Mechanical Treatment Impacts to Cultural
Resources in Central Arizona: The Marden
Brush Crusher1
J. Scott Wood2
Cultural resources are usually defined as the
locations and contents of prehistoric and historic archeological sites, historic buildings and
settlements, and areas which were the scene of
important historic incidents, such as battlefields or exploration routes. This definition is
legally adequate for management purposes, but
does not take into account the value of these
properties. First of all, they represent the
history and cultural heritage of the nation-an
important value in and of itself. However, the
primary importance of cultural resources is that
they are the physical remains of former times.
They represent the interaction of humans with
natural and cultural environments. By necessity,
then, they contain representations of the relationships within and between cultural/technological systems and environmental systems. In other
words, cultural resources can provide us with
information on the development of human social,
economic, political, and technological adaptations to changes in demography and natural
environment. They can also provide us with
information on environmental conditions at
various times in the past and on the extent to
which environmental changes and present conditions may have been affected by human activity. They do this by virtue of their being
composed of cultural and natural materials which
can be shown to be associated in both space and
time. Thus, besides documenting the development
of cultural systems, they also document the
development of environmental systems, both of
which can be useful as baseline or comparative
data for the management and use of those same
environments today. That is, they could if land
managers and other resource specialists would
begin to take cultural resources management and
archeological research seriously.
1Presented at the Symposium on Dynamics and
Management of Mediterranean-Type Ecosystems,
June 22-26, 1981, San Diego, California.
2Assistant Forest Archeologist, USDA Forest
Service, Tonto National Forest, Phoenix, AZ
85038.
Gen. Tech. Rep. PSW-58. Berkeley, CA: Pacific Southwest Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture; 1982.
Abstract: Forest Service management practices
have the potential for impacting cultural
resources in a variety of ways, ranging from
complete removal to alteration of the environmental context of prehistoric or historic proper
ties. Much of this impact is derived indirectly
from activities designed not so much to manage
the land surface (the location and contextual
matrix of cultural properties) but the vegetation
on that surface. This study examines one such
activity--brush crushing--and its effects on
archeological site integrity. Although much
remains to be done, it demonstrates that mechanical manipulation of vegetation has the potential
for serious disruption of cultural resources.
Present-day management of cultural resources
on the National Forests of the United States is
not directed toward the recovery and use of these
kinds of data, despite the long-demonstrated
similarity (at least in the Southwest) between
prehistoric land use patterns-some of which succeeded and many of which failed-and historic and
modern American land use patterns. Instead, the
job of cultural resources management in the U.S.
Forest Service is to integrate cultural resources
protection from damage and loss with the more
production-oriented management activities of the
Forests and their users. This job consists primarily of identifying the cultural resources of a
Forest and then working out ways to avoid
damaging or destroying them while attempting to
manage other resources.
Direct impacts to the land surface, and thus
to the cultural resources on and below that surface, are easy to see and understand. Any form
of construction or resource exploitation which
involves the removal, relocation, or compaction
of soils, sediments, or mineral materials, or
which require the modification of existing
topography, has the potential to damage, destroy,
or remove cultural properties and/or their artifactual content. Even the relocation of features
and/or artifacts can significantly alter an
archeological site, since it is from the spatial
arrangement of artifacts and other cultural and
non-cultural materials in surface and subsurface
topographic contexts that behavioral and environmental patterns are identified. These patterns
constitute the primary informational content of
any prehistoric or historic archeological site.
Since the goal of cultural resources management
on Federal lands is ostensibly to protect and
preserve this information, protection from
impacts should be directed at preserving the
integrity of behavioral contexts. Damage to
individual artifacts is one type of impact to the
data-producing value of a site, and, as such, it
merits study and prevention (c.f. Gallagher
1978). However, it appears to be less of a factor in data loss than dislocation, as the patterns of behavior involved in making an artifact
are much simpler and less informative about
human organization and adaptation than are the
515
behaviors which resulted in the eventual location
of that artifact and its associations with other
natural and cultural materials. It has been
demonstrated by several experiments (DeBloois,
Green, and Wylie 1974, Gallagher 1978) that tractor travel over an archeological site is
extremely destructive to surface artifact spatial
relationships and locations; markedly more so
than to artifact integrity. Thus, surface
disturbances such as road building, log skidding,
scarification, fence building, machinery operations, pipeline trenching, posthole digging,
stock tank construction, parking lots, recreation
areas, etc. are all activities or facilities
which could damage or destroy the contents and
context of any cultural property they encountered. Less obvious potential impacts can also
result from activities not specifically designed
to modify the land surface, but rather to modify
the vegetative cover of that surface. Examples
of such practices include chaining, juniper
pushing, crushing or "roller chopping" as it is
sometimes known, burning, and the use of herbicides in order to either change the composition
of the vegetation or to eliminate or reduce cover
to lessen fuel accumulation risks for fire management.
Several of these activities would seem to present fairly straightforward dangers to cultural
properties. The violent uprooting of trees by
chaining will necessarily damage any cultural
features that might have trees growing on or near
them. Others are less apparent or are indirect;
both burning and herbicide use, by removing
cover, may initiate erosion that could damage
cultural properties. In addition, burning will
destroy most standing historic structures, many
of which are built primarily out of wood, and it
may, under certain circumstances, affect the ability of some heat-sensitive cultural materials to
be dated. Because of the lack of information
about the effects of these vegetation manipulation practices, current Forest Service management
policies, at least in the Southwest, generally
assume that they will either damage cultural
resources or that their indirect effects are not
substantially more of a threat than natural processes. Thus, it is possible that some damages
are being overlooked while other activities which
may do no damage are being restricted. To provide efficient management of all resources, it
would be best to know just what kinds of impacts
can be expected from any particular kind of activity. It was in this context that the Prescott
National Forest in Arizona requested a study of
the impacts to be expected from their extensive
and continuing program of chaparral conversion by
mechanical treatment-specifically, the use of the
Marden Brush Crusher.
Little systematic observation of mechanical
treatment impacts to cultural resources has been
made to date, other than for several of the more
obvious ones-chaining (DeBloois, Green, and Wylie
1974) and scarification (Gallagher 1978). No
516
study had been made for treatments specific to
the chaparral vegetation type prior to this one.
The purpose of this study, therefore, was to
develop and describe a procedure by which any
mechanical impact to surface or subsurface
features and artifacts could be defined, and at
the same time to identify the impacts of brush
crushing on cultural properties. Specifically,
three types of cultural property were selected to
be observed for effect: surface artifact
scatter, surface (low relief) architectural
features, and subsurface artifactual materials.
This study was made in the simplest manner
possible, by means of "before and after" observational tests.
The area utilized for this study was a 60 acre
parcel of Prescott National Forest land, located
on the Walnut Creek Ranger District of that
Forest near its Camp Wood Administrative Site.
This area had been selected for conversion from a
chaparral to a grass type vegetation and was surveyed for cultural resources clearance in 1978
(Wood 1978). The archeological survey also
served to identify properties suitable for a test
of the impacts to be expected from the mechanical
treatment proposed for the parcel. The remainder
of this paper will describe the results of the
impact study carried out by myself and Harlow
Yaeger of the Prescott National Forest in 1979
(Wood and Yaeger 1978, Wood 1979).
Vegetation in the study area was primarily
turbinella oak and manzanita chaparral with a
discontinuous overstory of alligator juniper,
Emory oak, and an occasional pinyon, broken by
stringers of ponderosa pine in the basins and
drainages. The chaparral understory, locally
quite dense, also contained some mountain mahogany, Fendler ceanothus, datil yucca, silktassel,
beargrass, prickly pear and mammilaria cactus,
and occasional patches of ring muhly and grama
grasses. Substrate in this hilly area was primarily decomposed granite, ranging from rocky
residual clay sediments to more homogenous loamy
colluvium. Bedrock outcrops of granite were
common.
The conversion project which took place in the
study area involved the use of a Marden Brush
Crusher, a non-motorized device consisting of two
slightly offset tandem rollers fitted with hardened alloy "paddlewheel" blades. It is generally pulled as a trailer by a large
bulldozer-type tractor. The twelve blades on
each roller are of two different widths, providing an uneven gait, and the offset between
rollers provides for additional churning of the
surface and a variation in blade angles at contact. The combination of these features and a
gross weight of 44,000 pounds (20,000 kg) results
in the rapid removal of chaparral-type plants.
However, this device can operate only in relatively low-slope areas without a substantial tree
canopy, and where the ground surface is free of
exposed bedrock or talus. Rocky surfaces tend to
dull and break the blades. Thus, its normal
operation will often leave a pattern of irregular
patches of crushed and uncrushed stands of brush,
reflecting variation in topography and substrate.
IMPLEMENTATION AND DESIGN
Surface artifacts and architectural features
on cultural properties in central Arizona have a
long history of disturbance and modification by
various land use practices, including grazing,
farming (in some areas, for nearly 2000 years),
logging, mining, recreation, vehicle travel, and
pot hunting (the deliberate vandalism, destruction, and looting of archeological sites). Added
to this is an even longer history of natural
impacts such as weathering, erosion, soil movement, and alluviation all of which serve to alter
the physical characteristics and/or locations of
cultural materials. Sites located on hills or on
shallow soils are especially subject to this form
of alteration of behavioral contexts (McGuire
1977). Nevertheless, a Marden blade applies a
considerable amount of force, and so must be
expected to have at least the potential for
having at least as much of an effect on the
structural integrity of artifacts and features as
any of the "recognized" impacts.
In response to the problems and values discussed here, this study was designed to test what
appeared to be the two most critical potential
impacts expected from the use of the brushcrusher: surface artifact and architectural
displacement and subsurface artifact breakage and
displacement, taken as an indicator of all subsurface impacts.
Surface artifact damage was also recorded in
the form of breakage, used as a simple means to
represent all the various potential types of physical alteration which could be regarded as
damage to the informational content of a site.
At this stage in the development of impact
studies, it was felt best to emphasize the observation and assessment of dislocation impacts to
the presence, absence, and distribution of
cultural materials and architectural features,
since these factors relate directly to a site's
behavioral/informational integrity. A study of
this type, while it may not be suitable for quantified predictions, will nevertheless allow
generalizations to be made concerning the type of
impacts to he expected in surface-disturbing chaparral management activities. Thus, the test
program was designed to observe three specific
aspects of potential impact. The first two
involved recording the displacement, breakage,
addition, and loss of (1) surface artifacts and
(2) structural components (wall stones) located
within designated one meter square test plots.
The third aspect of the program was to investigate subsurface impacts. To do this, a series of
simple artificial test situations were constructed in the study area by burying sets of
large and small ceramic flower pots to simulate
subsurface artifactual remains. Specific burial
locations were selected according to considerations of soil texture and vegetative cover, as
were the test squares. Each burial contained two
sets of pots at different depths. These depths
were determined on the basis of excavation data
from similar environments nearby. In order to
control for the effect of time on surface artifact distribution, all the surface plots and
burials were laid out two months before the
crushing operation was scheduled to begin.
The location selected as a control was a
single room rock-outlined pit house habitation
site. It was selected because it contained a
well-defined architectural feature, surface artifact scatter, and showed a strong dichotomy
between areas with and without brush cover and
having rocky and fine textured soils. These criteria were used for all test selections in the
study. The only impacts expected here were time,
rainfall, and the curiosity of other Forest
personnel.
The location selected for the architectural
tests was a fairly well-defined architectural
feature with artifact scatter and little brush
cover. The area selected to test impacts to surface artifact scatter contained a fairly dense
concentration of surface artifactual material
situated near the architectural test locus. It
showed a strong dichotomy between brush cover and
bare ground within a minimal area as well as a
wide variety of artifact types.
In order to test impacts to subsurface
materials and features, a series of control and
test burials was laid out at various points; the
control site and in a separate testing locus some
distance away. The burial tests and controls
were all placed according to the same criteria as
the surface plots-brush cover and bare ground-with additional variations in depth and soil
texture.
THE CRUSHING OPERATION
The Marden device used in this study was
pulled by a D7H Caterpillar tractor. The strategy utilized was more or less representative of
standard Marden crushing operations. It involved
a series of concentric, overlapping paths being
followed throughout the treatment area. Those
areas too steep for safe operation (above 20 percent slope) were avoided, as were areas where the
vegetative cover was primarily trees or that were
too rocky for the somewhat brittle blades on the
crusher. The end product of this strategy was a
natural-looking mosaic of pine-oak stringers,
brushy hillsides and outcrops, and crushed flats
which should eventually be grassed over.
The crushing of test plots and burials was
more or less a sample out of a normal crushing
operation. Owing to unfortunately restrictive
time and budget limitations (so what else is
new?) the test loci could not be crushed entirely
517
in the overlapping pattern usually followed.
Instead, carefully choreographed paths were laid
out through them so that the crusher made a
single pass over each test plot and burial.
Because of this, the results of these tests must
be seen as representing only the minimum level of
impact that might result from use of the Marden
device.
RESULTS AND COMPARISONS
Fifty-two surface artifacts were originally
identified in the four surface test squares.
Almost 90 percent of these were impacted in some
way. Sixty-two percent of these impacts involved
the loss of artifactual materials from the observable surface inventory. Next in importance were
the displacement of original material and the
addition of new material brought up from below
the surface. Breakage was the least frequent
impact. The patterns of change at the control
site were exactly the opposite. Natural causes
produced primarily breakage of artifacts in
place-simple deterioration, with minor relocation
by runoff and periodic additions to surface
inventories from subsurface materials. Below
surface, no displacement of materials was
observed nor were there any changes in the structure. As expected, these observations indicate
that the natural transformation of cultural
features and materials into what are known as
archeological sites is a slow process which preserves much of the original integrity of the
property. The pattern of change produced by
Marden impact is characterized by a sudden loss
of a large amount of the surface artifact inventory with little replacement from the subsurface
population. This, combined with breakage and
displacement, means a sudden change in 90 percent
of the surface artifact inventory and nearly half
of the near-surface buried material. In
addition, disruption of surface architectural
remains runs from 40 to 100 percent, depending on
the amount of exposure. In the case of a very
shallow site with little or no post-occupation
deposition, this would constitute a near-total
disruption of the visible portion of that site.
The recognition of any site's character-shallow
or deep-would be seriously impaired if suddenly
over half of the visible artifacts were to
disappear. And whatever its direct effects are
on a site, the Marden device invariably opens and
softens soil surfaces, making them susceptible to
erosion and to subsequent further artifact
displacement by mixing from grazing, vehicular
travel, and even walking.
Factors in Assessing Impacts
Damage from the Marden device as observed here
was dependent on four factors: 1) contact with
the blades; 2) soil texture; 3) depth; and
4) artifact size. Contact with a blade
invariably produced both surface and subsurface
518
artifact damage or displacement. Rocky soils or
soils with expansive clays produced the highest
percentage of subsurface damage, while fine soils
produced the most damage (by displacement) in
surface inventories. However, it must be noted
that the presence of residual clays and large
numbers of rocks in a surface soil is usually a
result of natural, in place development rather
than cultural deposition, though this is not
always the case. While it was expected that certain amounts of cover would act to provide a protective cushion over cultural surfaces and fill,
the results of the tests indicate little or no
difference due to cover. It can be said, then,
that cover has no effect and is therefore not a
factor in the amount of impact. The third factor, depth, proved highly useful, as damage was
higher in shallow burials than in deep, and
highest on the surface itself-25 percent of deep
burials were impacted, 75 percent of shallow
burials, and 86 percent of the surface inventory.
Clearly, impact increases as depth decreases.
Finally, there was the factor of artifact size.
The ratio of large to small buried pots broken
(by expansive clay) at the control locus was 3 to
1; at the test burial locus, it was 5 to 1. As
well, only the larger surface artifacts were ever
physically damaged. This too indicates a strong
pattern-larger artifacts have a higher susceptibility to damage. In some cases this was
apparently due to larger area of potential blade
or rock contact, while in others it may have been
due to the larger artifact's having had less
structural integrity or resistance to stress than
a smaller one of the same materials. Unfortunately, these observations must remain tentative
and inconclusive, as much more variability was
encountered than expected in soil, surface, and
brush composition. Still, it can reasonably be
said that under the variety of conditions tested,
nearly 90 percent of the cultural surface and
50 percent of the subsurface inventories were
damaged by crushing.
SUMMARY AND CONCLUSIONS
Three direct impacts of brush crushing to
cultural properties were identified from this
study: 1) disruption of the spatial relationships of surface and subsurface artifacts;
2) disruption of structural elements in surface
and subsurface architectural features; and
3) physical damage to surface and subsurface
artifactual materials. The disturbance of artifact distributions is seen as the most critical
of these, since disrupting or destroying the spatial context of these remains severely limits any
attempt to characterize and identify not only
specific behaviors at sites but sites themselves
and regional and the regional and chronological
patterning of occupations and developments.
Indirect impacts to these properties may also
arise from erosion and, in some cases, from
increased site visibility, which tends to invite
vandalism. This being the case, it must be
recognized that the use of a Marden Brush-Crusher
in areas containing cultural resources cannot be
allowed without some measure of protection being
given to those resources.
Acknowledgements: This paper represents a
summary of the original report (Wood 1979) which
has been published in the USDA Forest Service
Southwestern Region's Cultural Resources Reports
series (No. 27). Beyond recognizing the support
provided by the Prescott and Tonto National
Forests, I would like to thank my crew, Raven
MacReynolds and Jim Mackin of the Prescott
National Forest and, especially, Harlow Yaeger,
formerly with the Prescott National Forest,
without whose knowledge and expertise this project could not have been carried out. I would
also like to acknowledge the editorial assistance
of my wife, Patti Fenner, a range conservationist
with the Tonto National Forest, who aided greatly
in adapting the paper to this forum.
LITERATURE CITED
DeBloois, Evan I.; Green, D.F.; Wylie, H.G. A
test of the impact of pinyon-juniper chaining
on archeological sites. Ogden, Utah:
Intermountain Region, Forest Serv., U.S. Dep.
Agric.: Archeological Reports; 1974.
Gallagher, Joseph G. Scarification and cultural
resources: an experiment to evaluate serotinous lodgepole pine forest regeneration techniques. Plains Anthropologist 23-82, Pt. 1:
289-299; 1978.
McGuire, Randall H. Prehistoric subsistence at
the Kiewit site, northwestern Arizona. Tucson,
Ariz.: The Kiva 43(1):49-82; 1977.
Wood, J. Scott. An archeological survey of a
brush control crushing area near Camp Wood
administrative site, Walnut Creek Ranger
District, Prescott National Forest. Prescott,
Ariz.; Prescott N.F. Cult. Res. Inventory
Report (Manuscript); 1978.
Wood, J. Scott. Chaparral conversion and
cultural resources on the Prescott National
Forest: an experimental study of the impacts
of surface mechanical treatment by Marden
Brush-Crusher. Albuquerque, New Mexico:
Southwestern Region, Forest Serv., U.S. Dep.
Agric. Cultural Resources Reports 27; 1979.
Wood, J. Scott; Yaeger, Harlow. Research design
for an experimental test of the impacts on
archeological sites of chaparral conversion by
Marden Brush-Crusher, Camp Wood, Walnut Creek
Ranger District, Prescott National Forest.
Prescott, Ariz.; Prescott N.F. Cult. Res.
Inventory Report (Manuscript); 1978.
519