Holistic perspective on juniper

Steve Nelle


People sometimes have tunnel vision on controversial or complicated subjects. Tunnel vision means viewing something with a close-up lens, looking at the object in detail, but being unable or unwilling to see the surroundings and associated parts. A holistic perspective means viewing something with a wide-angle lens, looking at an object and all of its parts in context with its surroundings.

Many people view juniper with tunnel vision, looking closely at some aspects while ignoring others. The purpose of this paper is to take a wide-angle, holistic look at some of the important but often overlooked relationships of juniper on rangeland.

Juniper: Natural Woodland or Grassland Invader?

One of the most basic aspects of understanding juniper is knowing to what extent it is a native, climax plant which belongs on the range or a noxious invader which does not belong. Range managers almost universally regard juniper as an invader wherever it is abundant. Range site descriptions developed by NRCS over the past 30 years portray the climax or potential plant communities of the entire Edwards Plateau as grasslands or savannas with scattered trees and shrubs. Juniper is only rarely listed as a climax species and then only in very small amounts. This grassland bias that the range profession has embraced has contributed to an incomplete understanding of the role of juniper and other rangeland trees and shrubs.

Table 1.

Reference to juniper Author County Year
“thickets of cedar” de Miranda Blanco 1756
“dense thickets of cedar” Moore Menard 1840
“dense forest of cedar” de Cordova Burnet 1858
“cedar covered mountains” Harrell Travis 1838
“cedar which covered the slopes” Roemer Comal 1849
“the cedar forest” Berlandier Kerr 1828
“mountains covered with cedar” de Cordova Bandera 1858
“the forest is composed of cedar” Berlandier Uvalde 1828
“heavy cedarbrake for six miles” Johnson ??? 1853
“cedar brake” Dewees Lampasas 1830
“the mountains are cedar bedekt” Claus Kendall 1855
“dense thickets of cedar” Nystel Bosque 1867

Despite assertions to the contrary, it is abundantly clear from the historical record that vast areas of Ashe juniper woodlands were present over much of the Edwards Plateau prior to white settlement (Smeins 1980). Forests or woodlands dominated by juniper and other woody species are apparently the climax, natural plant community for many of the steep and/or very shallow sites of the region. Several of the many references to widespread juniper in the Edwards Plateau prior to range deterioration are provided in Table 1.

It is also equally clear that grasslands and savannas are the historical climax plant communities on many other sites within the Edwards Plateau such as the gently rolling areas of deeper soil. On these naturally open or moderately wooded sites, juniper is truly an invader. A healthy cover of grass and the associated fires maintained an open or semi-open landscape.

So, the climax natural vegetation of the Edwards Plateau was neither a vast regional grassland, savanna, woodland nor forest. All of these vegetation types are climax and natural on different sites within the region.

Early settlers recognized the value of this natural mixture of vegetation types but did not know how to manage them. Juniper forests and woodlands were considered a commodity of great worth. Numerous enterprises were built around the many products derived from mature juniper (Bray 1904). As these woodlands were exhausted by non-sustainable harvest rates, grasses grew on the slopes creating a grazing resource (Weniger 1984). Conversely, the climax grasslands were valued for the large numbers of livestock that could be grazed. As the grasslands were exhausted by non-sustainable harvest rates, juniper invaded (Taylor and Smeins 1994). Now juniper seems to be everywhere, obscuring the previous distinction between climax grasslands and climax woodlands.

First Ask Why

Before a good strategy can be devised to combat juniper where it has invaded, we must ask why did juniper invade. It has been said that the cause of a problem suggests its cure. Some of the earliest observers accurately described the invasion of juniper into former grasslands and savannas as well as the cause of the invasion.

“As one passes from the canyons and hills to the level plateau divides, the timber gives place more and more to open prairie, which until within recent years, was free from woody growth of any kind. But these ranges have been overpastured, and the grass has consequently not succeeded in maintaining a continuous sod, and so has become unable to wage an equal fight against the shrubs … Again, settlement has stopped the periodic burning off of the grasses, which … drove out the shrubs and prevented the timber from gaining on the prairie.” (Bray 1904).

Overgrazing caused and continues to cause juniper invasions in two ways. As the grass was thinned out by heavy grazing, juniper seedlings had greater opportunity to establish and fires had less opportunity to burn. The original thicker cover of grass not only inhibited the germination and establishment of juniper, but allowed periodic large, hot fires to burn and keep cedar out.

Another factor in the invasion of juniper, especially redberry, is the relationship between grazing management and jackrabbits. Jackrabbits are a fairly reliable indicator of overgrazing. They tend to be abundant on heavily grazed, low condition range and uncommon on properly grazed, high condition range (Taylor and Buechner 1943). Jackrabbits play a major role in dispersal of redberry juniper seeds from wooded canyons into adjacent grasslands. Jackrabbit pellets were found to contain a large number of seeds in fall and winter when about half of the jackrabbit diet was thought to be the fruits of redberry juniper (Wolff 1948). In this same study, jackrabbits deposited seed laden pellets 1000 and less commonly 2000 feet from juniper stands. The combination of high numbers of jackrabbits on overgrazed range and their consumption and dispersal of juniper seeds magnifies the relationship between poor grazing management and juniper invasion.

The possible relationships between coyote control; the decline in fur trapping; the proliferation of berry-eating fox, raccoons, and skunks; and the spread of juniper have not been determined.

Its Ecological Roles

All native plants have natural functions which are important to the long term health and stability of the land. Where juniper woodlands are the natural climax vegetation on steep, thin soil sites, its function apparently includes protection from severe erosion and even the building of topsoil. Other important functions are discussed by Rollins and Armstrong (1994).

Even on sites where juniper has invaded into former grasslands and savannas, it is still fulfilling a natural ecological role. If mismanagement has allowed the loss of the protective cover of grass, juniper invades and provides some soil protection. The buildup of leaf litter adds organic matter, catches sediment and increases moisture holding capacity (Thurow and Carlson 1994). The physical barrier of juniper branches inhibit grazing and browsing of other plants which become established under and around the perimeter of juniper, thus protecting a seedbank. These are some of the reasons why such a good plant response is often seen after juniper control.

Much of the current emphasis on juniper control revolves around its role in the water cycle. Thick stands of juniper intercept large amounts of rainfall and utilize considerable soil moisture which decreases deep percolation, springflow and aquifer recharge. However, the removal of juniper does not automatically improve these watershed dynamics and could in fact make them worse.

If juniper control is followed by long term judicious management which promotes a good grass response, then watershed functions (runoff quality and subsurface quantity) should improve. However, a good cover of mid grasses also intercepts rainfall and utilizes soil moisture, so the net gain in aquifer recharge may be negligible.

If juniper control is not followed by proper grazing management, a good grass cover will not develop. Runoff will increase and deep percolation will be minimal. Much past juniper control has resulted in massive soil erosion (Marsh and Marsh 1992). A dense cover of juniper is better watershed protection than a poor cover of grass.

Methods of Warfare

Ever since juniper began to invade into the former grasslands and savannas, ranchers have tried every conceivable way to combat it. By far, the most common methods have been various mechanical means, many of which are highly effective. Chemicals have also been tried, some successfully, but these have played an insignificant role thus far.

For many years, ranchers have used goats to control juniper and other shrubs. In fact goats have been described by some as “nature’s herbicide”. A pasture can intentionally be converted from a shrubland to a grassland or savanna in a few years with the practice known as goating. All shrubs within reach of goats can be killed in this manner. The best browse plants such a hackberry, bumelia, elbowbush and redbud are the first to succumb to heavy browsing. Next, the medium quality plants such as oaks and sumacs will be killed. After the best and then the mediocre plants are eliminated, goats will turn to such unpreferred plants as juniper, persimmon, lotebush and algerita. If these less desirable “brush” plants are heavily browsed repeatedly, they too can be controlled or at least greatly suppressed.

This intentional use of goats to kill woody plants can be a legitimate practice when carefully managed, for those ranchers whose long term goals are to increase grass production and grazing capacity for cattle. The trade-off is that the value of the range for goats and deer and other wildlife will be greatly diminished for many years.

The unintentional loss of browse plants due to years of heavy goat stocking has had a far greater and more detrimental impact than has intentional goating. Traditional excessive stocking rates and unbalanced ratios of goats, deer, sheep and cattle have caused a slow but steady deterioration of the browse resource on many ranches. In these cases it is very common for the better browse plants to be eliminated while poor browse plants such as juniper proliferate (NRCS 1994). Sadly enough, it has often been the recommendations of range professionals within agricultural and conservation agencies which has contributed to these excessive stocking rates.

Well known Texas rangeman John “Chip” Merrill has said that in order for a range management practice to be good and proper, it must be both economically and ecologically sound. The use of goats to control juniper may fail both of these tests.

Goat numbers in Texas peaked during the 1960’s when they ranged between 3 and 4 million head. Since 1985 goat numbers have remained at about half that number (TASS 1990). With the increase of predators and other factors, the raising of large numbers of goats was deemed uneconomical to many Texas ranchers even with the mohair incentive. Now with the loss of the incentive program and the increasing value of good deer habitat, the economic prospect for making goats a large part of one’s livestock inventory is even less appealing.

Even if it were economical to raise enough goats to control juniper, the ecological soundness of this practice is very debatable. Juniper is among the least palatable, least preferred, and lowest in nutrition of browse plants although it does have some yearlong forage value. Goats will eat significant amounts of juniper only when there is little else of greater value available. When goats (or deer or exotics) consume large amounts of juniper, it is usually a symptom of deteriorated rangeland and having too many animals on too few acres. This type of heavy goating whether planned or unplanned has done great harm to rangelands. The benefits of keeping pastures “clean” is more than offset by the detrimental effects of overbrowsing, overgrazing, erosion and range deterioration.

More recently, a less destructive approach of using goats to control juniper has been suggested. Since goats will voluntarily eat moderate amounts of juniper especially during dry winters, it is reasoned that biological warfare can be successfully waged on more vulnerable young plants and regrowth. Selection for goats with a greater tolerance of repulsive and toxic juniper compounds is underway to increase juniper consumption.

Advocates of this more environmentally friendly version of biological control are quick to point out that they do not force goats to eat abnormally high amounts of juniper. Only moderate stocking is recommended, primarily in winter months. Under these conditions, it is doubtful that goats will do little more than light pruning of small juniper with limited harm done to seedlings recruitment or older plants.

With goats and juniper, you must either stock heavily enough to harm the juniper but abuse the resource or stock moderately to sustain the resource but watch juniper continue to increase. Goats are not the answer to the juniper problem.

If a lack of grass cover and a lack of fire is what caused the great spread of juniper, then a return to an improved grass cover and periodic fires must be a key part in controlling juniper (Ueckert et al. 1994).

The essential keys in the fight against juniper are grazing management and fire. Fire is the Achilles heel of juniper which will bring it to its knees. Despite this, less than one-tenth of one percent of the Edwards Plateau is burned with prescribed fire each year (Nelle unpublished data). Reasons for the low application rate of burning include the lack of fuel, reluctance to burn forage, a fear of fire and liability.

Major mechanical treatments (hand cutting, grubbing, dozing, chaining, shearing, roller chopping) will often be needed initially to allow grass to respond and fires to burn. Individual plant treatment with herbicides has a valid place for scattered small juniper. The careful use of goats may have a place in pruning back small juniper and reducing the number of seedlings. However, brush control per se is not the solution to the juniper problem. We in the range profession have done a good job of promoting and cost-sharing juniper control, but we have done a poor job thus far of correcting the real problem – overgrazing and lack of fire.


A holistic viewpoint of juniper might be expressed in this way:

1. Juniper woodlands are the native, natural vegetation type on many sites.

2. Juniper has invaded into other grassland and savanna sites where it is not natural.

3. The primary cause of the invasion is overgrazing and the lack of fire.

4. Under conditions of overgrazing, the invasion of juniper is the natural ecological response which provides a degree of protection and rehabilitation to the site.

5. Watershed response to juniper control can be positive or negative depending on subsequent grazing management.

6. Where juniper control is desired, it should be preceded by a firm commitment to proper stocking rates, regular planned rest periods and periodic prescribed burning. Otherwise it will quickly return.

7. The use of goats or other browsers to control juniper usually has harmful side-effects of eliminating other more desirable species.

Literature Cited

Bray, W.L. 1904. The timber of the Edwards Plateau of Texas. USDA Bureau of Forestry Bulletin No. 49. 30 p.

Marsh, W.M. and N.L. Marsh 1994. Juniper trees, soil loss, and local runoff processes. In. Soils, landforms, hydrologic processes and land use issues. Field report and guidebook. Society of Independent Professional Earth Scientists. Central Texas Chapter. Austin, TX.

Natural Resources Conservation Service 1994. The use and managment of browse in the Edwards Plateau of Texas. Temple, TX. 8p.

Rollins, D. and B. Armstrong 1994. Cedar through the eyes of wildlife. pp 53-60. In. Juniper symposium. TAMU Research Station technical report 94-2. 80p.

Smeins, F.E. 1980. Natural role of fire on the Edwards Plateau. pp 4-16. In. Prescribed burning of the Edwards Plateau. TAES. 74p.

Taylor, C.A. and F.E. Smeins 1994. A history of land use of the Edwards Plateau and its effect on the native vegetation. pp 1-8. In. Juniper symposium. TAMU Research Station technical report 94-2. 80p.

Taylor, W.P. and H.K.Buechner 1943. Relationship of game and livestock to range vegetation in Kerr County, Texas. In. The Cattleman. March 1943.

Texas Agrcultural Statistcs Service 1990. Texas historic livestock statstics 1867-1990. Austin, TX. 92p.

Thurow, T.L. and D.H. Carlson 1994. Juniper effects on rangeland watersheds. pp31-43. In. Juniper symposium. TAMU Research Station technical report 94-2. 80p.

Ueckert, D.N., S.G. Whisenant, and M.K. Owens 1994. Juniper control and management. pp61-68. In. Juniper symposium. TAMU Research Station technical report 94-2. 80p

Weniger, D. 1984. The explorers’ Texas. Eakin Press. Austin, TX. 224 p.

Wolff, S.E. 1948. An evaluation of some weedy Texas junipers. USDA-SCS. Ft Worth, TX 89p.

Comments: Allan McGinty, Professor and Extension Wildlife Specialist

Comments are closed.