Pima County’s Withdrawal from Its Past
Raymond M. Turner

Reconstruction of Pima County’s past vegetation from the time of the first European intrusion is an undertaking charged with obstacles.
Few detailed descriptions were made of the vegetation encountered by those early intruders.
These visitors often recorded the presence of conspicuous plants found throughout the entire
region, but specifics were missing. Thus, we
find written references to saguaros, greasewood
(creosote bush), sagebrush (saltbush), and palo
verdes by travellers passing through the region,
but these generalized references do little to resolve the mystery of where rough boundaries
between various plant communities might have
been, nor do they offer a baseline for judging what
changes might have occurred following those
early days.
Interest in the history of Pima County’s vegetation has increased in the wake of efforts to
retain certain habitats as last-ditch refuges for
waning populations of threatened organisms. In
addition, the rapidly expanding need for space
to accommodate the region’s growing human
population has made inroads into rural areas that
were but slightly impacted by humans only a
short time ago. Knowledge of a vegetation’s history may aid in directing its destiny by providing a hint of the ways in which we might steer
the forces of change. What is the effect of fire,
livestock grazing, and climate on the vegetation of our region? One method that provides
partial access to vegetation history is the art of
photographic matching or repeat photography.
Repeat photography as a tool for documenting landscape change has been with us for over

100 years when it was first used to document
changes in glaciers.[1] It was not used for studying shifts in vegetation until 1957, when Homer
L. Shantz, former president of the University of
Arizona, traveled across Africa, matching landscape photographs he had taken on a trip from
South Africa to the Belgian Congo in 1920.[2]
The vegetation of Pima County is as varied as
its topography, from peaks ascending to more
than 9,000 feet to desert valleys descending to
roughly 1,000 feet. At the higher elevations, quaking aspen and corkbark fir mingle with other conifers and forest plants whose main distributional
ranges extend far to the north; in the low valleys, desert species such as creosotebush and
saguaros, with affinities to the south, dominate
the landscape. Photographs of Arizona’s varied
landscape have been taken since 1867, although
the first from Pima County were probably taken
later. By selecting a few of the several hundred
historic landscape views available from within
Pima County and comparing those with exactly
matched recent views, we can detect trends that
may prove useful in making decisions about future management and use of Pima County land.
In this survey, we have obtained early photographs from various archives, with dates ranging from 1880 to 1942. We have attempted to
provide examples from the main vegetation
types that are represented in Pima County. This
survey of matched photographs will begin at
upper elevations and descend gradually to
lower points across the landscape, except that
riparian habitat comparisons will be set aside
and addressed separately.[3]

[1] For a review of the use of repeat photography through
the early 1980s, see Rogers et al. (1984).
[2] See Shantz and Turner (1958).
[3] One photograph (Figure 13) is presented without a photographic match because that site has not been visited to
obtain a more recent photograph.

1

2

3

nado National Forest
aro National Park
n Pipe Cactus National Monument
no O'odham Nation
za Prieta National Wildlife Refuge

(Mexico)

Map by Peter Griffiths.

31˚3

32˚30'

113˚

4

Figure
Town

Major road
Water course

6

4

ree
ints

7

9

8

12

13

15
4

Tucson

111˚

20 miles

3 2

5

11

1

Figure 1. Mica Mountain as seen from Man Head. Elevation: 8428 feet.

Figure 1a. Lawrence M. Huey photograph. June 1932. Courtesy of the Desert Laboratory Archive.

(1) Rincon Mountains
Mica Mountain, with an elevation of 8666
feet, is the highest point in the Rincon Mountains, on Tucson basin’s eastern rim. The peak’s
southern slope supports a forest of ponderosa
pine, the dominant tree in forests at this elevation. This tree is well adapted to survive low intensity ground fires once it becomes mature. One
set of records maintained by Saguaro National
Park shows that 411 lightning fires occurred in
the Rincon unit of the park between 1937 and
1986. Mostly small in extent, over 350 of these
fires occurred around Mica Mountain at eleva-

4

tions above 6500 feet. Centuries-long tree-ring
records show that the frequency of large fires on
Mica Mountain held at about one fire every six
years until the 1890s, after which the return interval lengthened due to fire suppression.[4] The
forest on Mica Mountain as seen in 1932 from
the top of Man Head Rock (Figure 1a) has therefore experienced roughly three decades of low fire
frequency. Figure 1b, taken in 1973, shows the
fire lookout tower that stood on Mica Mountain
for several decades but has now been removed.
The slope below the tower experienced a stand

Figure 1b. Martin A. Turner and Roger C. Wolf. April 15, 1973. Courtesy of the Desert Laboratory Archive.

replacing fire in 1943 (the Manning Camp Fire)
and a few new trees have become established
during the 30 years since then. This 5,741 acre
fire was the first large fire to occur following the
beginning of fire suppression. Several trees growing just below the camera station have died during the interval between photographs, making
patches of the forest floor more visible in the later
view. Many small trees in a fire-sensitive size range
have become established.

[4] See Baisin and Swetnam (1990).

5

Figure 2. Near the Helena Mine at the north end of the Santa Rita Mountains. View N40ºW. Elevation: 4600 feet.

Figure 2a. F. C. Schrader photograph. 1909. Courtesy of the U.S. Geological Survey Photographic Library, Denver.

(2) Helena Mine, Santa Rita Mountains
The Santa Rita Mountains have been an area
of active mining since the late 1800s. In 1909,
geologist Frank Schrader, working at the northern end of the mountain range, took photographs
in the vicinity of the Helena Mine (Figure 2a). At
an elevation of 4600 feet, this mountainous terrain is well above the Sonoran Desert and supported grassland, although incipient shrub invasion was evident by the time of Schrader’s visit.
Scattered mesquites and ocotillos, leafless in this
winter photograph, have attained a size consistent with the age of a decade or more. The camera is aimed toward the northwest and captures a

6

few evergreen oaks growing along minor northfacing canyons. The low, dark shrubs at
midground are fairyduster, a palatable plant that
is closely cropped by livestock.
When visited in November 2002, the mesquites had increased in density although the individual plants were relatively small considering
the century-long life of some (Figure 2b). Their
small size is undoubtedly partly the result of catastrophic freezes. The persistent dead branches on
many of the mesquites is typical of plants frozen
back by the severe freeze of December 1978.[5]
The live oaks are no longer present along the

Figure 2b. R. M. Turner photograph. November 5, 2002. Courtesy of the Desert Laboratory Photograph Archive.

north-facing canyons. Since this site is within a
few hundred feet of a former mine stamp mill
which used many ricks of firewood during its
mining heyday, we searched for signs that the
oaks had been removed for stoking furnaces.[6]
We found, lying on the ground where the oaks
had stood, carcasses of large oaks that had fallen
in place without signs of woodcutting activity.
Death of these long-lived trees, which are resistant to freezing temperatures and fire, is probably the result of the 1950’s drought.[7] The hills
support many more ocotillos than earlier; the
fairy duster has increased in number only slightly.

[5] For a description of the effect of the 1978 freeze on mesquite in the nearby San Pedro Valley, see Glinski and Brown
(1982).
[6] By turning his camera to the right, Schrader took another photograph that shows the stamp mill and ricks of
firewood. (F. C. Schrader #1533, USGS Photograph Library,
Denver). Also see Plate 37 in Turner et al. (2003).
[7] See Swetnam and Betancourt (1998).

7

Figure 3. Gently sloping bajada of the Santa Rita Mountains as seen from Huerfano Butte. View northeast. Elevation:
3900 feet.

Figure 3a. Photograph by David Griffiths. 1902. Courtesy of the National Archives.

(3) Huerfano Butte
Huerfano Butte is a conspicuous isolated
prominence east of Sahuarita on the Santa Rita
Experimental Range. David Griffiths climbed to
the top in 1902 and took a series of panning photographs. He wrote at the time that “close examination of the broad, gentle, grassy slopes between
the arroyos. . . reveals a very scattering growth of
mesquite (Prosopis velutina) which is in the form
of twigs 2 to 3 feet high with an occasional larger
shrub in some of the more favorable localities.”[8]

8

His suspicion that the shrub population was likely
to grow was confirmed after seven more years of
observation when he asserted that “the time is
coming when these foothill grassy areas, which
now have only an occasional small shrub, will be
as shrubby as the deserts and lower foothills . . .
if not more so.”[9] Burroweed, a small shrub, had
also “thickened and increased perceptibly during
the last five years.”[10]
Figure 3a , one of the Griffiths panning pho-

Figure 3b. R. M. Turner. 1986. Courtesy of the Desert Laboratory Photograph Archive.

tographs, faces toward the northeast. (The mining community of Helvetia lies at the base of the
Santa Rita Mountains, to the left of center.)
Griffiths’ description of the vegetation is borne
out by the scene above. The shrubs along the
water-courses are probably mostly mesquite, although the large darker shrubs in the foreground
are desert hackberry, identifiable because they persist to the present. The broad slopes between the
runnels are grass covered, as noted by Griffiths.

Figure 3b shows the same area after 84 years.
Griffiths’ foresight about the increase of shrubs
was remarkable. Shrubs, mainly velvet mesquite,
now dominate the landscape. The grassy matrix
has been largely filled by the small shrub,
burroweed, as foreseen by Griffiths. Note the persisting desert hackberries, many of which have
changed little in size since 1902.

[8] See Griffiths (1904).
[9] See Griffiths (1910).
[10] Ibid.

9

Figure 4. View west-northwest from the western edge of the Altar Valley toward Baboquivari Peak. Elevation: 3500 feet

Figure 4a. Grama grassland with a scattering of velvet mesquites some of which are probably a few decades old.
Photograph by David Griffiths. Ca. 1912.*

(4) Baboquivari Peak and the Altar Valley
David Griffiths, botanist with the Bureau of
Plant Industry in Tucson and early observer of
Pima County vegetation, took a photograph
in about 1912 at the south end of the Sierrita
Mountains on the edge of the Altar Valley,
showing a dense growth of grama grass (Figure
4a). Baboquivari Peak can be seen in the background. A scattering of velvet mesquites is already visible across the rolling hills of this grassland which is part of a cattle ranch. Roughly

90 years later the terrain is no longer grassland
and has been overtaken by woody vegetation,
including velvet mesquite, ocotillo, catclaw,
and gray-thorn (Figure 4b). Except for ocotillo,
all are relatively resistant to fire after reaching
large size. During earlier times when grass was
abundant as fuel and frequent fires were not
vigorously suppressed, these woody plants
would have been kept at low levels across this
landscape.

* Plate 71B in Griffiths, 1912.

10

Figure 4b. The grama grasses present earlier have been replaced by three-awn grasses. Woody plants such as
ocotillo and mesquite have markedly increased. R. M. Turner photograph. May 9, 2002. Courtesy of the Desert
Laboratory Photograph Archive.

11

Figure 5. View to the north across the northwest corner of the Rincon Unit of Saguaro National Park. The Santa
Catalina Mountains are in the background; the course of Tanque Verde Creek is marked by the distant cottonwood
trees. Elevation: 3040 feet.

Figure 5a. The saguaro forest that prompted the establishment of Saguaro National Park appears across the scene.
The recently constructed Mount Lemmon Highway ascends the mountain on the right. Ca. 1935. Courtesy of the
National Park Service.

(5) Saguaro National Park
Saguaro National Monument east of Tucson was established in 1932 to protect the dense
saguaro forest at the base of the Rincon Mountains. The Santa Catalina Mountains provide the
backdrop for a view of that same saguaro forest
(Figure 5a). The photograph was taken in about
1935, shortly after the Monument’s establishment. Investigators at that time noted the near
absence of young saguaros and predicted that this
population of predominantly old plants was not
sustainable. They predicted that the old plants,
many of which were nearing their upper age limit
(ca. 175 years), would inevitably die, leaving

12

Saguaro National Monument without a saguaro
forest.
By 1995 (Figure 5b), the saguaro forest had
indeed disappeared, upholding the biological
truth that populations cannot be sustained without reproduction. A permanent study plot, established in 1961 in the area seen in this view, recorded the saguaro’s decline. In 1961, this nineacre plot supported 209 saguaros; by 1983 the
number had plummeted to 100. Undetected
at that time, reproduction had begun and the
many small plants then present were not
counted. By 2001, however, when the plot was

Figure 5b. Change is widespread. Although Tanque Verde Creek still supports a line of cottonwood trees, most of the
saguaros have succumbed to old age and the distant bajada at the base of the mountain supports many houses. R.
M. Turner photograph. March 19, 1995. Courtesy of the Desert Laboratory Photograph Archive.

last examined, they were large enough to be
counted and the population stood at 227. Although this is more than the number present
in 1961, it is probably well below the value for
the1930s, when the first photograph was taken.
Because saguaros don’t become sexually mature
until they reach roughly 40-50 years of age,
the reproductive potential of this forest of
mainly immature saguaros will remain at an
ebb for several more decades.
The cause for the saguaro’s decline is possibly the result of two forces: livestock grazing,
which was present until the late 1950s, and

cutting of saguaro “nurse trees” (mesquites and
palo verdes) to fuel a nearby charcoal kiln,
which operated until after 1900. Removal of
“nurse trees” reduced the protective cover essential for establishment, and cattle seeking
shade under the remaining trees may have
trampled many of the seedlings and small
plants that managed to get started. Absent
these negative forces, periods of abundant rainfall during recent decades have brought back
the saguaro in large numbers.

13

Figure 6. Coyote Mountains as seen from the western edge of Avra Valley. Elevation: 2985 feet.

Figure 6a. Botanist David Griffiths photograph. 1903. Courtesy of the National Archives.

(6) Coyote Mountain
When botanist David Griffiths visited the
Avra Valley west of Tucson in 1903, he found a
dense sward of grass and forbs covering the
valley floor at an elevation of just under 3,000
feet (Figure 6a). At this elevation, the Sonoran
Desert meets the region’s semidesert grassland.
The distant saguaros reveal the desert’s tenuous grip here, just as scattered yuccas reveal
the station’s grassland ties. As with much of
the Avra-Altar Valley, this area had been grazed
by cattle for only two or three decades at the
time of Griffiths’ visit. This area became part
of the Papago Indian Reservation in 1916, and

14

has continued under a livestock management
regime since then.
Almost 100 years later (Figure 6b), the
change is glaring. Both saguaros and yuccas
persist, but the grass-forb turf is gone and many
new cacti and shrubs have become established.
Woody plants, such as creosote bush (extreme
right), white-thorn (leafless at left foreground),
velvet mesquite, and foothill palo verde, dominate, although succulents such as Arizona
yucca, prickly pear cactus (three species) and
cholla (two species) are also important constituents.

Figure 6b. R. M. Turner photograph. April 12, 2002. Courtesy of the Desert Laboratory Photograph Archive. Access
to this camera station was granted by the Tohono O’Odham Nation, and is gratefully acknowledged.

15

Figure 7. View to the northwest toward Panther Peak (extreme right) and the Avra Valley from a low hill in the
Tucson Mountains. Elevation: 2740 feet.

Figure 7a. Forrest Shreve photograph. March 20, 1916. Courtesy of the Desert Laboratory Photograph Archive.

(7) Panther Peak and the Avra Valley
The palo verde-saguaro desert, found on
hills and bajadas in much of Pima County at
elevations below 3,000 feet, is seen in a view
toward the northwest from a low hill in the
Tucson Mountains (Figure 7).
As noted in Figure 9, the limestone hill
called Picacho Calera has been removed as a
source of raw material for a cement plant at
nearby Rillito. By the time of the second photograph in 1995, the camera station had been
part of the National Park Service’s Saguaro National Monument (and then Park) for 34 years;
before that, it had been part of Pima County’s

16

Tucson Mountain Park. Widely scattered dwellings in Avra Valley can be seen just beyond the
Park boundary. During the 79 years between
photographs, several prickly pear cacti have become established near the camera station. Biomass differences in other plants (such as
saguaro, foothill palo verde, ironwood, ocotillo,
brittlebush, and triangleleaf bursage), as captured by these two scenes, are probably negligible. Such views illustrate the relative stability of the Sonoran Desert in situations free of
grazing impacts.

Figure 7b. Dominic Oldershaw photograph. January 18, 1995. Courtesy of the Desert Laboratory Photograph Archive.

17

Figure 8. Creosote bush dominated the area in the northwest corner of the Desert Laboratory grounds. View to the
west with the Tucson Mountains in the background. Elevation: 2430 feet.

Figure 8a. Forrest Shreve photograph. July 1928. Courtesy of the Desert Laboratory Photograph Archive. The arrow
marks a rock that is visible in both photographs.

(8) The Desert Laboratory
The Desert Botanical Laboratory of the
Carnegie Institution of Washington was created
in 1903. The 869-acre parcel on Tucson’s west
side was encircled by a fence in 1907, and a roadway crossing the northwest corner of the Desert
Laboratory was abandoned as a result. Botanist
Forrest Shreve photographed the road in 1928
to document its status at that time (Figure 8a).
Prior to fencing, the area had probably been
grazed by livestock for many years with dominance of the unpalatable creosote bush the re-

18

sult. After 21 years free from livestock grazing,
the area was still dominated by old creosote
bushes, leftovers from pre-fence days. The vegetation had probably changed little since fence
construction, although small plants of triangle
leaf bursage appear in the intervening spaces,
marking a resurgence of this more palatable species. In 1998, following seventy more years without livestock, the old roadway has almost disappeared (Figure 8b). Gone also, are most of the
creosote bushes. Only three of these long-lived

Figure 8b. R. M. Turner photograph. April 10, 1998. Courtesy of the Desert Laboratory Photograph Archive.

residents of North America’s warm desert regions
remain in view (circles). In its place are triangle
leaf bursage, fairy-duster, and range ratany, all
palatable to various classes of livestock. Also showing a notable increase are ocotillo, prickly pear,
cane cholla, foothill palo verde, and saguaro.
The Desert Laboratory property is now mostly
surrounded by houses; some of the dwellings
lining the western boundary of the Desert Laboratory are visible in this view. The arrows point
to a rock that is visible in both photographs.

19

Figure 9. View to the southwest from “The Island,” north of Tucson. The camera station is on a small hill near what
is presently Tangerine Road. Elevation: 2228 feet.

Figure 9a. The two isolated hills right of center are Picacho Gemelo (right) and Picacho Calera (left). Avra Valley lies
beyond. The Santa Cruz River flows from left to right across the scene, passing close to the base of the dark mountain
at left. The tiny settlement of Rillito is marked by the long clump of trees near the river. Forrest Shreve photograph.
June 20, 1930. Courtesy of the Desert Laboratory Photograph Archive.

(9) “The Island”
In 1930, botanist Forrest Shreve climbed a
small hill north of Tucson and took several photographs. The hill, known by Shreve and other
workers at the Desert Laboratory as “The Island,”
served as a perfect vantage point for viewing
Sonoran Desert vegetation. One of Shreve’s photographs looks toward the southwest with northern outliers of the Tucson Mountains in the distance. In 1930, the level plain is dominated by
foothill palo verde (Figure 9a). Note the absence
of saguaros, a Sonoran Desert dominant that accompanies the palo verde on rocky hills but is
often absent on level plains. The small shrubs near
the camera position are brittlebush and

20

triangleleaf bursage.
Sixty-five years later (Figure 9b), the plain is
still almost free of saguaros but the palo verdes
have increased in size, if not numbers. The spaces
between the trees support more small shrubs than
before. This recent increase in palo verdes has
been noted in other photograph pairs from sites
near Tucson, and is made all the more intriguing by the written observation of Forrest Shreve
suggesting that these plants might have been
newcomers a decade or so earlier. Shreve wrote
in 1911 that “I have had a great many thousands . . . come under my observation . . . and
have seen only two dead trees of full size.”[11]

Figure 9b. Picacho Calera is gone, having been the source of limestone for the Arizona Portland Cement plant
occupying the Rillito townsite since 1948. Dominic Oldershaw photograph. January 18, 1995. Courtesy of the
Desert Laboratory Photograph Archive.

Fourteen years later, however, he noted that
many palo verdes of all ages had died on
Tumamoc Hill.[12] Subsequently this decline
was reversed and then, in the late 1990s, another pulse of palo verde deaths was documented at Tumamoc Hill and elsewhere in the
Tucson region.[13] Thus, over almost a century,
observations of this desert tree show that its
populations undergo episodic dieback and revival. What apparently is seen in the photographs
from “The Island,” are two points in one or
more of these population pulses.
The reach of the Santa Cruz River crossing
this scene is just downstream from Tucson’s

sewage treatment plants; streamflow is perennial here after many decades of only intermittent flow. The trees marking the distant Santa
Cruz channel (near the base of the dark mountain) are mainly the riparian stalwarts Fremont
cottonwood and Goodding willow, which make
use of the abundant water at this site.

[11] See Shreve (1911).
[12] This reference to palo verde deaths at Tumamoc Hill is
in Shreve (1924/25).
[13] See Bowers and Turner (2001) for a description of recent dieback of palo verde at Tumamoc Hill.

21

Figure 10. View north-northwest of the creosote bush desert along the Camino del Diablo, Organ Pipe Cactus
National Monument. Tracks of the old Camino del Diablo can be seen crossing the foreground. The present-day
camino is about 250 yards to the north. Growler Mountains in the background. Elevation: 1280 feet.

Figure 10a. Tad Nichols photograph. November 1935. Courtesy of the Desert Laboratory Photograph Archive.

(10) Organ Pipe Cactus National Monument
The creosotebush copes with variations in
moisture availability by losing branches during
times of little rain and then producing new ones
when good times return. Over time, the stature
and density of stands of this shrub experience
slight pulses in biomass but maintain a size range
set by the conditions of the habitat. The shrubs
in a pair of photographs from west of Bates Well
on the Organ Pipe Cactus National Monument
in extreme western Pima County, illustrate this
drought pruning and regrowth phenomenon (Figures 10a and 10b). Fifty-eight years have elapsed
between the two photographs, and during this
time the shrubs have increased slightly in size

22

and number of branches. The increase in size
probably reflects rainfall during immediately preceding years rather than any long-term trend. No
new plants of this shrub became established during almost six decades. Much the same can be
said for the other plant species seen here. Jumping cholla, ironwood (center midground), white
ratany (small plants scattered among the creosote bushes), and ocotillo seem to have about
the same representation in both views. Saguaro
numbers are roughly the same although cacti at
midground have died. In this arid environment
with annual average rainfall of about 8 inches,
creosote bush, along with white ratany, occupy

Figure 10b. R. M. Turner photograph. April 1991. Courtesy of the Desert Laboratory Photograph Archive.

the low ridges between the intervening
drainageways. These runnels with their slightly
improved moisture tend to support saguaros and
desert trees such as foothill palo verde and ironwood, giving the landscape a distinctive lined appearance when viewed from above.

Changes on the region’s rivers
The magnificent cottonwood-willow forests
that grow along the rivers of southeastern Arizona are often described as endangered or declining. A frequently cited statistic states that today
they cover only 10% of their former large expanse.
The source of this statistic is unclear, but per-

haps it arose from studies along the lower Colorado River where dam construction and channel revetments have so drastically altered natural conditions that, indeed, this once abundant
riparian forest is nearly gone. The status of this
forest along the streams of southeastern Arizona is far different than in earlier years. There
has actually been a many fold increase in the
extent of this forest since the beginning of
Anglo settlement. Evidence of this expansion
comes from both written accounts and old photographs. Evaluation of landscape change using long interval photography has proven especially useful in documenting these changes.

23

Figure 11. Cienega Creek, east of Tucson, as seen from a low ridge. View to the northwest. Elevation: 3612 feet.

Figure 11a. C. E. Watkins photograph. 1880. © Huntington Library.

(11) Cienega Creek
One revealing pair of photographs shows
Cienega Creek east of Tucson in 1880 and again,
more than a century later, in 1998 (Figure 11).
Figure 11a shows the freshly laid tracks of the
Southern Pacific Railroad running alongside
Cienega Creek, represented here by a narrow, essentially treeless channel. The valley floor is covered by grasses and is free of mesquites and other
woody plants, although shrubs are common on
the adjacent uplands. (Note the brush fence run-

24

ning along the ridge at right midground.) The
later photograph records remarkable changes (Figure 11b). The grass-covered valley floor is now
occupied by a dense mesquite forest. The Cienega
Creek channel is deeper and broader and supports
a forest of cottonwoods and willows. The railroad
tracks were moved across the river after the original bridge washed away and a new bridge was
built upstream from this site in 1912.

Figure 11b. R. M. Turner photograph. January 24, 1998. Courtesy of the Desert Laboratory Photograph Archive.

25

Figure 12. The eastern-most pier on the Continental Bridge at the Santa Cruz River as seen from the north side of
the bridge. Elevation: 2818 feet.

Figure 12a. Photographer
unknown. June 1940.
Courtesy of the U. S.
Geological Survey, Tucson.

Figure 12b. The same pier with U.S.
Geological Survey stage-gage housing
attached to the bridge. The graduated
pole is marked in feet. The arrow shows
the approximate position of the 1940
streambed. R. M. Turner photograph.
November 1978. Courtesy of the
Desert Laboratory Photograph Archive.

(12) Santa Cruz River at Continental
The downcutting along the Santa Cruz River Channel is
shown by a pair of photographs taken of one of the piers on
the old Santa Cruz River Bridge at Continental. In June 1940,
the river bed lies perhaps five feet below the top of the pier
(Figure 12a). By November 1978, downcutting had removed
an additional five feet of streambed material (Figure 12b).
Matching these photographs today would not be possible;
the bridge was replaced following the flood of 1983, which
severely eroded approaches to the bridge.[14]
[14] The downcutting shown in this pair of photographs may merely
record a brief decline of streambed elevation during an otherwise steady
rise in this datum. In striking contrast to the decline shown by this pair of
photographs, Parker (1995), using measurements from three widely separated years at the Continental Bridge, shows an increase in elevation of 10
feet from1929 through 1976 to1985.

26

Figure 13. Santa Cruz River Channel downstream from the Pima Mine Road Bridge. This canyon was formed when the
old Tucson-Nogales wagon road was captured by the Santa Cruz River in about 1940. Note the vehicle for scale.
Photograph by R. M. Turner. December 15, 1981. Courtesy of the Desert Laboratory Photograph Archive. Access to this
camera station was granted by the Tohono O’Odham Nation, and is gratefully acknowledged. Elevation: 2650 feet.

(13) Santa Cruz River at Pima Mine Road
A story of downcutting along the Santa Cruz
River with a slightly different twist is illustrated
by viewing the Santa Cruz channel just downstream from the bridge at Pima Mine Road. This
station is approximately 25 miles downstream
from the Continental Bridge and seven miles upstream from our next station in this series at
Martinez Hill. First, a visit to this site will reveal
that the river’s course here is not at the lowest
point in the valley, which lies perhaps another
mile to the east. This seeming defiance of natural
design arose when the old Tucson-Nogales wagon
road, which paralleled the Santa Cruz to the west,
was gradually eroded by rainwater intercepted
from minor arroyos running out of the Sierrita
Mountains. (The road’s presence here is verified
as shown on the original land survey maps for

this region.) By 1940, the old wagon road had
eroded into a deep channel which later migrated
upstream to capture the main flow of the Santa
Cruz, thus rerouting the river to its present course
in the valley.[15] Figure 13 (the only photograph
in our series not offering a second look) shows
the channel as it appeared in 1981. The channel
through this reach is basically an artifact of man’s
activities. The absence of riparian vegetation is
probably the result of the resistance of the soil to
lateral erosion. Seedlings that become established
along the floor of the miniature canyon are swept
away by relatively minor floods. Thus, in this
reach of the Santa Cruz, man’s activities shifted a
river channel having the potential to support a
riparian forest to a new location where geologic
constraints precluded the growth of such a forest.

[15] Betancourt, Julio L., and Raymond M. Turner. Unpublished ms. Tucson’s Santa Cruz River and the Arroyo Legacy. On
file at U. S. G. S Desert Laboratory office.

27

Figure 14. View from the top of Martinez Hill to the south toward the Santa Rita Mountains. Elevation: 2598 feet.

Figure 14a. 1942. Courtesy of the Arizona Game and Fish Department.

(14) Martinez Hill
The Santa Cruz River passes along the base of
Martinez Hill east of the San Xavier Mission. This
is a new route for the river. Its earlier course had
been farther west, closer to the mission, until a
cut-off dike was built in 1915, diverting the river
eastward. Thus, like much of the river’s course
today, this reach to a large extent is man made.
In a photograph taken in 1942 (Figure 14a), the
channel had been deeply cut by erosion and sup-

28

ported a narrow forest of cottonwoods. Tall cottonwoods grow along the channel and a low mesquite woodland appears above the vertical bank
within the river’s meander. According to the photographer, the dense mesquite forest on the old
flood plain by this date had been greatly altered,
and, as he note: “This mesquite forest . . . is one
of the oldest stands for nesting and roosting of
the Whitewing [dove] in the U.S. Cutting of the

Figure 14b. R. M. Turner photograph. 1989. Courtesy of the Desert Laboratory Photograph Archive. Access to this
camera station was granted by the Tohono O’Odham Nation, and is gratefully acknowledged.

mesquite has greatly depleted the Whitewing
population from its once fabulous abundance.”
By 1989 (Figure 14b), the channel had been widened, eliminating the riparian forest.[16] The mesquite forest has continued to decline, perhaps by
cutting, but also because of heavy groundwater
pumping which commenced about 1940. The
dead or dying trees are harvested for firewood by
the local inhabitants.

[16] See Parker (1995), for a detailed description of channel
changes along the San Xavier reach of the Santa Cruz River
at Martinez Hill

29

Figure 15. Santa Cruz River as seen looking upstream from the Congress Street Bridge. Elevation: 2330 feet.

Figure 15a. November 22, 1930. Photograph courtesy of the U. S. Geological Survey, Tucson, Arizona.

(15) Congress Street Bridge
The Santa Cruz River bisects Tucson along its
historical midriff. In the city’s early days, the
river’s perennial flow and its cottonwood- willow forest provided Tucson’s citizens with a welcome change from the surrounding desert. By
1930, perennial flow at the Congress Street Bridge
had ceased but remnants of the riparian forest

30

still existed (Figure 15a). Sixty-five years later the
view from the bridge emphasizes man’s works and
his attempts to control the river’s force (Figure
15b). Both banks are stabilized, a trail runs along
one side of the channel, and the river is still
ephemeral here, although for short periods, as
on January 5, 1995, water flows beneath the bridge.

Figure 15b. Dominic Oldershaw photograph. 1995. Courtesy of the Desert Laboratory Photograph Archive.

31

Summary
Many photographic sets picturing the upper
San Pedro and upper Santa Cruz Rivers reveal a
surprisingly uniform sequence of changes. Before
the turn of the century, these valleys were largely
free of trees and shrubs, covered instead by
grasses, such as sacaton and tobosa. Prior to the
floods of August 1890, the Santa Cruz River above
present-day Green Valley supported much grass
and some forests.[17] Several early photographs
of these valley bottom grasslands show them to
be free of stumps or other woody plant remnants
that would be present had the absence of forests
been the result of recent wood cutting. As shown
by several photographic sets, by mid-twentieth
century, these channels were densely covered by
forests of mesquites, willows, and cottonwoods
and in general, remain densely covered today. Yet,
journalists and even some scientists have incorrectly reported the opposite trend.
The perception that the decline of this forest
extends upstream from the Colorado River
mainstem to its tributaries in southeastern Arizona is boosted by wildly inaccurate reports. For
example, one can read in an account about the
San Pedro River, that “steamboats once navigated
all the way to Charleston to supply Tombstone
and Bisbee with goods.”[18] In another, one reads
(concerning steamboat navigation): “When
Anglo Americans first came to the Southwest,
much of the Gila River was navigable.”[19] These
exaggerated claims overstate the volume of historic flow in these rivers. Undeniably, dam construction, water diversion, groundwater pumping, and other forces have altered the hydrologic
regimes of many of the region’s rivers, resulting
in dramatic losses of fish species and other aquatic
animals. However, along rivers that are still unobstructed by dams, these dramatic declines have
not occurred. In fact, a dramatic forest expansion
has taken place over the past century.
The sequence of changes along our major valleys appears to have started when sluggish flow
across flat, seasonally flooded swards of sacaton
became forcefully erosive during the last decades
of the 19th century. Rivers cut narrow defiles
through the grassy bottomlands. Downcutting
produced a drop in the water table and an end to

32

seasonal flooding of the grassy flats. The earlier
alternately dry and waterlogged conditions on
these low-lying heavy soils promoted the growth
of grasses and the exclusion of woody plants.
With the drop in the water table, conditions for
woody plants improved, and mesquite and other
shrubs were able to occupy the new habitat. In
addition, the floor of the newly deepened channel provided stretches of open mineral soil ideal
for the establishment of cottonwood and willow.
With subsequent widening, the bands of riparian trees broadened and became multi-aged, gallery forests. Thus, today we have cottonwoodwillow forests flanked by mesquite bosques where
a century ago the valleys supported grassy expanses of sacaton and tobosa grass.
Accompanying the changes from grassland to
forest have been dramatic changes in birdlife. The
lost grasslands were prime habitat for many sparrows including Baird’s and Botteri’s, both of which
are now “species of concern.” As noted in the
book, Birds of Arizona, “Until about 1878 [Baird’s
sparrow was] an abundant transient and doubtless winter resident in the grasslands of southeastern Arizona . . . ; until 1920 decidedly uncommon but still a winter resident about the bases
of the Chiricahua and Huachuca Mountains. Now
apparently much rarer . . . .” In Birds of Arizona,
one can also find that Botteri’s sparrow is a “rather
uncommon summer resident, . . . Usually in giant sacaton grass . . . Formerly much more common, especially before 1895 . . . . “
The changes in birdlife brought about by
growth of the new forests of cottonwood and
willow are probably even more dramatic. According to recent estimates for the San Pedro River,
the new riparian forest is home to a new biota
that includes half the birds known throughout
the United States and more mammals than occur “in any comparable area on the planet, save
for tropical cloud forest in Costa Rica.” Among
the birds in this new gallery forest is the rare
southwestern willow flycatcher, a “listed” species
and the yellow-billed cuckoo, whose numbers are
said to be dwindling (See The Arizona Daily Star,
August 2, 2000, “2nd Suit Filed over Cuckoo”).
The causes for the dramatic shift in the valley’s

biodiversity is probably the result of both climate
and land use. The downcutting on the Santa Cruz,
in August 1890, followed heavy month-long rains
falling across a region that had been severely denuded by livestock grazing. Although we may
never completely separate the impacts of livestock
grazing from those of climate, it is clear that the
change from grassland to forest came at a time
when livestock grazing was heavy. These animals
eat seedlings of the riparian trees, yet sufficient
seedlings were left behind to establish the grand
forests that are now the object of much well deserved interest. Furthermore, if livestock grazing
on the uplands was indeed a dominant factor in
causing downcutting along the region’s valleys,
we have the livestock to thank for the cottonwood-willow forest we are protecting today.
The mesquite forests flanking the narrow cottonwood-willow forests have much the same history as their neighbors. Present in a few places
when early travellers passed through our region,
they quickly expanded following the
downcutting of the late 19th century. Some of
these have disappeared in recent years. For example, the dense mesquite forest on the Santa
Cruz near San Xavier Mission slowly died as
groundwater pumping depleted the aquifer beneath it.
In many ways the Santa Cruz River presents a
special case when evaluating vegetation change.
The impact of man is especially strong along the
reach between Nogales and Tucson’s north side.
Sewage effluent from the treatment plant at
Nogales flows aboveground as it drifts northward.
Although underground by the time it reaches
Pima County, this artificial contribution probably
has an impact on the hydrology of part of the
Santa Cruz Valley. In addition, early in the 20th
century, a dike was built to direct flow diagonally
from the natural channel on the valley’s west side
across the valley to a lesser channel. Furthermore,
part of this valley segment just south of San Xavier
Mission had been used for irrigated agriculture
for centuries, which by itself greatly altered the
vegetation composition. Another man-induced
change arose when the Santa Cruz captured a
deeply eroded wagon road, abandoning its natu-

ral course for several miles along the valley. In
recent years, heavy groundwater pumping has
had a pronounced influence on vegetation growing along some parts of the valley floor. And over
its course past Tucson, the banks of the Santa Cruz
are shored up by cement walls that control both
flooding and growth of riparian vegetation.
Whether our forests and creosotebush stands
are shrinking or expanding, whether bird and fish
populations are declining or irrupting are all judgments that must be made in the context of appropriate time intervals. The starting point of an
ongoing biological or hydrological process often
must be arbitrarily assigned. Quite different conclusions can be reached depending on when the
biological stopwatch is started. For example, if
pre-downcutting conditions along our river valleys are considered the starting point from which
to judge change, then we might begin with grassland and end with forest today. By beginning at
a post-downcutting time, we might begin with
forest and then end with forest of lesser quality
today. More forest would be one conclusion; less
forest the other.
As we have seen in various photographic pairs,
changes can be rapid or slow and may involve
increases in some species and decreases in others. Location and areal extent are also important
qualifiers when describing change. As noted
above, changes over the past century in the cottonwood-willow forests are in opposite directions
on the lower Colorado River compared to the
headwaters of its undammed tributaries. And
along the Santa Cruz River, where human impacts
have been especially severe during recent decades,
anthropogenic forces have reversed the trends
seen along neighboring rivers of the region.
Photographic comparisons may be relatively
accurate measures of change but nonetheless
leave open the question of causation. The technique of repeat photography leaves little doubt
about what change has occurred, but often fails
to answer the question of just how the change
occurred.
[17] Cooke and Reeves (1976), Betancourt (1990), Betancourt
and Turner (1985).
[18] Steinhart (1994).
[19] McNamee (1994).

33

Acknowledgements
Several people have given support to this
project by providing archival access, by helping to locate old camera stations and to repeat
old photographs taken from those stations, by
supplying computer graphics assistance, and
by giving advice and direction when most
needed. For this backing, we especially thank
Chris Baisan, Janice Bowers, Art Brandt, Ed
Carter, Calvin Farris, Julia Fonseca, Peter
Griffiths, Dominic Oldershaw, Nathan Sayre,
Jeanne Turner, Martin Turner, Seth Turner,
Robert Webb, Brian White, and Roger Wolf.

Literature Cited
Baisan, C. H., and T. W. Swetnam. 1990. Fire
history on a desert mountain range: Rincon
Mountain Wilderness, Arizona, U.S.A. Canadian Journal of Forest Research 20:1559-1569.
Betancourt, J. L. 1990. Tucson’s Santa Cruz River
and the Arroyo Legacy. Ph.D. Dissertation,
University of Arizona, Tucson.
Betancourt, J. L., and R. M. Turner. 1988. Historic
arroyo-cutting and subsequent channel
changes at the Congress Street crossing, Santa
Cruz River, Tucson, Arizona. Pages 1353-1371,
in E. E. Whitehead, C. F. Hutchinson, B. N.
Timmermann, and R. G. Varady (eds.), Arid
Lands: Today and Tomorrow. Westview Press:
Boulder, Colorado.
Bowers, J. E., and R. M. Turner. 2001. Dieback and
episodic mortality of Cercidium microphyllum
(foothill palo verde), a dominant Sonoran
Desert Tree. Journal of the Torrey Botanical
Society 128(2):128-140.
Cooke, R. U., and R. W. Reeves. 1976. Arroyos
and Environmental Change in the American
Southwest. Clarendon Press: Oxford.
Glinski, R. L., and D. E. Brown, 1982, Mesquite
(Prosopis juliflora) response to severe freezing
in southeastern Arizona. Journal of the Arizona-Nevada Academy of Science 17:15-18.
Griffiths, D. 1904. Range Investigations in Ari-

34

zona. USDA, Bureau of Plant Industry Bulletin no. 67.
Griffiths, D. 1910. A Protected Stock Range in
Arizona. USDA, Bureau of Plant Industry Bulletin no. 177.
Griffiths, D. 1912. The Grama Grasses,
Smithsonian Institution, U. S. National Museum, Vol. 14, Part 3.
McNamee, G. 1994. Gila: The Life and Death of
An American River. Orion Books:New York.
Parker, J. T. C. 1995. Channel Change on the Santa
Cruz River, Pima County, Arizona, 1936-1986. U.S.
Geological Survey Water-Supply Paper 2429.
Rogers, G. F., H. E. Malde, and R. M. Turner. 1984.
Bibliography of Repeat Photography for Evaluating Landscape Change. University of Utah
Press: Salt Lake City.
Shantz, H. L., and B. L. Turner. 1958. Photographic
Documentation of Vegetational Changes in Africa over a Third of a Century. The University of
Arizona, College of Agriculture Report 16 9.
Shreve, F. 1911. Establishment behavior of the
palo verde. Plant World 14:289-296.
Shreve, F. 1924/25. An unusually arid season in
southern Arizona, p. 164-165 in Carnegie
Yearbook, Vol. 24. Carnegie Institution of
Washington, Washington, D.C.
Steinhart, P. 1994. Two Eagles/Dos Aguilas: The
Natural World of the United States-Mexico
Borderlands. University of California Press:
Berkeley.
Swetnam, T. W., and J. L. Betancourt. 1998. Mesoscale disturbance and ecological response to
decadal climatic variability in the American
Southwest. Journal of Climate 11:3128-3147.
Turner, R. M., R. H. Webb, and J. E. Bowers. 2003.
The Changing Mile Revisited, University of
Arizona Press: Tucson.

Ray Turner at work.

R. H. Webb photograph.

Raymond M. Turner
A westerner by birth, Ray Turner attended
Utah State University, the University of Utah
(B.S.1948: botany) and Washington State University (Ph.D.1954: botany). He taught at the University of Arizona (1954-1962) before joining the
U. S. Geological Survey.
Ray Turner’s interest in desert vegetation dynamics has resulted in studies of long-term permanent vegetation study plots as well as in his
use of repeat photography for documenting landscape change. His interest in repeat photography
was aroused during preparation, with Rod
Hastings, of The Changing Mile (University of
Arizona Press). Turner is also author of publica-

tions describing changes in riparian vegetation
along the Gila and Colorado Rivers, and changes
in permanent vegetation study plots at the Desert
Laboratory, Tucson, Arizona, and in MacDougal
Crater, Pinacate Preserve, Sonora, Mexico.
Retired since 1989, he has subsequently coauthored two books, Sonoran Desert Plants: An
Ecological Atlas and Kenya’s Changing Landscape
(both published by the University of Arizona
Press). A second edition of The Changing Mile,
coauthored by Turner, is currently in press. He is
still using the camera to recapture old landscape
scenes in New Mexico’s “bootheel” country and
in Kenya.

35