juniper crown fires by decreasing the moisture content of the juniper leaves, hence increasing their flammability (M.K. Owens, pers. comm.). Researchers in Oklahoma have shown that applications of paraquat sprays prior to prescribed burns in the spring significantly increased the damage to eastern redcedar (J. virginiana) (Engle et al. 1988). Bryant et al. (1983) found that crown fires could be generated in live Ashe junipers by burning windrows of dozed juniper that had been pushed into live Ashe juniper stands. Optimal conditions for crown fires were: wind speeds exceeding 10 miles/hr, juniper canopy cover greater than 35%, relative humidities of 20 to 40%, and air temperatures of 36 to 90°F. The crown fires usually stopped where distance between the ashe junipers exceeded 23 to 33 ft.

Using individual-plant treatments, such as leaf sprays of picloram or soil applications of hexazinone or picloram, for maintenance control of juniper regrowth after prescribed fire is another potential avenue for the effective integration of fire with proven herbicide technology. Individual plant treatments, such as the Brush Busters "leaf spray" (McGinty and Ueckert 1996), are much less labor intensive and require less herbicide after the stature of brush has been reduced by fire.

Biological Control

Biological control of redberry and Ashe juniper by the classical approach, i.e. introduction of alien control agents, is not considered a viable alternative because: these junipers are native rather than introduced plants; because Ashe juniper has beneficial uses (fence posts and aromatic oils); these junipers have value for wildlife habitat and food; and these junipers are so closely related to other junipers valued as ornamentals and for lumber production (DeLoach 1980). The "native plant" status of redberry and Ashe juniper does not rule out the possibility that alien control agents might successfully control these species. One of the classical examples of biocontrol of a native plant by introduced organisms often mentioned in the biological control literature is the "essential annihilation of Bermuda cedar (Juniperus bermudiana) on some islands of Bermuda by the accidentally introduced scale insects Carulaspis visci and Lepidosaphes newsteadi" (Huffaker 1957). The Bermuda cedars were highly susceptible even to fairly low densities of the scale insects.

Even though junipers are protected from grazers and insects by secondary plant compounds called monoterpenes that render them of relatively low palatability and that interfere with their digestibility (Launchbaugh et al. 1997; Taylor et al. 1997), it has been known for 40 to 50 yr that goats can effectively control junipers and other brush species that are relatively unpalatable to other livestock species (Magee 1957). Research at Sonora has shown that the winter diets of goats generally include about 10% juniper (in some years as much as 20 to 30%) and that Spanish goats are much more efficient than Angora goats for controlling woody plants (Straka and Taylor 1994).

Critical elements in successfully controlling juniper with goats include reducing the stature of the junipers to within reach of the goats and grazing with a high ratio of goats to juniper. Achieving these critical elements on rangelands supporting dense stands of mature junipers requires initial treatments such as chaining, tree dozing, or in isolated cases fire, to reduce the stature of the junipers so they can be defoliated by goats and to reduce the volume of juniper foliage available. Secondary treatments, such as prescribed burning, individual-plant herbicide treatments, or hand cutting or grubbing may then be necessary to further reduce the densities of live juniper so that a high goat:juniper ratio can be achieved for the number of goats available.

The obvious potential hazard in using goats for juniper control is that the desirable browse species may be excessively grazed (Scifres 1980; Nelle 1997). Use of the desirable browse plants should be carefully monitored when using goats for juniper control.

Research at Sonora during 1969-1974 demonstrated that 2-way chaining followed by alternate grazing with Spanish goats, cattle, and sheep at 45 animal units/section (85% of the animal units consisting of Spanish goats) reduced juniper cover by an average of 97% (Merrill and Taylor 1976). Total canopy cover of all woody plants and pricklypear was reduced 83% during this 5-yr period.

The foliage of juniper seedlings and regrowth contains lower concentrations of monoterpenes than that of mature junipers (Straka and Taylor 1994). This finding helped explain observations that goats preferred juniper seedlings and regrowth over mature juniper foliage. Riddle et al. (1996) found that Spanish goats consumed more juniper than did Angora goats and that both breeds consumed more Ashe juniper than redberry juniper. They concluded that goats probably ate less redberry juniper because it contained higher concentrations of the essential oils sabinene + -pinene and myrcene than did Ashe juniper. Their findings indicate that Spanish goats will be more effective than Angora goats for controlling junipers, and that goats will be more effective for controlling Ashe juniper than redberry juniper.

We have observed severe defoliation of redberry junipers in pastures near San Angelo where the Angora goat to redberry juniper plant ratio was high. Although severe defoliation will not kill mature redberry junipers, goating during the winter, when grasses and forbs are dormant, could provide substantial control of very young redberry juniper seedlings still in the "needle leaf" stage whose meristematic regions are not yet covered by soil or mulch. Simulated browsing (hand clipping) of redberry juniper seedlings averaging 4 in. in height (range 2 to 7.5 in.) to ground level killed 82% of the seedlings (D.N. Ueckert, unpublished data). Simulated browsing to 0.4 in. above ground killed 53% of the seedlings, and removal of half the foliage from juniper seedlings resulted in 15% mortality.

In "cafeteria-style" feeding trials, we found that Spanish goats generally preferred to eat redberry juniper seedlings that had developed "scale" leaves compared to younger seedlings that were in the "needle-leaf" stage. Six out of 10 Spanish goats consumed significantly more scale-leaf juniper than needle-leaf juniper during three, 2-hr feeding trials, even though the concentrations of volatile oils were greater in scale-leaf compared to needle-leaf junipers (D.N. Ueckert and J.L. Petersen, unpublished data). These findings suggest that the sharp tips on needle-stage juniper leaves may protect young redberry juniper seedlings in the needle-leaf stage from browsing by goats. Therefore, the most effective time to use goats to kill redberry junipers may be soon after seedlings emerge during those autumn-early spring periods with above-average precipitation.

Field grazing trials near San Angelo, Texas have demonstrated the potential short-term effects of heavy Spanish goat grazing upon populations of small redberry junipers. A single high-intensity short-duration grazing event with Spanish yearlings during winter (290 to 580 goat-days/acre) reduced the abundance of needle-leaf seedlings 25 to 41% and all juniper seedlings 20 to 28% (D.N. Ueckert and J.L. Petersen, unpublished data). The abundance of needle-leaf seedlings was reduced 34 to 39%, and that of all juniper seedlings was reduced 30 to 34% by grazing at these stocking rates with Spanish goats during the winters of 2 consecutive years. Average pre-treatment heights of redberry juniper seedlings in these trials were 4.0 in. (needle-leaf plants) and 9.1 in. (scale-leaf plants). All larger junipers had been removed from the experimental plots prior to these grazing trials to achieve a high goat:juniper ratio.

Using Our Knowledge on Juniper Biology to Achieve Management Objectives

Knowledge generated by scientists on juniper biology and ecology is critical for designing economically sustainable juniper management strategies. "Population modeling" is a scientific technique that allows us to synthesize and apply the existing biological information in developing more effective management strategies. Population models facilitate the organization and synthesis of large amounts of information and the identification of knowledge gaps where research activities should be focused. This approach allows scientists to develop management strategies that focus on the weak link - or Achilles heel - of juniper's life cycle (Whisenant 1991).

The life cycle of juniper includes four stages: seeds, seedlings, non-reproductive juveniles, and adults. Population models are a series of equations that describe the growth and development of each stage and the transition between life stages. The information critical to modeling include seed production, female-to-male ratio, seed predation, seed death, germination requirements, soil factors, precipitation, grass cover, competition, etc. Modeling allows us to examine the biotic and abiotic factors that affect mortality and the rate of change from one life stage to another.

Preliminary modeling efforts on Ashe juniper at Texas A&M University suggested that our management activities would be most effectively focused on strategies that reduce seedling establishment, such as fire, goats, and competitive interactions with grass (Whisenant 1991). If the juniper infestation consists of dense stands of mature trees, we obviously must initially focus on controlling the mature trees, then implement strategies to reduce establishment of new plants. Juniper population models will undoubtedly become increasingly important in developing juniper management strategies in the future.

A recently developed model for Ashe juniper indicated that maintenance of grass-dominated communities requires cool-season fires at a return interval of less than 25 years, whereas the return interval could be longer with warm-season fires (Fuhlendorf et al. 1996).

Economics of Redberry Juniper Control

Juniper control must be economically feasible to be acceptable to rational and profit-maximizing land managers. Land owners and managers invariably seek to maximize returns, minimize costs, or both when investing in range improvements such as the control of juniper. A recent in-depth economic analysis of redberry juniper control concluded that control was economically feasible (Gerbolini 1996). The study evaluated 2-way chaining as the initial juniper control practice, with prescribed burning 3 years after chaining and periodically thereafter over a 30-year planning horizon. Variables evaluated in sensitivity analyses included: yearly rate of increase in juniper canopy cover (1.6 - 5%); initial juniper canopy cover (10-30%); weighted price of livestock ($0.75 - $0.89/lb); discount rate (4.9 - 10.8%); cost of chaining ($15.21 - $18.25/acre); and cost of burning ($3.53 - $4.24/acre). Juniper control by 2-way chaining with periodic prescribed fire was determined to be economically feasible under all the variables specified above. The investment's total net present value becomes positive in 4 to 8 years. Total net present value at optimum burning cycles of 6 to 7 years for the 30-year planning horizon was estimated to be about $37/acre on very shallow range sites and about $103/acre on sandy loam range sites. The internal rates of return were 27% and 63% for very shallow and sandy loam range sites, respectively (Gerbolini 1996).

The results from Gerbolini's (1996) economic analysis of juniper control were not consistent with those from 3 other Texas studies (Vantassel and Conner 1986; McPherson 1987; Rowan and Conner 1994). Reasons for discrepancies in conclusions among the economic studies include the use of different relationships between juniper canopy cover with forage production, different rates of increase in juniper canopy cover over time, the use of different livestock prices, whether or not labor savings following juniper control was factored in as reduced cost, and different planning horizons.

Another recent economic study (Reinecke et al. 1997) compared the economics of controlling 6 different degrees of Ashe juniper infestations, varying from a sparse stand of saplings (90/acre; 3% juniper canopy cover) to a very dense stand (46 large junipers/acre + 317 medium-size junipers/acre + 1,022 small junipers/acre; 22.5% juniper canopy cover). The researchers concluded that net cash flow could be maximized by controlling Ashe junipers during relatively early stages of invasion which were susceptible to repeated prescribed fires or to a combination of individual plant herbicide treatments and prescribed fires. Net cash flows dramatically decreased, and treatment costs dramatically increased for juniper situations that required initial mechanical treatments with repeated prescribed fire follow-up treatments. The economics of controlling all degrees of Ashe juniper infestations were more favorable under rotational grazing than continuous grazing because there were no deferment costs where rotational grazing was practiced.

To know that only 2 of 5 economic analyses showed juniper control in Texas to be economically feasible causes a lot of disconcertedness to those of us who have witnessed productive grasslands and savannahs become essentially worthless juniper woodlands or "cedar brakes" and to permanently loose potential to produce useful products. However, the economic studies should stimulate landowners and rangeland resource managers to set aside their old paradigms about brush control and begin to think as ecologists to successfully manage their juniper. Those who have ignored their juniper and now are plagued with "cedar brakes" may simply have to invest more capital in their juniper control than they will be able to recover in their lifetime if their goal is to turn the land resource over to their heirs in better condition than it was when they acquired it. The uncertainty of economically feasible means for controlling mature juniper infestations should be a stimulus to landowners to control juniper invasions or reinfestations in their early life stages, when they are easy to kill and most susceptible to the ecological tools and/or individual plant treatments that cost little.

Some economic analyses of juniper control may show less-than-favorable feasibility because they assume livestock production is the only ranch enterprise. Most ranches also have the potential for generating revenue through wildlife and recreational enterprises, and a substantial percentage of the landowners in Texas are capitalizing upon these opportunities. Juniper management may be essential to develop, maintain, or improve these wildlife and/or recreational enterprises. Thus, present and future revenues from these additional ranch enterprises should improve the economics of juniper control. Another aspect to consider is the impact of juniper control on the real estate and aesthetic value of the land investment. Excessive juniper control may actually detract from real estate value of ranchland, but careful sculpting of the landscapes to create mosaics of grasslands, savannahs, and juniper woodlands will invariably enhance land values. This should be an important consideration to those who plan to convert their real estate to cash for retirement.

Summary

A brief overview of alternative technologies (mechanical, herbicidal, fire, and biological) for management of redberry and Ashe juniper has been presented, along with information on how effective juniper management systems may be developed through the proper selection and sequencing of 2 or more of these technologies in a long-range plan. There are no magic bullets for juniper eradication, but technologies for sustainable juniper management systems are available. The solution to the juniper problem is "site specific", i.e. the best treatment or series of treatments will vary from one ranch to another as well as among pastures or among range sites within a pasture on a given ranch. The "best" juniper management system depends upon the nature of the juniper infestation, availability of capital, the land owner's (or manager's) attitudes or personal preferences, and several other factors. Technology is available to allow landowners to selectively control junipers where they occur in association with desirable browse plants, to thin junipers to an acceptable density, and to create grasslands interspersed with juniper savannahs and juniper woodlands. This "sculpting" of landscapes allows the land owner or manager to optimize the value of his resource for livestock, wildlife, aesthetics, recreation, and real estate.

Ecologically sound and economically sustainable juniper management systems must be based upon knowledge of its biology and ecology, using good grazing management, fire and goats to the fullest extent possible, and upon selecting and sequencing alternative technology to maximize synergisms among the various treatments. Successful long-term juniper management will be achieved by those who most effectively implement strategies that reduce seedling establishment, which is the weak link of juniper's life cycle.

Literature Cited

Bryant, F.C., G.K. Launchbaugh, and B.H. Koerth. 1983. Controlling mature ashe juniper in Texas with crown fires. J. Range Manage. 36:165-168.

Dye, K.L., II, D.N. Ueckert and S.G. Whisenant. 1995. Redberry juniper - herbaceous understory interactions. J. Range Manage. 48:100-107.

DeLoach, C.J. 1980. Prognosis for biological control of weeds of southwestern U.S. rangelands, p. 175-199. In: Proc. V Int. Symp. Biol. Contr. Weeds, Brisbane, Australia, 1980.

Engle, D.M., J.F. Stritzke and P.L. Claypool. 1988. Effect of paraquat plus prescribed burning on eastern redcedar (Juniperus virginiana). Weed Technol. 2:172-174.

Fuhlendorf, S.D., F.E. Smeins, and W.E. Grant. 1996. Simulation of a fire-sensitive ecological threshold: a case study of Ashe juniper on the Edwards Plateau of Texas, USA. Ecol. Model. 90:245-255.

Gerbolini, A.J. 1996. Economic evaluation of redberry juniper control in the Texas Rolling Plains. M.S. Thesis, Dept. of Agr. Econ., Texas Tech Univ., Lubbock, Tex. 119 p.

Huffaker, C.B. 1957. Fundamentals of biological control of weeds. Hilgardia 27(3):101-157.

Jenkins, R.B. 1939. Cedars and poverty. The Cattleman. 1939. p. 61-62.

Launchbaugh, K., C. Taylor, E. Straka and R. Pritz. 1997. Metabolic basis for juniper consumption by livestock. In: Proc. 1997 Juniper Symposium. Jan. 9-10, 1997. San Angelo, Tex. Texas Agr. Exp. Sta.

Magee, A.C. 1957. Goats pay for clearing Grand Prairie rangelands. Texas Agr. Exp. Misc. Pub. MP-206.

McFarland, M.L. and D.N. Ueckert. 1982. Honey mesquite control: use of a three-point-hitch mounted, hydraulically assisted grubber. Texas Agr. Exp. Sta. Prog. Rept. PR-3981.

McGinty, A. and D. Ueckert. 1996. Brush Busters how to master cedar. Three safe and effective ways to control small cedar (juniper). Texas Agr. Ext. Service and Texas Agr. Exp. Sta. Leaflet L-5160.

McGinty, A. and D. Ueckert. 1997. Brush Busters: a program to market tactical brush management technology to landowners. In: Proc. 1997 Juniper Symposium. Jan. 9-10, 1997. San Angelo, Tex. Texas Agr. Exp. Sta.

McPherson, G.R. 1987. Plant interaction in redberry juniper-mixed grass communities. Ph.D. Diss., Dept. Range & Wildlife Manage., Texas Tech Univ., Lubbock, Tex.

Merrill, L.B. and C.A. Taylor. 1976. Take note of the versatile goat. Rangeman's J. 3(3):74-76.

Nelle, S. 1997. Holistic perspective on juniper. In: Proc. 1997 Juniper Symposium. Jan. 9-10, 1997. San Angelo, Tex. Texas Agr. Exp. Sta.

Rasmussen, G.A., G.R. McPherson, and H.A. Wright. 1986. Prescribed burning juniper

communities in Texas. Dept. of Range & Wildlife Management, Texas Tech Univ., Lubbock, Tex. Manage. Note 10. 6 p.

Rasmussen, G.A. and H.A. Wright. 1989. Succession of secondary shrubs on ashe juniper communities after dozing and prescribed burning. J. Range Manage. 42:295-298.

Reinecke, R.K., J.R. Conner, and A.P. Thurow. 1997. Economic considerations in Ashe juniper control. In: Proc. 1997 Juniper Symposium. Jan. 9-10, 1997. San Angelo, Tex. Texas Agr. Exp. Sta.

Riddle, R.R., C.A. Taylor, M.M. Kothmann, and J.E. Huston. 1996. Volatile oil contents of ashe and redberry juniper and its relationship to preference by Angora and Spanish goats. J. Range Manage. 49:35-41.

Rowan, R.C., and J.R. Conner. 1994. Economic considerations in the control of Ashe juniper, p. 69-78. In: Proc. 1994 Juniper Symposium. Texas A&M Univ. Agr. Res. Sta. Tech. Rep. 94-2. Sonora, Tex.

Steuter, A.A. and C.M. Britton. 1983. Fire- induced mortality of redberry juniper (Juniperus pinchotii Sudw.). J. Range Manage. 36:343-345.

Scifres, C.J. 1980. Brush Management Principles and Practices for Texas and the Southwest. Texas A&M University Press. College Station, Tex. 360 p.

Straka, E. and C.A. Taylor. 1994. Using goats to manage brush. Ranch & Rural Living Mag. 75(5):19-20.

Taylor, C., K. Launchbaugh, E. Huston, E. Straka and R. Pritz. 1997. Improving the efficacy of goating for biological juniper management. In: Proc. 1997 Juniper Symposium. Jan. 9-10, 1997. San Angelo, Tex. Texas Agr. Exp. Sta.

Ueckert, D.N. and S.G. Whisenant. 1982. Individual plant treatments for controlling redberry juniper seedlings. J. Range Manage. 35: 419-423.

Vantassel, L.W. and J.R. Conner. 1986. An economic analysis of brush control practices and grazing systems in the Rolling Plains of Texas. Texas Agr. Exp. Sta. Misc. Pub. MP-1619. 13 p.

Welch, T.G. 1995. Chemical weed and brush control suggestions for rangeland. Texas Agr. Ext. Service Bull. B-1466. 27 p.

Whisenant, S.G. 1991. An approach to population modeling for developing brush management strategies, p. 176-178. In: Proc. 4th International Rangeland Congress. Apr. 22-26, 1991. Montpellier, France. Association Francaise de Pastoralisme.

Wiedemann, H. 1997. Elevated chaining of juniper. In: Proc. 1997 Juniper Symposium. Jan. 9-10, 1997. San Angelo, Tex. Texas Agr. Exp. Sta.

Wiedemann, H.T. and B.T. Cross. 1982. Performance of front-mounted grubbers on rubber-tired equipment. Texas Agr. Exp. Sta. Prog. Rept. PR-3982.

Wiedemann, H.T. and B.T. Cross. 1996. Draft requirements to fell juniper. J. Range Mange. 49:174-178.

Wink, R.L. and H.A. Wright. 1973. Effects of fire on an ashe juniper community. J. Range Manage. 26:326-329.

Wright, H.A., F.M. Churchill and W.C. Stevens. 1976. Effect of prescribed burning on sediment, water yield, and water quality from dozed juniper lands in central Texas. J. Range Mange. 29:294-298.