TA245 - 7 33133;; 1978

873
0.1191 y

i

Range
Vegetation
After

Mechanical

Brush
Treatment

on
the

Coastal

Prairie

The Texas Agricultural Experiment Station, Neville P. Clarke, Director, College Station, Texas, The Texas A&M University System

[Blank Page in Original Bulletin]

Contents

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . 2

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Brush Management in South Texas: An Overview . . . . . . . . . . . . . . . . . . . .. 3

The Study Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5

Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 6

The Original Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 6

Procedures for 1977 Evaluations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7

Brush Cover and Composition . . . . . . . . . . . . . . . . .  . . . . . . . . . . . . . . . .. 7

Herbaceous Vegetation and Range Condition . . . . . . . . . . . . . . . . . . . . .. 10

Management Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Literature Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Authors

Mutz, I. L., assistant range scientist, The Texas Agricultural Experiment Station,
Corpus Christi

Scifres, C. ]., professor, The Texas Agricultural Experiment Station (Department
of Range Science)

Drawe, D. L., assistant director, The Rob and Bessie Welder Wildlife Founda-
tion, Sinton, Texas

Box, T. W., dean, College of Natural Resources, Utah State University, Logan

Whitson, R. E., associate professor, The Texas Agricultural Experiment Station
(Departments of Range Science and Agricultural Economics)

Summary

In 1963, several mechanical brush management practices were installed on a
Blackland range site which supported a "Chaparral-bristlegrass" community of
the Rob and Bessie Welder Wildlife Refuge near Sinton. The original plots have
since been maintained free of major disturbances (except for a maintenance burn
in 1974), allowing a direct comparison of vegetation changes after 14 years. In
1971, huisache on a portion of each plot was removed by oiling, allowing evalua-
tion of its impact on vegetation improvement. Vegetation changes were assessed
relative to changes in botanical composition. Although botanical composition
changes have occurredlin brush cover during this period (reflecting major
changes in rainfall pattern), the total canopy cover in 1977 was remarkably
similar to that in 1963. With increasing wetness since 1963 and associated ex-
tended inundation of lowlands and swales, a portion of the honey mesquite
cover has been replaced by huisache, lime pricklyash, blackbrush acacia,
bluewood, and other species, regardless of treatment.

Roller chopping and shredding, for the most part, did not change brush
canopy cover or range condition appreciably after 14 years even though a

maintenance burn was installed in 1974. Based on past research accounts and |

present stature of the woody plants, life of shredding or roller chopping treat—
ments probably did not exceed 5 years, using shrub canopy coverage as a crite-
rion. Improvement in herbaceous vegetation may last somewhat longer but
probably does not exceed 7 years. Extent of improvement following scalping the
rangeland with a ’-’KG" blade was intermediate between that of the simple top
removal treatments and root plowing.

Reduction of brush canopy cover on root-plowed areas and those treated
with root plowed and raking combined averaged 57 percent in 1977, based on
comparison with adjacent untreated areas. However, successional change of
herbaceous vegetation was relatively slow on root-plowed rangeland. Composi-
tion of herbaceous vegetation following initial soil disturbance from root plow-
ing in 1963 was roughly equivalent to that of untreated plots in 1977. Huisache
increased in importance over the study period, regardless of treatment, but the
relative change was most dramatic following root plowing. Removal of huisache
in 1971 by oiling allowed development of only a 5- to 8-percent total canopy
cover of brush compared to 27.7 percent on areas root plowed only and to 48.6
percent on untreated plots.

Present livestock carrying capacities, estimated as a function of herbaceous
composition and brush cover, differ among treatments, primarily because of
differential brush control effectiveness. Where the only treatment was huisache
oiling, carrying capacity is estimated to be one animal unit per 23.5 acres year-
long. On shredded, roller-chopped, and scalped areas where the huisache was
oiled, carrying capacity is one animal unit per 15.5 to 16 acres. Carrying capacity
on root-plowed areas is one animal unit per 14.1 acres and on areas root plowed
and raked is one per 12.6 acres. The importance of huisache on livestock produc-
tion on the Coastal Prairie is exemplified by reduced carrying capacities follow-
ing simple top removal methods or scalping (one animal unit per 22.4 to 22..
acres) and following root plowing (one animal unit per 17.9 acres) where the”
huisache was not oiled.

Range Vegetation
After Mechanical Brush Treatment

on the Coastal Prairie

Brush and its management on rangeland is a prob-
lem that can best be approached from an ecological
perspective within a well-defined economic
framework (Scifres, 1977). The goal of brush man-
agement is to manipulate the range vegetation such
that optimum proportions and amounts of native
plants are restored and sustained relative to potential
for a given range site. Thus, brush management is
applied to expedite secondary succession on range-
land. Compared to row crop agriculture (mainte-
nance of a monoculture, usually of annual plants, by
depending on a system of intensive inputs) which
strives to retard succession, success in range im-
provement depends on a low level of synthetic inputs
applied over a relatively long period. Consequently,
long-tenn vegetation change most accurately reflects
the degree of success from range improvement treat-
ments. Monitoring of vegetation change on range-
land through time following manipulation is the only
means of fully understanding the ecological impact
and potential utility of any brush management prac-
tice. However, maintaining continuity in a brush
management project over several years is difficult.
Rarely does one have the o portunity to evaluate
vegetation 10 or more years a ter installation of range
improvement practices.

Initial influence of the selected brush management
techniques, shredding, roller chopping, scalping with
”KG" blade, root plowing, and the combination of
root plowing and raking, may be summarized as fol-
lows (Powell, 1966; Powell and Box, 1967):

1. The brush control methods which minimized
soil disturbance, shredding and roller chop-
ping, most effectively improved successional
stage and increased forage production for the
first and second years after treatment.

2. Soil disturbance from root plowing or scalping
retarded plant succession and caused large
fluctuations in annual forage production.

3. Brush cover was initially reduced by more than

50 percent by the treatments with greatest re-
ductions after root plowing and root plowing

5 followed by raking. Root-plowing treatments,

however, increased the cover of pricklypear.

Based on these results, Powell and Box (1967) con-
cluded that complete eradication of brush in south

Texas was not feasible. They suggested that a sys-
tematic brush management program would more
likely be successful in continually increasing range
forage quantity and quality.

The objective of this research was to evaluate and
discern the management implications of changes in
Coastal Prairie vegetation 14 years after application of
several mechanical brush control practices to a
mixed-brush (Prosopis-Acacia) infestation.

BRUSH MANAGEMENT IN SOUTH TEXAS:
AN OVERVIEW

Excessive woody plant cover lowers range forage
production and availability throughout the West.
However, nowhere is the interest in brush control
more keen than in the Rio Grande Plains and Coastal
Prairie, the ”brush country," of Texas. Since livestock
and wildlife are the primary economic products of
these rangelands, brush management programs are
directed towards optimizing economic returns by
maintaining a balanced mix of vegetation that will
sustain these different kinds of animals.

Until the early 1970's, most brush management in
south Texas was accomplished mechanically with
heavy equipment. Energy was relatively inexpensive,
and herbicides did not effectively control many of the
species in the mixed-brush complex. Although
2,4,5-T [(2,4,5-trichlorophenoxy)acetic acid] effec-
tively controlled honey mesquite) associated woody
species of the genera Acacia, Condalia, Lycium, Aloysia,
Zanthoxylum, Celtis, Diospyros and others were not ef-
fectively controlled at conventional rates of the
phenoxy herbicide. In fact, it was generally believed
that controlling honey mesquite alone with 2,4,5-T
released herbicide-tolerant species, creating more
troublesome management problems than originally
existed.

As new herbicides were developed, chemical con-
trol of south Texas brush appeared more promising.
A benzoic acid herbicide, dicamba (3,6-dichloro-0-
anisic acid), broadened the spectrum of species con-
trolled somewhat compared to phenoxy herbicides.
With the advent of picloram (4-amino-3,5,6-trichlo-
ropicolinic acid) and its commercial availability in
combination with 2,4,5-T, brush spraying became

lScientific names follow Gould (1975) and are presented in Appen-
dix for plants mentioned in text.

3

more widely adopted in south Texas. The introduc-
tion of new compounds such as tebuthiuron [N-(5-
[1, 1-dimethylethyl]-1,3,4 -thiadiazol-2 -yl)-N-N ’-di-
methylurea], effective for control of some woody
species which are not highly spsceptible to most
sprays, may lend impetus for expanded herbicide use
in the area (Scifres, Mutz, and Hamilton, 1978). Sys-
tems have been developed for chemical brush man-
agement which are compatible with maintenance of
quality wildlife habitat in south Texas (Beasom and
Scifres, 1977) and which are economically acceptable
(Whitson, Beasom, and Scifres, 1977). However, even
with the improved brush control levels following ap-
plications of 2,4,5-T + picloram, followup brush
management may be required within 5 years after
treatment (Scifres , Durham, and Mutz, 1977).

Prescribed burning offers potential for protracting
the initial brush control effectiveness achieved with
herbicides and reducing costs of repeated treatment
with chemicals (Box and White, 1969; Scifres and
Merkle, 1975; Scifres, 1975). Like herbicides, con-
trolled and prescribed burning are gaining favor for
range improvement in south Texas. Apparently,
range burning is considered economically superior to
other methods in certain situations and provides ben-
efits in addition to brush suppression (Gordon and
Scifres, 1977). Prescribed burning promotes use of
forages which are otherwise not utilized efficiently by
livestock (Oefinger and Scifres, 1977), improves food
availability (seed crops of herbaceous plants for ufp-
land game birds and succulent regrowth browse or
white-tailed deer [Odocoileus virginianus] [Gordon and
Scifres, 1977]) for wildlife, and may reduce the impact
of parasites on range animal populations. Continued
research and producer experience will undoubtedly
result in more widespread application of fire for range
improvement. However, even with recent technolog-
ical advances in herbicides and fire, many brush
management programs in south Texas continue to
focus on mechanical methods.

Mechanical brush management methods can be
categorized, relative to action on woody plants, as
those which are designed to simply remove current
year's topgrowth and as those which remove the
entire woody plant. Shredding and roller chopping,
methods of simple top removal, kill only those woody
species which are not capable of resprouting from
basal stem segments, roots, or rhizomes. Methods
such as grubbing, root plowing, and chaining are de-
signed to remove entire plants. However, if impro-
perly applied, chaining may also result in simple top
removal and should be followed by raking and stack-
ing of debris for maximum range improvement (Sci-
fres, Mutz, and Durham, 1976).

Variables such as treatment life expectancy (rela-
tive to brush cover replacement and concomitant car-
rying capacity adjustment) and initial costs vary con-
siderably among mechanical brush management
practices. Also, since degree of soil disturbance varies
among methods, the rate of improvement in herbace-
ous vegetation and associated livestock carrying

4

capacities vary within a given set of edaphic and
climatic conditions.

Although methods of simple top removal such a€
e

shredding and roller chopping cause minimal surfac
disturbance and are relatively rapid and inexpensive,
they typically result in relatively short-term brush
suppression when applied without provisions for fol-
lowup brush management procedures. For example,
huisache sprouts attained half their original height
within 5 months after top removal (Powell, Box, and
Baker, 1972). Although sprout growth rate was highly
dependent on rainfall, canopy replacement was
probably complete within 5 years. Significant biotic
suppression of sprout growth (increased use by in-
sects and mammals compared to their use of mature
branches) may occur the season of top removal.
White-tailed deer utilize succulent branch tips,
leaves, and mast crops of woody plants including
huisache, twisted acacia, and honey mesquite
(Drawe, 1968). Although only transient advantages,
browse value is improved and visibility on rangeland
is increased following simple top removal of brush
(Box and Powell, 1965; Powell and Box, 1966).

Contrasted to methods such as shredding or roller
chopping, root plowing causes maximum surface soil
disturbance, is relatively slow and expensive, but
usually results in comparatively long-term brush
suppression. Soil disturbance usually encourages an
influx of annuals, especially annual grasses and
broadleaved weeds, which may dominate the her-

baceous vegetation for two to three growing seasons, Q

depending on rainfall, soils, and the species and
abundance of residual perennial grasses. Thus, carry-
ing capacity may not be maximized until 5 or more
years after root plowing. The maximum treatment ef-
fect may then be realized for an additional 5 years.
Carrying capacity may then steadily decline for an
additional 5 to 7 years under over-grazing and no
followup improvement practices. Good grazing man-
agement and additional improvement efforts such as
prescribed burning may significantly increase the
longevity of maximum range improvement following
root plowing.

The ”KG” blade, similar to brush dozers, is de-
signed to skim the ground surface and shear off
woody plants leaving the top growth in stacks or
windrows (Drawe, 1977). This method removes the
brush topgrowth nearer the ground line than conven-
tional methods of simple top removal and causes
more surface soil disturbance.

The combination of prescribed burning and
mechanical brush control offers much promise in
south Texas. On the Welder Foundation, brush cover
was significantly reduced when areas were burned 4
years after mechanical treatment (Box and White,
1969; White, 1969). Fire may be effectively combined
with any mechanical method which releases herbace-
ous production to improve the fine fuel load and con
tinuity (Dodd and Holtz, 1972). Relatively cool fires
("maintenance burns") applied within 2 to 4 years
after installation of mechanical practices will usually

control woody resprouts and invading seedlings. If
woody plant infestations are relatively dense and
composed primarily of mature individuals, a hot fire
(‘reclamation burn”) may be required for range im-
provement.

THE STUDY AREA

The Coastal Prairie, tall grass prairie in the climax
state, is a major division of the Gulf Prairies and
Marshes Area (Gould, 1975). Climax grasses include
big bluestem, yellow Indiangrass, seacoast bluestem,
eastern gamagrass, various species of Panicum, and
gulf muhly. Herbaceous invaders include threeawns,
horsetail conyza, western ragweed, bushy bluestem,
tumblegrass, and whorled dropseed. Woody species
include honey mesquite, huisache, blackbrush acacia,
spiny hackberry, lotebush, bluewood, lime prickly-
ash, wolfberry, and several others.

Growing season of the Coastal Prairie is about 300
days. The area is typified by warm temperatures,
coastal air movements, and high relative humidity.
Rainfall averages about 37 inches annually, with
highs during late spring and early fall.

The study was located on a Blackland range site
supporting a "Chaparral-bristlegrass" community
(Box and Chamrad, 1966). The woody plant compo-
nent is typified by various mixtures of blackbrush
acacia, honey mesquite, huisache, twisted acacia,
agarito, creeping mesquite, spiny hackberry, lote-
bush, bluewood (often referred to as ”brazil" [Box
and Chamrad, 1966]), Texas persimmon and lime
pricklyash (Figure 1). Drawe, Chamrad, and Box
(1978), comparing site photographs taken in 1939
with those taken in 1960 and 1969, decided that these
woody plant stands are relatively stable. Herbaceous
vegetation underneath the woody mottes is generally
limited to scattered plains bristlegrass plants and
forbs. Vegetation between, the mottes on heavily
grazed areas is usually dominated by buffalograss,
Roemer threeawn, and filly panicum. Common
curlymesquite is also common between the brush
mottes. Under light grazing, silver bluestem, little

bluestem, meadow dropseed, Texas cupgrass, love-
grass tridens, Texas wintergrass, and other tall- and
midgrasses dominate. According to Drawe et al
(1978), silver bluestem, meadow dropseed, plains
bristlegrass, and sourgrass are the first grasses to in-
crease when stocking rates are decreased on over-
stocked areas. Species diversity and abundance of
forbs vary seasonally, but upright prairie-coneflower
and spotted beebalm usually occur on undisturbed
interspaces between the brush mottes. The high for-
age production during the growing season provides
excellent grazing, but nutritional value of the forages
is low during late fall and winter.

Plains bristlegrass, buffalograss, and common
curlymesquite apparently produce the bulk of forage
on the site (Box, 1960). Plains bristlegrass apparently
becomes of less importance, and filly panic becomes
one of the more important species from midsummer
(July) until early fall (September). During winter and
early spring, Texas wintergrass produces a substan-
tial amount of green forage. Based on harvests in 1957
and 1958, air-dry grass standing crop was approxi-
mately 2,100 pounds per acre for the fall-early winter
period (Sept.-Dec.), about 100 pounds per acre for
the late winter-spring period (Dec.-Mar.), 2,200
pounds per acre for the spring-summer period
(Man-June), and about 900 pounds per acre for the
summer-fall period (Iune-Sept.) (Box, 1960).

This community occurs on nearly level uplands
typified by Victoria clay. Other series present in
minor amounts may include Monteola clay, Orelia
sandy clay loam, Orelia fine sandy loam, or Willacy
fine sandy loam. Victoria clay is a member of the fine,
montmorillonitic, hyperthermic family of Udic Pellus-
terts. These soils have high water-holding capacities
and shrink-swell characteristics. The surface 6 inches
are dark gray clays with weak, very fine, subangular
blocky structure. A mulch of fine, discrete aggregates
occurs on the surface. The surface soil is typically
hard, highly plastic and sticky with a few fine
strongly cemented calcium carbonate concretions and
snail shell fragments, and is moderately alkaline.
These highly fertile soils are widely used as cropland.

Figure 1. Mixed-brush
stand in November 1977 on
the Rob and Bessie Welder
Wildlife Refuge where
huisache was oiled in 1971.

5

MATERIALS AND METHODS

The Original Study

In June 1963, three locations were selected in a
"Chaparral-bristlegrass” community on the Rob and
Bessie Welder Wildlife Refuge on the Coastal Prairie.
The refuge is located approximately 8 miles north of
Sinton in San Patricio County, Texas. The pasture
containing the study was grazed yearlong at the rate
of one steer per 17 acres and supported white-tailed
deer at the rate of one animal per 6 acres. Brush con-
trol treatments, each applied to two 20-acre plots dur-
ing late June and early July 1963, included shredding,
roller chopping, ”scalping” with a ”KG" blade, root
plowing or root plowing, followed by raking. The
original experiment, designed as a randomized com-
plete block, included comparison of fertility treat-
ments, but those subplots are not considered in this
analysis. Strips 225 feet wide were burned across the
original treatments in 1965 and 1966 (Box and White,
1969); those treatments also were not included in this
evaluation. In 1971, the huisache on approximately
one-third of each area was selectively treated by oil-
ing. In 1974, the entire management unit containing
the study plots was burned in an attempt to suppress
woody plant cover and remove excess mulch. Species
composition and woody species canopy cover were
recorded prior to mechanical treatments and in the
summers of 1964 and 1965. Also, herbage standing
crop by species was determined on each plot to
within 10 percent of the mean by a weight-estimated
method (McIntyre, 1952) in late August 1964 and
1965.

Based on species composition by weight, a ”her-
bage score" was determined for each plot. Herbage
score was based on ”relative desirability” for grazing,
utilizing relative proportions of forage species, and is
comparable to standard determinations of range con-
dition. Particulars of the method will be discussed in
the following section.

Procedures for 1977 Evaluations

Inasmuch as possible, data compatible with those
collected at initiation of the study were collected dur-
ing mid-August 1977. The plots were stratified to as-
sure separation of areas on which the huisache was
oiled in 1971 from those on which the huisache was
not treated. Burning treatments installed in 1965 and
1966 were not considered. Ten line transects, 100 feet
long, were randomly placed on the areas where only
the original mechanical brush control practice had
been applied, and ten transects were placed on the
adjacent area where the huisache had been oiled in
each plot. Since the entire area was burned in 1974,
there was no way to avoid confounding of that treat-
ment with the mechanical practices. Canopy cover of
woody plants was recorded by species and converted
to relative values for comparison of botanical compo-
sition differences.

Development of range condition estimates are
usually based on weight or foliar coverage by species

6

(Dyksterhuis, 1949). However, prior to and during
this evaluation, the pasture containing the plots was
subjected to grazing at one animal unit per 12 acres.
Therefore, basal contacts using an inclined 10-point
frame were utilized to estimate species composition.
This method avoids the influence of short-tenn ef-
fects of weather and preferential top removal by graz-
ing. Its weakness is that the role of broadleaved
species is biased downward and the role of sodgras-
ses is usually biased upward. Also, basal cover is not
necessarily directly related to top growth production.
Fifty 10-point frame samples were recovered from the
portion of each plot where the huisache plants were
not oiled and 50 from the portion where the huisache
was oiled in 1971.

A herbage score was developed for each treatment
according to the original method as described by
Powell (1966). Species included in the calculations
and percentage allowable of each deviate somewhat
from local Soil Conservation Service guides presently
used (Table 1). However, the scoring system ap-
peared to have a high degree of utility, and valid
comparisons depended on uniformity in use of the
system. A work sheet was developed for each plot
which included percentage composition by species
and percentage of that composition allowable in cal-
culating the herbage score. The percentage allowable
values were summed, and a range condition value
was assigned to each plot. Carrying capacities were
estimated from local Soil Conservation Service guides
as follows:

Herbage Closest associated Carrying capacity
score range condition class (acres/A.U.)
24-34 Low Fair 17
35-42 Fair 15
43-50 High Fair

13
51-58 Low Good
59-66 Good 12
67-75 . High Good 10

The original herbage scoring system (Powell,
1966) included the influence of the brush cover on
carrying capacity since stocking rates were not cal-
culated. There are no available data applicable to the
study area to facilitate accurate adjustments in car-
rying capacity for brush cover. Therefore, brush cover
and forage cover were assumed to be directly related
— any land surface covered by brush was assumed
void of usable forage. This method resulted in conser-
vative estimates of carrying capacity since some for-
age species do exist within the brush mottes and no
browse value was assigned the woody plants. The
method is probably biased more toward single-
stemmed species such as huisache under which some
forage will grow as contrasted to mixed brush mottes
which allow little, if any, forage species to exist. By
this method, a plot with a her age score of 45 (high
fair condition and potential carrying capacity of one

€

€

Table 1. Percentage of various forage species allowable in calculation of herbage score in August 1977 following various mechanical brush
control treatments in 1963 and/ or oiling of huisache in 1971 on the Rob and Bessie Welder Wildlife Refuge near Sinton, Texas. (Adapted from

the scheme of Powell [1966]).

esirable grasses
(All allowable in
herbage score)

Less desirable grasses
(Allowed in percentages
shown in parentheses)

Perennial forbs
(Total 5 percent
allowed)

Undesirable (invader)
grasses (No amount
accepted)

Buffalograss(10)b
Common curlymesquite (10)b
Filly panicum“ <5)“
Green sprangletop (5)a
Knotroot bristlegrass (5)d
Longtom (10)a

Meadow dropseed (10)
Pink tridens (5)

Plains bristlegrass (5)
Texas cupgrass (10)
Texas wintergrass (10)
Vine mesquite (5)

White tridens (5)

Big cenchrusa

Bunch cutgrass
Hairyseed paspalum
Little bluestem
Lovegrass tridens
Mourning lovegrass
Sideoats grama
Silver bluestem
Sourgrass

Bundleflower
Texas snoutbean

Common Berrnudagrass
Fringed chloris

Texas grama
Threeawns

Tumble windmillgrass
Whorled dropseed

aSpecies not listed by Powell (1966) or Soil Conservation Service Technical Guide but listed by Gould and Box (1965) and encountered in 1977

evaluations.

bUp to 10 percent total allowable for buffalograss and common curlymesquite in combination.

CP. halli var. filipes according to Gould (1975).

dA total of 5 percent of any combination of filly panicum and knotroot bristlegrass was allowed.

animal unit per 13 acres) and a brush cover of 33
percent would be assigned a carrying capacity of

13 x  ]= 19.4 acres per animal unit. The

final calculated values were tested by sight estimating
standing crop in November 1977 and reviewing pub-
lished research on forage production trends for the
site (Box, 1960). Comparative carrying capacities were
calculated on the basis that 26 pounds per day is re-
quired to sustain an animal unit and that 25 percent of
the forage produced annually is usable by the ani-
mals. These values were in general agreement with
carrying capacities based on the herbage scores. On
the average, the estimates varied between the two
methods by less than 10 percent.

RESULTS AND DISCUSSION

Brush Cover and Composition

The maintenance burn applied in 1974 had little
apparent effect on the woody plant population. Occa-
sional charred logs were found during the evalua-
tions, but burned stumps were generally less than
0.25 inch in diameter, indicating that only the smaller
woody plants were severely damaged.

Brush canopy cover in 1963 on untreated plots
was 48.6 percent (Powell, 1966) with honey mesquite
(38.4 percent), ;~e_huisache (2.2 percent), blackbrush
acacia (1.2 percent), agarito (1.2 percent), and tasajillo
(1.5 percent) the primary species. The remainder of
the cover consisted of twisted acacia, spiny

ackberry, lotebush, bluewood, Texas persimmon,
erlandier wolfberry, pricklypear, creeping mesquite,
and lime pricklyash. Twisted acacia was the only
species encountered in 1963 that was not recorded in

1977 (Table 2). Twisted acacia essentially replaces
huisache in the western portions of south Texas, and
its presence in 1963 may have been the result of pre-
vious rainfall patterns.

That these woody plant stands are relatively static
as proposed by Drawe et al (1978), at least relative to
total canopy cover, may be evidenced by the similar-
ity of cover values, 48 percent in 1977 compared to
48.6 percent in 1963. However, there were differences
in botanical composition of the brush stand. Where
the huisache was oiled but no mechanical treatment
was applied, honey mesquite provided a consid-
erably lower percentage of the composition (Table 2).
On a relative basis, honey mesquite furnished 79 per-
cent of the total brush cover in 1963 (Powell, 1966)
and only about 16 percent in 1977. This cover differ-
ential was compensated by an increase in species
such as blackbrush acacia, 2.5 percent in 1963 and
27.4 percent 14 years later. Substantial increases also
occurred with species such as lime pricklyash and
bluewood. All other species except creeping mesquite
increased in canopy cover by 2 to 3 percent.

The decline of honey mesquite may be explained
partially by rainfall conditions preceding each evalua-
tion. Relatively dry conditions, with the exception of
1960, prevailed for the 5-year period preceding instal-
lation of the experiment (Table 3). For the 5-year
period including 1963, annual precipitation averaged
7.91 inches below the annual average for 1957
through 1976. These dry conditions persisted until
1966, after which annual rainfall was above the 20-
year average with the exceptions of 1969 and 1975.
During the 5-year period preceding reevaluation of
the experiment (1972 through 1977), annual rainfall
averaged 4.93 inches above the 20-year average of

7

Table 2. Relative botanical composition (%) of woody plant stands based on canopy cover in August 1977 after mechanical brush control
treatments in 1963 and/or oiling of huisache in 1971 on the Rob and Bessie Welder Wildlife Refuge near Sinton, Texas.“

Huisache oiled

Root plowedfl‘

No Roller Root
Common name treatment Shredded chopped Scalped plowed and raked
Agarito 6.9 6.5 5.9 0 7.2 0
Blackbrush acacia 27.4 16.6 12.2 5.7 19.1 11.4
Brazil (bluewood) 11.3 3.3 8.5 0.7 4.8 0
Carolina wolfberry 0.6 0.8 1.6 0 3.6 9.4
Creeping mesquite 1.5 0.2 0 0 3.6 1.9
Huisache 8.6 4.1 3.7 9.4 19.1 30.3
Honey mesquite 15.9 46.6 47.3 78.2 14.3 30.2
Lime pricklyash 13.6 5.7 48.8 3.8 19.1 7.6
Lotebush 4.0 2.8 1.6 18.8 O 1.9
Spiny hackberry 4.6 9.3 9.1 0.6 6.0 5.7
Tasajillo 0 0.4 O O O 0
Texas persimmon 5.6 3.7 5.3 0 3.6 1.9
Absolute canopy cover (%) 48.8 24.7 18.8 16.0 8.4 5.3
Huisache not oiled
No Roller Root Root plowed

Common name treatment Shredded chopped Scalped plowed and raked
Agarito 6.5 6.7 8.1 0.7 0
Blackbrush acacia 26.6 13.9 14.4 2.2 0
Brazil (bluewood) 3.0 2.6 4.4 0 0
Carolina wolfberry 0 0 0 O 3.6
Creeping mesquite 0 0 O 1.5 0
Huisache 33.6 22.4 31.7 80.3 83.3
Honey mesquite 14.8 33.4 21.5 5.1 12.0
Lime pricklyash 10.4 12.9 15.3 5.4 0
Lotebush 0.9 3.1 0.2 3.3 0
Spiny hackberry 2.6 5.0 4.4 0.4 0 ‘-
Texas persimmon 1.6 0 0 1.1 1.1
Absolute canopy cover (%) 43.2 41.9 43.2 27.7 27.6

“The area containing the research plots was maintenance burned in 1974.

36.62 inches. Thus, lowlands which were dry preced-
ing and during the original experiment held water for
extended periods each year, especially during the
spring growing season. Honey mesquite does not tol-
erate inundation exce t for short eriods (Scifres,
1973), and on lowland) areas (signi icant in area on
cumulative basis) along the coast, honey mesquite
was killed by inundation following hurricane Beulah
(Scifres and Mutz, 1975). Kill of honey mesquite by
flooding was evidenced by remnants of trees on the
lowlands on the Welder Refuge during the 1977
evaluations (Figure 2). Honey mesquite on the study
plots varied from 2.1 to 3.1 feet tall, and there was no
difference in height of honey mesquite on untreated
plots and that of regrowth on treated plots (Table 4).
The lowland areas were excluded from the 1977
evaluation but were undoubtedly included in the
original evaluation. Honey mesquite was present in
proportions originally reported only on areas receiv-
ing the scalping treatment (Table 2).

Huisache is apparently adapted to moist sites and
favored by wet years. The year after installation of the
study, huisache contributed 4.6 to 8.7 percent to the
woody plant cover on all plots except those roller
chopped (12 percent) or root plowed and raked (23.2

8

Table 3. Annual rainfall on the Rob and Bessie Welder Wildlife
Refuge near Sinton, Texas from 1957 to 1976.

Precipitation (inches)

Deviation from

Year Total 20-yr average
1957 ‘ 41.51 + 4.89
1958 38.90 + 2.28
1959 35.02 - 1.60
1960 46.96 +10.34
1961 26.30 -10.32
1962 18.51 -18.11
1963 16.75 -19.87
1964 36.23 + 0.39
1965 27.87 - 8.75
1966 32.02 - 4.60
1967 43.05 + 6.43
1968 48.17 +11.55
1969 33.81 - 2.81
1970 40.43 + 3.81
1971 39.14 + 2.52
1972 39.30 + 2.68
1973 49.06 +12.64
1974 39.54 + 2.92
1975 30.59 - 6.03 V
1976 49.06 +12.44
Average 36.62

 

percent) (Powell, 1966). The dramatic increase in
huisache cover on the Welder Wildlife Refuge
prompted management personnel to begin an oiling
operation in 1971. Oiling held the relative proportions
to 3.7 to 9.4 percent except on the root-plowed plots
(Table 2). Root plowing of mixed brush on the more
fertile soils of the Coastal Prairie may result in almost
solid stands of huisache within 5 to 7 years, especially
under higher than normal rainfall. Huisache cover
presently accounts for more than 80 percent of the
total brush canopy cover on the root-plowed plots
which were not subjected to the oiling treatment (Ta-
ble 2). This allowed total woody canopy cover to in-

crease to over 27 percent since treatment in contrast

to brush covers of 5.3 to 8.4 percent on root-plowed
plots where the huisache was oiled. In comparison,
huisache canopies account for 22.5 to 33.6 percent of
the total brush cover following other treatments
where the species was not oiled (Figure 3).
Compared to original stands, the change in rela-_
tive proportion of the brush cover furnished by
huisache the past 14 years has been relatively minor
except on root-plowed plots where the oiling treat-
ment was applied (Table 5). Where oiling was not
practiced, the relative contribution by huisache has
increased by 11 to 25 percent following mechanical

Figure 2. Depression on
which honey mesquite was
killed by extended inunda-
tion on the Welder Wildlife
Refuge.

Eigure 3. Hiusache formed
almost solid stands follow-
ing root plowing of Coastal
Prairie in 1963 on the Wel-
der Wildlife Refuge (upper
photo). Oiling of areas in
1971 which were root
plowed in 1963 effectively
reduced the presence of
huisache (lower photo).

9

Table 4. Height (ft) of honey mesquite encountered in August 1977
on the various areas mechanically treated in the summer of 1963 on
the Rob and Bessie Welder Wildlife Refuge near Sinton, Texas?

Huisache treatment

Treatment Oiled Not oiled
None 2.5 —
Shredded 2.3 3.1
Roller chopped 2.1 3.0
Scalped 2.1 2.4
Root plowed 2.1 2.3
Root plowed and raked 2.4 2.5

aThe area containing the research plots was maintenance burned in
1974.

Table 5. Change in relative botanical composition (%) represented
by huisache by August 1977 after mechanical treatments in 1963
and/or oiling in 1971 on the Rob and Bessie Welder Wildlife Refuge

near Sinton, Texas?

Huisache treatmentb

Treatment Oiled Not oiled
None + 4 -
Shredded — 5 +25
Roller chopped -— 8 +11
Scalped + 1 +24
Root plowed +15 +76
Root plowed and raked + 7 +60

“The area containing the research plots was maintenance burned in
1974.

bValues rounded to nearest whole number.

treatments other than root plowing. Following root
plowing, huisache now provides 6O to 76 percent
more of the relative cover than immediately after
treatment. These data and recent observations indi-
cate that huisache may be considered one of the pri-
mary, if not the most serious, brush problems on the
Coastal Prairie. Brush management methods which
disturb the soil surface apparently release huisache to
increase in the absence of competition from other
woody plants.

Lime pricklyash was relatively unaffected by
mechanical treatment. It provided 1.1 to 2.5 percent
of the brush cover on untreated plots and those
shredded or roller chopped in 1963 (Powell, 1966). In
1977, relative foliar cover averaged 9.1 percent, and
ranged from 3.8 to 19.1 percent among the treat-
ments, except on plots which were root plowed and
raked (Table 2). Even after adjustment to absolute
ground cover, a notable increase in lime pricklyash
has occurred since installation of the experiment. The
relative cover of lime pricklyash was reduced by
chaining, raking, and stacking on sites similar to
those in thepresent study (Scifres et al, 1976), but the
species is a root sprouter and must be completely
uprooted for effective control. Texas persimmon,
bluewood, and lotebush responded similarly to lime
pricklyash. These species are also difficult to control
with other brush control procedures such as her-
bicides (Scifres et al, 1977) and may present signifi-

10

cant livestock management problems where their
abundance exceeds that necessary for quality wildlife
habitat.

Herbaceous Vegetation and Range Condition

Some differences in herbaceous vegetation were
expected between species composition in 1977 and
that in 1963 and 1964 since rainfall conditions differed
considerably. Chamrad and Box (1965) evaluated the
influence of drouth in 1963 on mortality of grasses on
the Welder Refuge. Mortality of grasses was lowest
on Victoria clay sites compared to that on Nueces fine
sand or Miguel fine sandy loam. Mortality also varied
among species within sites from 34.7 percent for
silver bluestem on Victoria clay to 76.8 percent for
seacoast bluestem on Miguel fine sandy loam. On
Victoria clay, the highest mortality, 51.9 percent, oc-
curred with filly panicum. Mortality of seacoast blue-
stem was 48.7 percent and of buffalograss 43.4 per-
cent.

The effects of weather patterns were not reflected,
to the extent anticipated, in differences in herbaceous
vegetation between the initial and 1977 evaluations.
The primary grass species encountered by Box and
Powell in 1963 were encountered in the 1977 evalua-
tions. The greatest difference between evaluations
was the frequent occurrence in 1977 of common Ber-
mudagrass and longtom. These species, not em-
phasized by Powell (1966), apparently increased in
importance during the series of wet years prior to the
reevaluation. Longtom and common Bermudagrass
were reported by Scifres and Mutz (1975) to be initial
stabilizers of previously inundated lowlands in south
Texas. These species grow on the immediate
shorelines of inundated lowlands, and longtom sto-
lons occasionally spread over the surface of standing
water. Although they are of fair to good forage value,
both species are considered invaders of the Blackland
range site. However, based on its distribution and
recently reported ecological role (Gordon and Scifres,
1977; Scifres and Mutz, 1975), longtom was utilized in
calculating the .1977 herbage score (Table 1). Recently
published information on its grazing value was fur-
ther justification for not excluding longtom from the
herbage score. Durham and Kothmann (1977) re-
ported that although little bluestem was usually the
dominant grass in cattle diets on the Coastal Prairie,
longtom afforded from 14 to 21 percent of botanical
composition of diets from February to early March.
Utilization during this period ranged from 25 to 37
percent. As additional evidence for the recent influ-
ence of increased rainfall on vegetation change since
initiation of the study, every plot supported signifi-
cant cover of various species of rushes and sedges
(Table 6), which are abundant only under relatively
wet conditions.

Herbage scores changed little from 1963 to 197 7 on
shredded, roller chopped, or untreated plots where
the huisache was oiled (Table 6). When adjusted for
presence of longtom (not encountered in 1963), her-
bage scores in 1977 are essentially the same as re-

Q

ti‘

9‘

Table 6. Herbage score and percentage composition of herbaceous vegetation in August 1977 after installation of mechanical brush control
treatments in 1963 and/or oiling of huisache in 1971 on the Rob and Bessie Welder Wildlife Refuge near Sinton, Texas?

Huisache oiled

Hegetation No

Roller Root Root plowed
class treatment Shredded chopped Scalped plowed and raked
Desirable grassesb 75.0 84.0 80.2 74.9 69.5 66.3
Invader grassesc 0.2 1.4 1.4 2.1 3.1 10.7
Forbs 18.8 13.9 12.7 23.0 26.9 21.4
Rushes/sedges 6.0 0.6 5.7 0 0.5 1.6
Herbage scored 59.3 60.3 50.9 53.2 53.8 59.2
Change since 1964e — 1.7 — 5.9 — 2.6 +12.3 + 2.4 +35.5
Huisache not oiled
Desirable grassesb 75.4 71.0 62.3 61.1 48.0
Invader grassesc 0.3 0.9 18.4 7 16.2 19.0
Forbs 20.1 18.0 15.9 17.0 22.4
Rushes/sedges 4.2 10.1 3.4 5.7 10.6
Herbage scored 47.2 45.2 45.1 47.1 47.7
Change since 19648 —18.9 — 8.3 — 4.2 — 4.3 +24.0

aThe area containing the plots was maintenance burned in 1974.
bIncluded green sprangletop and big cenchrus.
clncluded common Bermudagrass and longtom.

dDesirable forbs, bundleflower, and Texas snoutbean included in herbage score.

eI-Ierbage score reported by Powell (1966).

ported for 1963, following no treatment except
huisache oiling or the mechanical top removal treat-
ments followed by huisache oiling. This is under-
standable after inspection of absolute brush covers
following mechanical treatment where the huisache
was not oiled (Table 2). The brush cover has been
replaced, and range condition is essentially the same
as when the study was initiated.

Absolute herbage scores in 1977 following root
plowing or root plowing and raking differed little
from those where other mechanical treatments were
applied (Table 6). However, herbage scores in 1977 on
root plowed and raked areas were 24 to 35.5 points
higher than reported 1 year after treatment. The low
initial herbage scores were attributed to extensive soil
disturbance by root plowing which invariably is con-
ducive to establishment of invaders. It should be em-
phasized that the herbage scores are based on relative
proportions of species present and do not reflect the
influence of brush canopy cover which varies consid-
erably among the treatments (Table 2). The relatively
low brush covers on the root-plowed plots are indica-
tive of even greater potential for range improvement
than reflected by 1977 evaluations.

On the portions of the plots where the huisache
was not oiled, herbage scores were lower in 1977 than
reported in 1963 with the exception of the root-
plowing-plus-raking treatment (Table 6). On most
plots, the herbage score difference was substantial

ven after including the contribution of longtom

@1666 was not present in 1963. Thus, huisache infes-

‘tations are seriously retarding achievement of poten-
tial improvement in range condition.

The state of range vegetation in 1977 was also as-
sessed relative to grazing value of the grasses present
(Table 7). Regardless of treatment, a relatively low
percentage of the grass composition was afforded by
grasses of poor grazing value. Such species included
Roemer threeawn, tumble windmillgrass, Texas
grama, and whorled dropseed. Grass species
encountered in 1963 but not encountered in 1977 in-
cluded common curlymesquite and Nealley sprangle-
top. The only other notable differences in grasses
present in 1977 compared to 1963 were that scattered
little bluestem clones were encountered in 1977, and
substantially higher amounts of knotroot bristlegrass
were present. Presence of little bluestem indicates an
upward trend in range condition, but knotroot
bristlegrass is usually an indicator of yearly weather
fluctuations or minor disturbances of the vegetation.

The relative contribution of grasses of good to ex-
cellent grazing value varied little among plots except
for those shredded on which the huisache was oiled
(Table 7). The shredded plots supported a relatively
high percentage of hairyseed paspalum and relatively
low amounts of longtom and common Bermudagrass.
This occurrence shifted the proportion of forages in
the fair grazing class downward and the proportion
provided by the good to excellent class upward.

Forbs accounted for 16 to 22 percent of the botani-
cal composition on plots where the huisache was not
treated and 13 to 27 percent of the botanical composi-
tion on plots where the huisache was oiled in 1971.
The variation in forb populations appeared to be
more closely associated with localized site conditions
and season of sampling than with specific treatment.

11

Table 7. Relative botanical composition (%) of grasses by grazing value after installation of mechanical brush control treatments in 1963
and/or oiling of huisache in 1971 on the Rob and Bessie Welder Wildlife Refuge near Sinton, Texas.a

Huisache oiled

Grazing No Roller Root
value treatment Shredded chopped Scalped plowed and raked
Good-Excellent 43.8 60.6 28.7 35.8 35.0 42.1
Fair 51.6 37.8 66.4 63.5 63.3 52.1
Poor 4.6 1.6 4.9 0.7 1.7 5.8
Huisache not oiled

Good-Excellent 37.6 30.6 24.5 38.0 17.2
Fair 57.6 66.2 71.6 53.8 68.9
Poor ‘ 4.8 3.2 3.9 8.2 13.9

aThe area supporting the plots was maintenance burned in 1974.

Nineteen species of broadleaf herbaceous plants were
encountered of which western ragweed, saltrnarsh
aster, erect dayflower, seacoast sumpweed, sawtooth
fogfruit, common ruellia, prairie gerardia, and up-
right prairie-coneflower were the most common. Na-
tive legumes, Texas snoutbean and bundleflower,
were present in low amounts (less than 0.1 percent of
the composition).

Condition of the range (based on composition of
herbaceous vegetation) which had not been mechani-
cally treated but on which the huisache had been
oiled was rated as high fair to low good. However,
the heavy brush cover caused the estimated carrying
capacity to be relatively low, one animal unit per 23.4
acres (Table 8). This conservative stocking rate was
developed to allow range improvement, inasmuch as
possible, without brush management. In contrast, the
root-plowed plots where the huisache had been oiled
were also in high fair condition but, because of the
relatively low brush cover, the average estimated car-
rying capacity was one animal unit per 14.1 acres. The
huisache oiling operation was of critical importance in
determining livestock carrying capacity. The carrying
capacity was significantly lower where the huisache
was not oiled 8 years previously except on root-
plowed plots. In that single instance, although not
statistically significant, only 3.8 acres would be re-

Table 8. Calculated carrying capacities (acres/animal unit, year-
long) in August 1977 after installation of mechanical brush control
treatments in 1963 and/ or oiling of huisache in 1971 on the Rob and
Bessie Welder Wildlife Refuge near Sinton, Texas.“

Estimated carrying capacities (acres/A.U.)b

Treatment Huisache oiled Huisache not oiled
None 23.4 d

Shredded u 15.9 ab 22.9 cd
Roller chopped 16.0 ab 22.4 cd
Scalped ‘ 15.5 ab 22.9 ccl

Root plowed 14.1 ab 17.9 bc

Root plowed and raked 12.6 a 17.9 bc

aThe area containing the plots was maintenance burned in 1974.

bMeans followed by the same letter are not significantly different at
the 95% level according to Duncan's New Multiple Range Test.

12

quired per animal unit where the huisache was not
oiled.

MANAGEMENT IMPLICATIONS

Apparently, canopy cover of mixed-brush stands
on Chaparral-bristlegrass communities is relatively
stable under a given set of management conditions.
Shifts in botanical composition occurred over the 14-
year period in response to rainfall patterns. Honey
mesquite did not tolerate extended inundation, and
during wet years was replaced, relative to its contribu-
tion to botanical composition based on foliar cover,
primarily by huisache. Other species such as lime
pricklyash and bluewood also increased, but not so
dramatically as huisache.

Simple top removal by roller chopping or shred-
ding resulted in only short-term suppression of the
brush canopy and provided no lasting improvement
in range condition (Figure 4). These methods must be
applied systematically, either alone or in conjunction
with other techniques such as burning, and repeated
periodically to facilitate progressive improvement in
range condition. Apparently, the maintenance burn

., applied in 1974 was too late after treatment to sub-
 stantially improve vegetation of roller-chopped or

shredded areas. Burning the year following mechani-
cal treatment improved the level of brush“ suppres-
sion (Box and White, 1969), but burning apparently
must be practiced regularly to maintain range im-

.1 provement. These data and results from other studies
“indicate that simple top removal practices, if not fol-

lowed with prescribed burning, should be repeated at
3- to 5-year intervals, depending on rainfall. Al-
though simple top removal results in only short-term
effects on the vegetation, some of the transient effects
are of value. Shredding or roller chopping can be
applied to create desirable patterns of brush suppres-
sion. This improves visibility on portions of the area
which facilitates livestock handling and care while al-
lowing maintenance of some mature brush cover.
New sprouts stimulated by top removal improve
browse value (Powell and Box, 1966), and reduced
woody plant stature improves accessibility to th

sprouts. Reduction of brush topgrowth releases he

baceous vegetation which provides fine fuel for instal-
lation of prescribed burns. The additional fuel and

Root plowefi

localizing of the brush canopy near ground line al-
lows maximum exposure of the woody plant top-
growth to the hottest part of grass fires.

Scalping, essentially a compromise between sim-
ple top removal methods and root plowing, does not
appear to be advantageous over either of the other
approaches (Figure 5). A possible contradiction
would be the case of initially high brush covers where
felled topgrowth would be left as litter unless raked.
The ”KG" blade shears the woody plants off at
ground line and allows localizing of the debris with-
out use of a brush rake. However, soil disturbance

Figure 4. Brush canopy
cover of stands which were
roller chopped (upper
photo) or shredded (lower
photo) in 1963 and on
which the huisache was
not oiled in 1971 were not
different in 1977 from un-
treated areas on the
Coastal Prairie.

following scalping allows invasion of opportunists
such as willow baccharis (Drawe, 1977) which retard
the rate of range improvement.

Relative only to brush control effectiveness based
on reductions in brush canopy cover, root plowing is
the most effective mechanical method known. Since
it removes at least a portion of the brush root system,
treatment life expectancy is much longer than from
other treatments.- For example, the brush canopy
cover 14 years after root plowing the study site is ap-
proximately 6O percent of the original cover. Where
the huisache was removed by oiling, the present

Figure 5. Mixed-brush
stand in November 1977
which was scalped in 1963
on the Welder Wildlife
Refuge near Sinton, Texas.

13

canopy cover is only 15 percent of the original. Thus,
a conservative treatment life estimate would be 20
years. Followup improvement efforts such as pre-
scribed fire and/or individual-plant treatment with
herbicides might suppress the brush canopy indefi-
nitely. However, there are some distinct disadvan-
tages of root plowing:
1. The surface soil is left extremely rough for
several years following treatment. Although this
facilitates water infiltration and retention, it is
sometimes a problem in livestock management.
Raking localizes debris following root plowing of
dense brush and facilitates management.

2. The initial cost of treatment is high. Although
economic returns may be favorable based on
treatment life expectancy, cash flow within the
operation at the desired time of treatment may
limit the investment.

3. Soil disturbance is excessive, and secondary
succession is reinitiated in its early stages. Thus,
for several years after root plowing, low value for-
age species dominate the herbaceous stands. This
often results in reduced stocking rates until desir-
able, perennial range forage species become rees-
tablished.

4. The increase in huisache, which may form al-
most pure stands within a few years after root
plowing, is a severe problem. Thus, huisache
management is an important consideration when
root plowing is accepted as the brush manage-
ment alternative. Individual-plant treatments can
be most economically and easily applied in the
early stages of huisache infestation.

5. Root plowing essentially eliminates the brush
cover and seriously reduces browse availability for
an extended period. Its use requires careful plan-
ning where wildlife habitat is important to the
range management strategy.

Each of these methods has particular adaptability
to specific situations, and all are most effective when
considered relative to land resource management ob-
jectives and when their use is carefully planned with
provisions for followup treatment and proper grazing
management for most effective response.

CONCLUSIONS

Absolute brush covers in mature ”Chaparral-
bristlegrass” communities as described on the Welder
Wildlife Refuge on the Coastal Prairie are apparently
relatively stable. Species composition of the stands
may shift with prolonged changes in rainfall patterns.
Based on a 20-year average of about 36 inches per
year, honey mesquite dominated the brush stands
during a 10-year period when the annual average was
about 31 inches. During the next 10-year period, the
annual average was over 41 inches, resulting in ex-
tended inundation of depressions. Honey mesquite,
as evidenced by dead remains, did not tolerate the
wet cycle and was largely replaced by huisache. Dur-
ing wet periods, bunchgrasses may be reduced in rel-

14

ative importance and water-tolerant sod formers such
as longtom increase in importance.

Effects of roller chopping and shredding are show
g v

lived, and retreatments are necessary for lastin
range improvement. These practices require repeti-
tion, probably at least on 3- to 5-year intervals, for
maximum range improvement. Prescribed burning
can be used for maintaining initial brush manage-
ment effectiveness from shredding and roller chop-
ping.

Scalping with a “KG" blade is intermediate in
brush control effectiveness, soil disturbance, and re-
sponse of desirable vegetation between simple top
removal methods and root plowing. Utility of the
method, compared to root plowing, probably would
hinge on relative costs.

Root plowing provides long-term reductions in
brush cover. However, range condition initially de-
clines because of the influx of low-value range forage
species. With effective followup improvement efforts
and sound grazing management, the method offers a
moderate to long-term approach to brush manage-
ment.

Huisache is a particular problem when forrnulat-
ing brush management strategies for Coastal Prairie
rangeland. Apparently, control of other woody
species with root plowing releases huisache to domi-
nate the woody plant stand. If root plowing is
selected as a brush control alternative, the brush
management program should include provisions for
huisache treatment after 3 to 5 years.

ACKNOWLEDGMENTS

Ieff Powell collected the original data. These con-
trasts in time would not have been possible without
preservation of the study plots by the Rob and Bessie
Welder Wildlife Foundation. Permission from Caleb
Glazener, Director of the Foundation, to pursue this
work is gratefully acknowledged. The authors are
grateful to Roger Smith and Clayton Phelps for their
assistance in data collection and to Iulia Scifres for
typing and helping prepare the manuscript.

LITERATURE CITED

Beasom, S. L. and C. I. Scifres. 1977. Population reactions of
selected game species to aerial herbicide applications in south
Texas. I. Range Manage. 30:138-142.

Box, T. W. 1960. Herbage production in four range plant com-
munities in south Texas. I. Range Manage. 13:72-76.

Box, T. W. 1961. Relationships between plants and soils of four
plant communities in south Texas. Ecology 42:794-810.

Box, T. W. and A. D. Chamrad. 1966. Plant communities of the
Welder Wildlife Refuge. Contrib. No. 5, Series B. Welder
Wildlife Foundation, Sinton, TX. 28 p.

Box, T. W. and I. Powell. 1965. Brush management techniques for
improved forage values in south Texas. Trans. N. Amer. Wildl.
and Nat. Res. Conf. 30:285-296.

Box, T. W. and R. S. White. 1969. Fall and winter burning of south
Texas brush ranges. I. Range Manage. 22:373-376.

Chamrad, A. D. and T. W. Box. 1965. Drought-associated
mortality of range grasses in south Texas. Ecology 46:780-785.

Dodd, I. D. and S. T. Holtz. 1972. Integration of burning with
mechanical manipulation of south Texas grassland. I. Range
Manage. 25:130-136.

Drawe, D. L. 1968. Mid-summer diet of deer on the Welder
Wildlife Refuge. I. Range Manage. 21:164-166.

Drawe, D. L. 1977. A study of five methods of mechanical brush
control in south Texas. Rangeman's I. 4:37-39.

Drawe, D. L., A. D. Charnrad and T. W. Box. 1978. Plant com-
munities of the Welder Wildlife Refuge. Contrib. No. 5, Series B
(2nd Ed.). Welder Wildlife Foundation. Sinton, TX. 40 p.

Dyksterhuis, E. I. 1949. Condition and management of range land
based on quantitative ecology. I. Range Manage. 2:104-115.

Durham, A. I., Ir. and M. M. Kothmann. 1977. Forage availability
and cattle diets on the Texas Coastal Prairie. I. Range Manage.
30:103-106.

Gordon, R. A. and C. I. Scifres. 1977. Burning for improvement of
Macartney rose infested Coastal Prairie. Texas Agr. Exp. Sta.
Bull. 1183. 15 p.

Gould, F. W. 1975. Texas Plants. A checklist and ecological sum-
mary. Texas Agr. Exp. Sta. MP-585 (Rev.). 129 p.

Gould, F. W. and T. W. Box. 1965. Grasses of the Texas Coastal
Bend. Texas A&M Univ. Press, College Station. 189 p.

McIntyre, G. A. 1952. A method for unbiased selective sampling
using ranked sets. Austral. I. Agr. Res. 3:385-390.

Oefinger, R. E. and C. I. Scifres. 1977. Gulf cordgrass production,
utilization and nutritional value following burning. Texas Agr.
Exp. Sta. Bull. 1176, 19 p.

Powell, I. 1966. Forage production following selected mechanical
brush control practices in south Texas. MS Thesis, Texas Tech
Univ., Lubbock. 73 p.

Powell, I. and T. W. Box. 1966. Brush management influences for-
age value of south Texas woody plants. I. Range Manage.
19:212-214.

Powell, I. and T. W. Box. 1967. Mechanical control and fertilization
as brush management practices affect forage production in
south Texas. I. Range Manage. 20:227-236.

Powell, I., T. W. Box and C. V. Baker. 1972. Growth rate of sprouts
after top removal of huisache (Acacia furnesiana [L.] Willd.)
(Leguminosae) in south Texas. Southw. Nat. 17:

Scifres, C. I. 1977. Herbicides and the range ecosystem: Residues,
research and the role of rangemen. I. Range Manage. 30:86-91.

Scifres, C. I. 1975. Systems for improving Macartney rose infested .

Coastal Prairie rangeland. Texas Agr. Exp. Sta. MP-1225. 12 p.

Scifres, C. I. (Ed). 1973. Mesquite. Growth and development,
management, economics, control, uses. Texas Agr. Exp. Sta.
Res. Mon. 1:84.

Scifres, C. I., G. P. Durham and I. L. Mutz. 1977. Range forage
production and consumption following aerial spraying of mixed
brush. Weed Sci. 25:48-54.

Scifres, C. I., I. L. Mutz and G. P. Durham. 1976. Range im-
provement following chaining of south Texas mixed brush. I.
Range Manage. 29-418-425.

Scifres, C. I. and M. G. Merkle. 1975. Herbicides and good man-
agement improves range pastures. Weeds Today. 6(1):5-7.
Scifres, C. I. and I. L. Mutz. 1975. Secondary succession following
extended inundation of Texas coastal rangeland. I. Range Man-

age. 28:279-282.

Scifres, C. I., I. L. Mutz and W. T. Hamilton. 1978. Control of
mixed brush with tebuthiuron. I. Range Manage. 31:(In press).

White, R. S. 1969. Fire temperatures and the effect of burning on
south Texas brush communities. MS Thesis. Texas Tech. Univ.,
Lubbock.

Whitson, R. E., S. L. Beasom and C. I. Scifres. 1977. Economic
evaluation of cattle and white-tailed deer response to aerial
spraying of mixed brush. I. Range Manage. 30:214-217.

15

Common Name

Cacti

Pricklypear
Tasajillo

Forbs

Bundleflower
Common ruellia
Erect dayflower
Prairie gerardia
Saltmarsh aster
Sawtooth fogfruit
Seacoast sumpweed
Spotted beebalm
Texas snoutbean

Upright prairie-coneflower

Western ragweed
Horsetail conyza

Grasses

Big bluestem

Big cenchrus
Bristlegrass
Buffalograss

Bunch cutgrass

Bushy bluestem
Common Bermudagrass
Common curlymesquite
Eastern gamagrass

Filly panicum

Fringed chloris

Green sprangletop
Gulf muhly

I-Iairyseed paspalum
Knotroot bristlegrass
Little bluestem
Longtom

Lovegrass tridens
Meadow dropseed
Mourning lovegrass

16

Appendix

SCIENTIFIC NAMES OF PLANTS
MENTIONED IN TEXT

Scientific Name

Opuntia sp.
Opuntia leptocaulis

Desmanthus virgatus
Ruellia runyoni
Commelina erecta
Gerardia heterophylla
Aster subulatus
Phyla inczsa

Iva annua

Monarda punctata
Rhynchosia texana
Ratibida columnaris
Ambrosia psilostachya
Conyza canadensis

Andropogon gerardii

Cenchrus myosuroides

Setaria sp.

Buchloe dactyloides

Leersia monandra

Andropogon glomeratus

Cynodon dactylon

Hilaria berlangeri

Tripsacum dactyloides

Panicum hallii var. filipes

Chloris ciliata

Leptochloa dubia

Muhlenbergia capillaris var. filipes
Paspalum pubiflorum

Setaria geniculata

Schizachyrium scoparium
Paspalum lividum

Tridens eragrostoides

Sporobolus asper var. drummondii
Eragrostis lugens

Common Name

Scientific Name

Grasses (continued)

Nealley sprangletop
Pink tridens

Plains bristlegrass
Roemer threeawn
Seacoast bluestem

Sideoats grama
Silver bluestem

Sourgrass

Texas cupgrass
Texas grama
Texas wintergrass
Threeawns
Tumblegrass

Tumble windmillgrass

Vine mesquite
White tridens
Whorled dropseed
Yellow Indiangrass

Leptochloa nealleyi

Tridens congestus

Setaria macrostachya

Aristida roemeriana

Schizachyrium scoparium var.
littoralis

Bouteloua curtipendula

Bothriochloa saccharoides var.
longipaniculata

Digitaria insularis

Eriochloa sericea

Bouteloua rigidiseta

Stipa leucotricha

Aristida sp.

Schedonnardus paniculatus

Chloris verticillata

Panicum obtusum

Tridens albescens

Sporobolus pyramidatus

Sorghastrum nutans

Woody Plants

Agarito

Berlandier wolfberry
Blackbrush acacia
Bluewood

Carolina wolfberry
Creeping mesquite
Honey mesquite

Huisache

Lime pricklyash
Lotebush

Spiny hackberry
Texas persimmon '
Twisted acacia
Willow baccharis
Wolfberry

Berberis trifoliola ta 

Lycium berlandieri
Acacia rigidula

Condalia obovata
Lycium carolinianum
Prosopis reptans
Prosopis glandulosa var.

A glandulosa

Acacia farnesiana
Zanthoxylum fagara
Zizyphus obtusifolia
Celtis pallida
Diospyros texana
Acacia tortuosa
Baccharis salicina
Lycium sp.

Mention of a trademark name or a proprietary product does not
constitute a guarantee or warranty of the product by the Texaafi
Agricultural Experiment Station or by the Welder Wildlife Founda- a ‘ '
tion and does not imply its approval to the exclusion of other
products that also may be suitable.

[Blank Page in Original Bulletin]

All programs and information of The Texas Agricultural Experiment Station are available to everyone
without regard to race, color, religion, sex, or national origin.

The Texas Agricultural Experiment Station, Neville P. Clarke, Director, College
Station, Texas

3M—10-78