The upper desert grassland of Southern Arizona; a basic ecological analysis
Item Type text; Thesis-Reproduction (electronic)
Authors Kincaid, David Reed, 1931-
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 29/09/2021 12:45:19
Link to Item http://hdl.handle.net/10150/553930 THE UPPER DESERT GRASSLAND OF SOUTHERN ARIZONA A BASIC ECOLOGICAL ANALYSIS
, . ' - by
David Reed Kincaid . ■ : U '- '4'
, - j:-: 1 : v : ' . .. ^ :r ..
" *■ - —f v ‘ - : . •
^ O" . ' . ... " ' - . :.. - '■.vv:*; . Cc : : . s .
' ,
A Thesis Submitted to the Faculty of the
DEPARTMENT OF WATERSHED MANAGEMENT
In Partial Fulfillment of the Requirements
For the Degree of
MASTER OF SCIENCE
In the Graduate College
UNIVERSITY OF ARIZONA
:« • .
. p . 1959 AMOSLfiA T/m3HTtJU8 3 0 Q/iAaadAJIO TJ1383a H33RU 3HT GJ.C JAHAa/xOI )JOO i DIdAfi A
hif.Dmvi hivcQ
is rj: -i. L /'*
'4LW 7: ,: 4 J / >, /
3ffi TCO'i
30M31D8 30 M3T2AM
^gyiloD abstC orli nl
AVia IJriJS 3 0 YTI2M3VIVTU
f e d £ 979/ 1959 n
STATEMENT BY AUTHOR
This thesis has been submitted in partial fulfillment of require ments for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for perm ission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in their judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.
SIGNED: / ^ V ^
APPROVAL BY THESIS DIRECTOR
This thesis has been approved on the date shown below:
____J f . ______4 ^ / X T ? Robert F. Wagle^ Date Assistant Professor of Watershed Management
ii ACKNOWLEDGMENTS
Appreciation is expressed to Dr. Robert F. Wagle for his assistance in the preparation of this paper. Gratitude is also extended to Dr. Robert R. Humphrey for guidance in setting up this study and for his valuable suggestions on the presentation of included data.
C;
iii TABLE OF CONTENTS
P a g e
List .of Tables...... v
List of Figures...... v
INTR O D U C TIO N ...... 1
LITER A TU R E REVIEW ...... 3
DESCRIPTION O F THE STUDY A REA...... 5
* L ocation ...... 5 C lim a te ...... 5 G eology and S o ils ...... 6
PRO CEDURE...... 8
CHARACTERISTICS O F INDIVIDUAL P L O T S...... 9
RESULTS...... 16 -
DISCUSSION...... • • ...... • • 21
Altitudinal lim its of the Oak Savanna...... 21 Distribution of Species Within the Oak Savanna...... 22 Floristic Relationship of the Oak Savanna...... 26
S U M M A R Y ...... 31
LITERATURE C ITED ...... 33
A P P E N D I X ...; ' ...... 35 LIST OF TABLES
T able P a g e
1. Certain site characteristics as related to plant densities 17
2. Major plant species in the study area and relative occur rence frequencies*...... 20
LIST OF FIGURES
Figure : : ' ^ Page
1. General view of the oak’savanna showing scattered trees with grass understory ...... 2
2. Portion of study area with Muhlenbergia emersleyi (tall, plume-like seedheads) and Andropogon cirratus co -d o m in a n t...... 10
3. Plot VI. Typical oak savanna with Andropogon cir ratus dominant. Trees consist primarily" of Quercus oblongifolia...... 13
4. Comparative rainfall and temperature from Porter-: . ville, California and Ruby, Arizona, showing similar winter conditions...... 28
5. Comparative rainfall and temperature from Hays, Kansas and Bear Valley, Arizona, showing similar summer growing conditions...... 30
v INTRODUCTION
There are almost no basic ecologic or floristic studies on the upper desert grassland of New Mexico and Arizona. Such studies are fundamental to the development of ecologic methods for the control of invading species. Analysis of vegetation changes and of the factors primarily responsible may indicate effective measures needed to con trol the woody plants that have invaded many of these ranges.
In order to make a thorough analysis of vegetation changes in the upper desert grassland, it is first necessary to know the species prevalent in the area before the arrival of the large herds of livestock before and during the 1880*8. Since early literature provides few detailed descriptions, this study necessarily began with an analysis of grassland ungrazed by domestic animals. The rare places where such situations exist today, are isolated spots physio graphic ally inaccessible to livestock.
The purpose of this study was to obtain and interpret basic data on the floristic characteristics and environment of the upper desert grassland as typified by the oak savanna of southern Arizona
(Figure 1).
1 V 14 y»': ■ :% i / jY #- ,r B atigs* ^ irM f k /A
Figure 1. General view of the oak savanna showing scattered trees with grass under s to r y . LITERATURE REVIEW
Travelers through southern Arizona about the time of the Civil
War admired the upper desert grassland for its beauty and praised it
for its productivity. J. Ross Browne (1864) wrote —
The day’s journey through the valley of N ogales... was one of the most pleasant of our trip. Every mile we travelled the country improved in beauty and fer tility. Grass up to our horses’ shoulders covered the valley, and the hills were clothed with luxuriant groves of oak. Much of the country reminded me of the Coast Range in California.
The foothill m esas east of the Santa Rita Mountains also drew special
atten tion .
It was a matter of surprise to us how luxuriant the grass was on this m esa and what an inexhaust ible support it affords for innumerable herds of cattle. ; ; : : . :
Unfortunately, too many people believed the grass was inex
haustible. The ’’innumerable herds” of 1864 continued growing until
by 1901 there were few areas of virgin vegetation left in southern
Arizona. In that year, David Griffiths (1901) of the U. S. Department
of Agriculture wrote--
On all the Western stock ranges which the writer has visited there have existed many small areas in cultivated fields, unused pastures, fenced railroad
3 4
rights-of-way and sim ilar situations which are in their virgin state or have so far recovered from overstocking as to bear testimony to the original productivity of the soil. Thingst are far, different in large areas of Southern Arizona. Here unused pastures are very rare, cultivated fields are fewer in number and the destruction is so complete that in many localities even the railroad right-of-way has recovered but little in three or four years* ? i tim e .
Although Griffiths (1901), Thornber (1909), and Shreve (1915)
described the upper desert grassland floristically, they did not dis
cuss it from a successional point of view. The area they described
as the ''upper foothills", "live-oak belt", or "encinal" were locations throughout southern Arizona and New Mexico characterized by large
grassy slopes dotted with evergreen oaks and nut pines. -.As a rule, >
these oak savanna areas occurred as belts connecting the treeless
valleys or plains with the more heavily forested mountaintops.
Shreve (1915) placed the lower lim it of the "encinal" of the
Santa Catalina mountains at 4, 300 feet and the upper at 6, 000 to 6, 500
■feet. ■ - r: v; :■ ! . v : ' -.;■
. ' - ' ' ' :::: ' ■ v r: , ;:0' - . - . :■ ;.Gv DESCRIPTION OF THE STUDY AREA
L ocation .
The study was made in Township 23 South, Ranges 11 and 12
East, in southwestern Santa Cruz County, Arizona. In this area nine plots were so located as to permit analysis of vegetation at different
elevations on a variety of exposures. Elevations of the plots ranged
from 5, 200 feet to 5, 500 feet.
C lim a te.
Climatic records were available for Nogales, Arizona, about 18
miles southeast of the study area, and for Ruby, Arizona, seven m iles
northwest of the area (Smith, 1956). Rainfall records were available
for a nine-year period at Bear Valley Ranch, about four m iles south
west of the study area. The data from these stations indicate that an
increase in elevation is accompanied by an increase in total precipita
tion. The rainfall pattern, however, remains essentially the same,
about 60 per cent falling in the summer.
Temperatures at the study area followed the same pattern as at
Nogales, but averaged about 3 to 4 1/2°F. lower. Nogales is at an
5 6 elevation of 3, 800 feet, while this study was conducted at elevations from 1, 400 to 1, 700 feet higher. Robbins (1917) has shown that in con tinental climates the average decrease in temperature is about 2 1/2°F. for every 1, 000 feet increase in elevation. This figure is not constant throughout the year, but is least in winter (1 1/2°F .), next in autumn
(2 1/2°F .), then sximmer (3°F.), and greatest in spring (3 1/2°F .).
The temperature difference brings about a retardation in the develop ment of vegetation amounting to about 13 days for every 1, 000 feet increase in altitude.
The range between day and night temperatures is large, averag ing from 30° to 40°F. yearlong. At Nogales, Arizona, 18 miles from the study area, January is the coldest month with a mean minimum of 29. 5°F.
The mean maximum for January is 61. 4°F. If sun temperatures, were considered, the range between day and night temperatures would be much
g r ea ter .
The mean annual evaporation at Nogales is 99. 92 inches. Evapor
ation at the study area would probably, correspond roughly to this figure.
...
Geolo gy and Soils.
Darton (1909) describes the geology of the central range of the
Tumacacori Mountains as being of the Tertiary volcanic succession
type. Its geology is sim ilar to that of the nearby and better known Santa
Rita Mountains of Arizona. 7
The study area immediately south of Atascosa Peak is in rolling but quite steep and very rocky terrain. Observations by Gumming (1951) indicated that the most common rocks in the vicinity are rhyolites, tuffs, andesites, felsites, and agglomerates.
Soils in oak woodlands of the Southwest are varied. Climate, not soil type, apparently is the major factor determining the vegetation.
Soils of these areas are usually of the grey-brown podzolic group at higher elevations (6, 000 feet) and Shantung-brownlike soils around 5, 000 feet (Martin and Fletcher, 1942). PROCEDURE
Nine areas were so located as to provide samples from slopes facing north, south, east, and west. In each area a 100-by-50-foot plot was laid out. Each of these was divided into five 20-by-50-foot compartments. Two 50-foot line transects were located randomly in each compartment, and the plants recorded and measured by the line intercept method (Canfield, 1942). Soil samples were taken to deter mine soil type. The field work was carried out from September 1957 through January 1958.
The collected data were then analyzed or interpreted in terms of climate, soil, exposure, and altitudinal relationships. In order to verify certain of the interpretations, comparisons were made and analogies drawn between this and other areas.
8 CHARACTERISTICS OF INDIVIDUAL PLOTS
Plot I was laid out on a steep (80%), rocky, west-facing slope near the top of a ridge at an elevation of 5, 300 feet. The plot was shel tered by ridges from north, east, and south winds (Figure 2). 1 2 The dominant grasses were Andropogon cirratus , Muhlenbergia em ersleyi, and Bouteloua curtipendula in that order. Other grasses found in quantity were Aristida orcuttiana and Eragrostis intermedia.
The dominant trees and shrubs were Quercus oblongifolia, Nolina m icro-
carpa, and Das ylirion wheeleri. The most abundant forb was Cheno-
podium watsoni.
3 In this plot the density of grasses was . 0978; that of trees and
shrubs .1148. Density of forbs was .0348. Although the density of
annual plants was measured, as it was on all plots, it has not been *
* Most abundant. In the case of grasses, this means the species with the greatest basal intercept (Canfield, 1942). For trees, shrubs, and forbs it means the species with the greatest crown intercept. The terms “principal", "most common", and "most abundant" are used syn onymously with "dominant". 2 Plant names correspond with those in Kearney, T .H ., and R. H. Peebles. 1951. Arizona Flora. Univ. of Calif. Press, Berkeley, Calif. 3 Grasses: basal density expressed as a decimal fraction of com plete ground cover. Trees, shrubs, and forbs: crown density expressed as a decimal fraction of complete ground cover.
9 Figure 2. Portion of study area with Muhlenbergia emersleyi (tall, plume-like seed- heads) and Andropogon cirratus co-dominant. o 11 presentee! as a dependable figure because of the great annual variability of these species depending on seasonal precipitation. Also, measure ments made in winter might show many species as abundant that would not be present if a similar study were carried out during the summer season, or at times even during the same season of another year.
Plot II was on the same slope as Plot I, at an elevation of 5, 200 feet with an exposure to the west-southwest. The terrain was steep (80%) and rough and received less sunlight during the morning than Plot I, due to shading from the ridge to the east. The dominant grasses were Muh- lenbergia em ersleyi, Andropogon cirratus, and Aristida orcuttiana.
The most abundant trees and shrubs were Quercus oblongifolia, Nolina microcarpa, and Garrya wrightii. Forbs consisted mainly of Cheno- podium watsoni and Artemisia ludoviciana. The grass density was . 1042; shrubs and trees . 1093; and forbs . 0332.
Plot III was on a very steep (80-90%) north-facing slope at 5, 300 feet elevation. The area received less sunlight than any other plot studied. Dominant grasses were Muhlenbergia em ersleyi, Aristida orcuttiana, and Tridens muticus. A few plants of Muhlenbergia longili-
gula were found here. The most common trees and shrubs were Quercus oblongifolia. Artemisia ludoviciana was the main forb.. Plot III had a
grass density of . 0757, shrub and tree density of . 0866, and forb density
of . 0293. 12
Plot IV was on an essentially level area on a ridge at 5, 300 feet elevation. It was protected on the eastern side by a higher ridge run ning north and south. There was a very slight slope to the west; the ground was not as rocky as on the previous plots. The dominant grass was Andropogon cirratus, followed in order of abundance by Muhlen- bergia em ersleyi, Bouteloua curtipendula, and Aristida orcuttiana.
The most abundant trees and shrubs were Quercus emoryi, Quercus oblongifolia, and Mimosa biuncifera. Basal density .of grasses was
. 1677; tree and shrub density was . 1667; and forb density was only . 0062.
Plot V was at 5, 300 feet on the same ridge as Plot IV. It was on a .slope (45%) facing southwest but was still partly protected from sun on the east by a ridge. Here, again, Andropogon cirratus was the dom- » inant grass, followed in order of abundance by Muhlenbergia em ersleyi,
Bouteloua curtipendula, and Aristida orcuttiana. Trees and shrubs were mostly Quercus oblongifolia and Quercus emoryi. The most com mon forb was Chenopodium watsoni. Grass basal density was . 1512;
density of trees and shrubs .1634; and density of forbs . OI96.
Plot VI was near the top of a ridge on a gentle slope (15%) facing
west, with an elevation of 5, 400 feet. Muhlenbergia em ersleyi and
Andropogon cirratus were dominant grasses (Figure 3), followed by
Aristida orcuttiana and Andropogon hirtiflorus. Quercus oblongifolia
and Quercus emoryi were the dominant trees. The most abundant forb Figure 3. Plot VI. Typical oak savanna with Andropogon cirratus dominant. T r c consist prim arily of Quercus oblongifolia. 14 was Artemisia ludoviciana. Grass density was . 1362. Shrubs and trees ' comprised . 1035 of the ground cover. Forb density was . 0179.
Plot VII was laid out on a steep (70%) rocky slope with a southeast exposure at 5, 400 feet elevation. The area was not protected from sun or wind on any side. Muhlenbergia em ersleyi was the most abundant grass species, followed in order of abundance by Bouteloua curtipendula,
Andropogon cirratus, and Bouteloua hirsuta. The dominant trees and shrubs were Quercus oblongifolia and Mimosa biuncifera. The most abundant forb was Chenopodium watsoni. Density of grasses was . 0983; that of shrubs and trees . 0472, and forbs . 0058.
Plot VIII was at 5, 300 feet elevation with an east exposure, also on a steep (70%), rocky slope. Muhlenbergia em ersleyi was the domin ant grass, followed closely by Andropogon cirratus and Bouteloua curti pendula. Dominant trees and shrubs were Quercus oblongifolia and
Dasylirion wheeleri. There was only a trace of Viguiera annua and
Chenopodium watsoni. Grass density was .1171; that of trees and shrubs
.0 9 1 3 .
Plot IX was the highest plot in the series at 5, 500 feet elevation.
This plot faced directly south and received no shading or protection from
wind. Basal grass density was . 0836, the most abundant grass by far
being Bouteloua curtipendula.- The next two in abundance, Muhlenbergia
em ersleyi and Andropogon barbinodis together did not equal the amount 15 of Boutejoua curtipendula. Dominant trees and shrubs were Quercus oblongifolia and Dasylirion wheeleri. Large amounts of two annual forbs, Viguiera annua and Chenopodium watsoni, were present. Crown density of trees and shrubs was . 0780; that of forbs . 0467. .. ' RESULTS
Taken as a whole, the plants along the entire 4, 500 feet of line transect provide a good cross section of relict vegetation in the oak savanna. Basal density of all perennial grasses was . 1147. Four species, Andropogon cirratus, Muhlenbergia em ersleyi, Bouteloua curtipendula, and Aristida orcuttiana, comprised slightly more than
75 per cent of this density. Density of trees and shrubs was . 1054, more than half of which was,provided by the oaks Quercus oblongifolia and Q. emoryi. Density of forbs was . 0213) the most abundant species being Chenopodium watsoni and Artemisia ludoviciana (Appendix Tables
A, B, and C). ' '
Although the study was made to obtain a picture of the type as a whole, certain inter plot differences resulting from site characteristics can be noted in Table 1.
The highest grass density occurred on the most nearly level areas,
Plots IV, V, and VI. In addition to being more level, these plots were also less rocky than the other plots. Although soil erosion was not noticeable on any of the plots, runoff following summer-storms on the
16 17
Table 1. Certain site characteristics as related to plant densities.
! 1 1 T r e e s and E xp o- P e r ,G r a sse s 1 F orb s * Cent 1 Shrubs P lo t su re 1 Slope 1 b a sa l 1 crow n crow n 1 dens it v 1 d en sitv t d en sity 1 t t ! t I ' W . V 80 ' .0 9 7 8 ' .1 1 4 8 .0 3 4 8 1 1 1 t I II ' WSW ' 80 ' .1 0 4 2 ' .1 0 9 3 .0 3 3 2 t 1 i 1 III ' N ' 85 ' .0 7 5 7 ' .0 8 6 6 1 . .0 2 9 3 1 I IV ' . yr ' t.1677 ' .1 6 6 7 I .0062 0 1 I 1 V ; s w ' 45 ’ -1 5 1 2 * .1 6 3 4 .0196 | VI ' w '. 15 • .1 3 6 2 ’ . .1 0 3 5 t .0 1 7 9 t | i 1 . VII ' . SE : V i 70 ' .0 9 8 3 ' .0472,, .0 0 5 8 1 t f
VIII ' 1 ; E 70 ' .1 1 7 1 .0 9 1 3 . T f ♦ • " • ■ - ; 1 IX ' s ' 75 ' .0 8 3 6 ' .0 7 8 0 .0 4 6 7 1 i i 1 I 1 1 1 * :: . -• - '* t 1 AVG. 1 .1 1 4 7 * .1 0 5 4 .0 2 1 3 i ? 1
. ‘ . ; / ‘ 18 steeper slopes may have resulted in less available moisture and, con sequently, a lower density of grasses. There was, a general decrease in density of forbs accompanying increased grass density.
No significant changes in density of shortgrasses could be corre lated with decreases in density of mid grasses on the study area. In seven of the nine plots, when the midgrass density was below the nine- plot average, the shortgrass density was above. Except on Plot III, an inverse relationship between the density of Muhlenbergia em ersleyi and
Bouteloua curtipendula was noted. However, the effect of less favorable environmental conditions that resulted in low mid grass densities seemed to result in reduced grass densities over-all, rather than a marked increase in shortgrasses, making these inverse relationships rather •
vagu e. ; - i , :: r.-'.'r-':' : .
There was no inverse relationship between the density,of trees and shrubs as opposed to grasses. 'When the density of mid grasses was plotted against the density of oaks, a reduction in oaks corresponded to a reduced density of m idgrasses. These results indicate that even with in a very narrow altitudinal range, the generally better environmental conditions that favor the growth of oaks also increase the abundance of midgrass. If the oak stand is moderately heavy, and particularly if
Quercus emoryi forms an appreciable part of the oak population, the area should be capable of supporting a good stand of m idgrasses.
\ 19
Of all the perennial grass species encountered on one or more of the nine plots, Muhlenbergia em ersleyi and Andropogon cirratus were the most widely adapted to the variety of sites represented (Table 2).
Although Andropogon cirratus had a greater total density than the
Muhlenbergia, it did not occur as consistently, nor maintain a high den sity as consistently.
Aristida orcuttiana, although having a lower density than Bouteloua curtipendula, was more generally distributed throughout the study area.
Quercus oblongifolia was distributed throughout the study area. Other
. • v ' : species of trees and shrubs were less widely distributed. Quercus :
emoryi tended to be restricted to the more mesophytic areas at this alti- tude (north-facing slopes, more level areas) and was accompanied by the highest midgrass densities and a relatively higher proportion of the forb : C v - . - ' 5 ’ ' ' Artemisia ludoviciana as compared with Chenopodium watsoni.
V J
u 20
Table 2. Major plant species in the study area and relative occurrence frequencies.
Vegetation , Dominance Priority Rating* L ife form 1 S p ec ie s 1 1st 1 2nd ' 3rd ' 4th ' 5th ' 6th** ! f 1 1 i i 1 1 Andropogon barbinodis 1 ' 0 1 0 ' 1 ' 0 ' 0 ' 8 1 Andropogon cirratus 1 3 1 3 ' 1 ' 2 0 ' 0 1 Andropogon hirtiflorus 1 0 1 0 ' 0 ' 1 ' 0 ' 8 ' Aristida orcuttiana 0 t 1 . ' 2 ' 1 ' .4 ' 1 Bouteloua curtipendula 1 1 1 1 ' ’ 4 ' 0 ' 0 ' 3 GRASSES 1 Bouteloua hirsuta 1 0 I 0 ' 0 ' 2 1 0 ' 7 1 Eragrostis intermedia 1 0 1 0 ' 0 ' 3 ' 0 1 6 1 Heteropogon contortus 1 o 1. 0 ' 0 ' 0 ' 1 ' 8 1 Lycurus phleoides 1 0 1 0 ' 0 ' 0 ' 2 ' 7 ' Muhlenbergia emersleyL' 5 1 4 ' 0 ' 0 ' 0 V .o 1 Muhlenbergia longiligxla1 b ::| 0 ' 0 ' 0 ' 1 ' 8 1 1 Tridens muticus . 1 0 0 ' 1 ' o ' 1 ' 7 I , 1 Baccharis pteronioides 1 0 0 ' 1 ' 1 ' 1 ' 6 1 Dasylirion wheeleri ' 0 t 2 ' 2 ' 0 1 2 ' 3 ' D ale a sp . ... .' 0 t 0. • ' 0 . 4 I 0 ' 5 TREES ' Qarrya wrightii 1 0 1 0 ' 1 ' 1 ' 0 ' 7 ^ 1 Mimosa biuncifera 1 0 1 1 ' 3 ' 2 ' 0 ' 3 SHRUBS i Nolina microcarpa 1 0 1 3 r 0 ' 0 ' :i 0 ' 6 1 Quercus emoryi 1 i ! 2 ' 1 ’ o ' 0 ' 5 1 Quercus obiongifolia 1 8 1 1 » 0 ' O ' 0 ' 0 1 1 Yucca schottii 1 . 0 0 ' 0 .' 1 ' 2 ' 6 . ' ■■ r 1 1 1 i 1 1 1 , t ArtemisiaTudoviciana J 3 3 0 3 ! 0 0 • •• Chenopodium watsoni 4 3 ' 0 ' 0 0 FORBS ( Penstemon sp. ( 0 1 V o 1 o ' o V . 8 ( Sisymbrium sp. 0 0 ' i 1 o ' 0 ' 8 . ■ ■ V ig u iera annua ... ■ 2 ’ 2 ' 0 ’ 0 ' 4 ;v 1 . i 1 1 1
* Numbers following each species indicate the number of plots on which this species was rated as occurring in the relative dominance position indicated. * * ' '' • ' ** Plots placed in this column include; those not having-the.named species.
Bouteloua hirsuta, Lycurus phleoides and Tridens muticus are classed as shortgrasses. The other grasses in this table are midgrasses. DISCUSSION
Altitudinal Limits of the Oak Savanna.
The oak savanna in southern Arizona occurs primarily between elevations of 4, 000 and 6, 500 fe e t. S h reve (1915) believed that the upper altitudinal lim its of the oak savanna were established by low;winter tem peratures. Pearson (1920), on the basis of knowledge regarding the natural occurrence, of various tree species in the Fort Valley, Arizona, area, concluded that the upper altitudinal range of all trees is deter mined by low temperature. Whether low temperature or competition from forest trees is actually more important in establishing the upper limit of the oak savanna has not been determined. The fundamental cause in either case is climatic, since it is the increased precipitation at higher altitudes that enables the forest species to survive. In addition to altitude, factors such as slope, exposure, and cold-air drainage greatly influence the distribution of oaks by extending or decreasing the vertical lim its of favorable growing conditions.
The lower lim it of the oak savanna, on the other hand, apparently is determined by moisture deficiency. According to Pearson (idem. ), the distribution of a species along its lower elevational limits is governed
21 22 not by the higher temperatures there, but by the lower available m ois ture. The decrease in available moisture at lower elevations is com pounded from several factors. In the Southwest, as in most areas, a decrease in elevation is accompanied by a decrease in precipitation and an increase in evaporation because of higher temperatures. This increase in temperature may further intensify moisture deficiency by increasing transpiration in some species. It has been determined that continental climate temperatures increase with decreasing altitude at a rate of about 2 1/2°F. per 1, 000 feet (Robbins, 1917). The effect of temperature differences of this magnitude on transpiration rate has not
been determined. It would seem reasonable to assume, however, that
the physiological stress on a given tree species at the lower elevations
of its range is greater than at higher elevations. V
Exposure and topography exert microclimatic influences on plant
distribution by modifying soil-m oisture and air-moisture conditions.
Oak trees, for example, particularly Quercus emoryi, may often be
found growing near intermittent streams and in drainages well-below
their usual altitudinal lim its.
Distribution of Species Within the Oak Savanna.
Although the oak savanna typically occurs as a belt between the
desert and forest, and inter grades with plants characteristic of the 23 desert at its lower limit and with forest species at its upper limit, it ' contains many species that are characteristic of the oak savanna alone and are not found above or below it. This does not imply that these species are uniformly distributed or even occur throughout the altitudinal range of the type. Within the fairly narrow lim its of 4, 000 and 6, 500 feet elevation, where the most typical oak savanna occurs, plant distri bution is controlled in large measure by small environmental differences that result from insolation, topography, exposure, and soil. As altitude and exposure both affect temperature, conditions prevailing on north facing slopes at 4, 000 feet seem to be characteristic of east-facing and west-facing slopes at about 5, 000 feet and of south-facing slopes at higher elevations.
Quercus oblongifolia is typical of the lower elevations and drier sites and rarely grows above an altitude of 5, 500 feet. Juniperus d&ppeana grades upward from the Quercus oblongifolia sites into areas containing much Quercus emoryi and Pinus cembroides, species seldom found below
5, 000 feet, except in drainages or other areas where moisture is more plentiful. Thick stands of Fouquieria splendens, a desert shrub, occur on rocky, tight soils on south-facing, droughty slopes as high as 5, 000 feet.
: ; Rainfall is the critical factor in initiating summer growth in the oak savanna. By the time summer rains start in July, temperatures have long been favorable for plant growth throughout the oak savanna. 24
The grasses in this area rely on summer rainfall for the bulk of their growth. One would expect to find midgrasses at higher elevations where rainfall is more plentiful and the drought-resistant shortgrasses at lower elevations where rainfall is less. This is generally true if at any given altitude the modifying effect of exposure is considered. As moisture is available for a longer period following summer rains on north-facing slopes, m esic plants are more typical of these slopes than of other exposures;
Bouteloua curtipendula, a typical desert grassland drought- resistant midgrass, demonstrated a definite distributional response to this "exposure factor1' as related to an increase in altitude. Gumming
(1951) noted that Bouteloua curtipendula was dominant on north-facing slopes in the bajada zone (about. 3, 800 feet elevation), comprising more than 50 per cent of the grasses. It also was the most.abundant species in the north-facing slopes in the mountain-foothill zone at 4, 000 fe e t e le vation. Here, however, three of the four areas he studied supported less than 50 per cent Bouteloua curtipendula, which tended to be replaced by other m idgrasses. ! Above 5, 000 feet, where the present study was con ducted, Bouteloua curtipendula was more abundant on the drier, south facing slopes, a.hd was replaced on m oister sites by Muhlenbergia em ersleyi, Andropogon cirratus, and Aristida orcuttiana. Both of these .
studies point to the fact that Bouteloua curtipendula is more drought tol
erant than the other midgrasses with which it was associated. 25
■ Only three of the twelve most common perennial grass species recorded on the nine plots are classed as short grasses. It is of interest to note that none of them occurred in the top two dominance classes.
Only one, Tridens muticus, fell in the third class, and all were rated seven times out of nine in the lowest priority class. Insofar as this study site represented an .undisturbed oak savanna area,: m idgrasses are strong dominants. It is recognized that most midgrasses are coarser and less palatable to cattle than associated shortgrasses. Although usually less heavily grazed, they require more me sic conditions for survival. When plants are near the limits of their genetically established tolerance ranges, their distribution is controlled by what may appear to be very minor differences in environmental conditions. In the oak savanna, the .. - r. . '. ■ • v : .. . . ■ - midgrasses are sensitive to rather slight changes in the physical environ ment in terms of available moisture. McGinnies, Parker, and Glenden- ing (1941) stated that the side-effects of grazing, i. e ., increased light intensity at the soil level and therefore higher soil temperature and evaporation, decreased humidity near the ground, and increased runoff because of lower plant density probably have more to do with changing the plant cover from midgrass to short grass than the immediate effects of grazing.
In the upper desert grassland, environmental changes accompany ing an increase in elevation often increase the proportion of midgrass to 26
shortgrass (Gumming, 1951). As areas topographically inaccessible to cattle are higher than surrounding accessible grassland, a higher pro portion of midgrasses to shortgrasses on some relict areas might be more correctly ascribed to increased altitude than to the absence of
grazin g.
Regardless of the specific causes for midgrass or shortgrass -
dominance, the present study does show conclusively that ungrazed oak
savanna, growing under climatic conditions favorable to growth of these
oaks, contains a minimum of shortgrasses and is strongly dominated by
the larger m idgrasses.
Floristic Relationship of the Oak Savanna.
The oak savanna in general appears to be an area of scattered
woody plants whose origin and major distribution are in the southwestern
United States and northern Mexico, with an understory of grasses and
forbs, many of which are characteristic of the great plains.
Shreve (1915) observed that the oak savanna floristic relationship
is almost wholly confined to the foothills of the northern Mexican states
of Sonora, Chihuahua, Sinaloa, and Zacatecas. The oak savanna of
southern Arizona is sim ilar to that existing throughout the mountainous
portions of these Mexican states. Very few of its plants are found north
of the . Mo go lion Rim. The common woody or succulent perennials have
their major distribution in northern Mexico, southern Arizona, and New
Mexico. Most of them occur in west Texas. 27
The oak savanna contains several plants that reach their maximum distribution on the great plains. These are principally summer-growing grasses and forbs, including Bouteloua curtipendula, Bouteloua hirsuta,
Sitanion hystrix, Artemisia ludoviciana, and Chenopodium watsoni.
Bouteloua gracilis, which did not occur on the study plots but is-common in the oak savanna east of the Santa Rita Mountains, also is characteris- tic, of the great plains. : -<
Few of the species have ;their major distribution in the mountains , ...... " i ;A;! ' of Colorado and Utah; almost none are common in California.;
■■■■• . . > > ■ Precipitation in the oak savanna falls during two distinct seasons-- winter and summer. The winter rains are similar in amount and distri bution to those in the California oak savanna regions (Figure 4).
• Although the species are dissim ilar, the general aspect of the
California oak sa.vanna strongly resem bles that of the Arizona oak savanna. 0 However, the grass understory in California is composed almost entirely of annual grasses. These grasses begin to grow early in the spring when temperatures become high enough, and end their life-cycle by the time the water that fell in the winter has been exhausted. The few perennial grasses that remain in relict areas and areas long protected from grazing are cool-season grasses that are dormant during the summer.
The grasses in the Arizona oak savanna, on the other hand, are almost exclusively warm-season perennials, making the bulk of their 28
PORTERVILLE, CAL.-
RUBY, ARIZ. MEAN DEGREES MEAN TEMPERATURE-
Figure 4. Comparative rainfall and temperature from Porter ville, California and Ruby, Arizona showing sim ilar winter conditions. 29 growth during July, August, and September. July-October rainfall and temperature distribution closely approximate conditions at the western edge of the midgrass area of the’ great plains .(Figure 5). Many species of grasses and forbs are common to both areas'.
. v
: r: 0
i ;go o ' t
: o o V:
o > 1 ; : i
- > ',1, i‘ /. vV
V.'; c TV " . - ; ' %'
Univ. of Arizona Library 30
HAYS,KANS. BEAR VALLEY, ARIZ.
90 TEMPERATURE Ik 80
70
60
50 RAINFALL 40
30
20
10 MEAN - DE6REES MEAN TEMPERATURE
MONTHS
Figure 5. Comparative rainfall and temperature from Hays, Kansas and Bear Valley, Arizona showing similar summer growing conditions. SUMMARY
A study was made of the floristic characteristics of a relict area in the upper desert grassland oak savanna type near Ruby, Arizona.
The plant cover was analyzed by the line-transect method, using a total of ninety 50-foot line-transects.in nine separate locations. Twenty
species of grass were encountered. The density of perennial grass was
. 1147. Over 75 per cent of the grass consisted of four species -- Andro- pogon cirratus, Muhlenbergia em ersleyi, Bouteloua curtipendula, and
Artis tida orcuttiana.
Density of trees and shrubs was . 1054, almost half of which was
Quercus oblongifolia. Density of forbs was . 0213, the most abundant
species being Chenopodium watsoni and Artemisia ludoviciana.
The oak savanna in southern Arizona is restricted to between the
elevations of 4, 000 and 6, 500 feet. Within this altitudinal range, the
factors of slope, exposure, and cold-air-drainage control, the distribu
tion of oaks by extending or decreasing the lim its of climatic conditions
in which they can grow.
Although the oak savanna typically occurs as a belt between the
desert grassland and forest, and intergrades with plants typical of the
31 32 desert at its lower limit and with forest species at its upper limit, many of the species are found only within the oak savanna. The distribution of each species is determined by the environmental factors of altitude, topo graphy, exposure, and soil. Soil effects in the oak savanna appear to be secondary to climatic effects.
The proportion of midgrasses to shortgrasses is often higher on relict areas than on nearby grazed portions of the oak savanna. However, grazing by domestic livestock does not appear to be the sole cause of a reduction in m idgrasses. Changes in physical environment caused by climatic fluctuations and an indirect effect of grazing on m icroclimate appear to be more important in effecting these grass species composition changes than the direct effects of grazing. LITERATURE CITED
Browne, J. Ross. 1950. A tour through Arizona, 1864. Arizona Silhouettes, Tucson, Arizona. . ; „
Canfield, R. H. 1942. Sampling ranges by the line intercept method. Research Report No. 4, Southwestern Forest and Range Exper iment Station, Tucson, Arizona. , .
Gumming, K. J. 1951. The effect of slope and exposure on range vegetation in the desert grassland and oak savanna areas of Santa Cruz County, Arizona. M.S. Thesis, University of Arizona, Tucson.
Darton, N. H. 1909. A resume of Arizona geology. University of Arizona Bulletin 119, Tucson.
Griffiths, D. 1901. Range improvement in Arizona. USDA Bureau of Plant Industry Bui. 4. W ash., D. C.
______. 1904. Range investigations in Arizona. USDA Bureau of Plant Industry Bui. 67. W ash., D. C.
Martin, W. P. and J. E. Fletcher. 1942. Vertical zonation and great soil groups on Mt. Graham, Arizona, as correlated with climate, vegetation, and physical characteristics. Ariz. Agr. Exp. Sta. Tech. Bui. 99, Tucson.
McGinnies, W. G. , K. W. Parker, and G. E. Glendening. 1941. Southwestern range ecology. Southwestern Forest and Range Exp. Sta. , Tucson.
Pearson, G. A. 1920. . Factors controlling the distribution of forest types. Part II. Ecology, 1:289-308.
Robbins, W. W. 1917. Native vegetation and climate of Colorado. Colo. Agr. Exp. Sta. Bui. 224, Fort Collins.
33 34
Shreve, F. 1915. The vegetation of a desert mountain range as con ditioned by climatic factors. Carnegie Inst. Wash. Publ. 217, Wash. , D. C.
Smith, H. V. 1956. The climate of Arizona. Ariz. Agr. Exp. Sta. Bui. 279, Tucson.
Thornber, J. J. 1909. Range conditions in Arizona 1900-1909. (As recorded by various observers in a series of miscellaneous papers.) Univ. of Ariz. Library, Tucson.
United States Department of Agriculture. 1941. Climate and Man, USD A Yearbook 1941, W ash., D. C. t
APPENDIX, .
< • , ‘ "j ‘ ; w « . - • 3 r',.. > •* • ; , 4 i y ; 1 { ) . " / . V: . : . . ; # 4 ;
j . : i i . ■ t • ; • V-' ■. ■ v , : J » 7 0 f . \ ; . s v ?
f , f 1 : \ • - . : 3 u :
;■ :>
• ..it
35 ) 36
Appendix Table A. Perennial Grasses (in order of abundance).
i S p ecies .F req u en cy Line Intercept ■1 N o • F e e t ' P e r c e n t1
Andropogon cirratus ' 534 138. 62 i 26. 86 Muhlenbergia em ersleyi i 527 137. 12 ; . 26. 57 Bouteloua curtipendula ' 272 6 1 .3 7 ' iiV 89;; Aristida orcuttiana ' 252 , 52. 75 • 10.22: Eragrostis intermedia < 147 29.04 « 5.63... Bouteloua hirsuta i 100 16.58 ' 3.21 : Andropogon barbinodis , 4o 12. 54,' - 2.43 Tridens muticus i 49 12.33 ; ' 2.39 Lycurus phleoides ' 55 12.08 ' 2.34 Andropogon hirtiflorus I 41 f 10.42 - 2. 02 Heteropogon contortus • 50 9.79 ' 1. 90 Elyonurus barbiculmis • 31 8.79 ' 1.70 Muhlenbergia longiligula ' 9 ’ 3.83 i .74 Trachypogon montufari • 10 3. 04 . •5 9 Leptochloa dubia • 13 2.79 ' •54: Panicum bulbosum • id 1.57 ' .30 Muhlenbergia monticola ' 5 1. 33 - . 2 6 : Muhlenbergia rigens j 3 ’ 1.25 ' .24 Bromus anomalus ' 6 .5 0 > ...... 0 9 Sitanion hystrix ' 2 ; .2 5 : , : .05
T otal f 2,156 515. 99 2 ' 99.97
Based on analysis of 90 50-foot line transects. “
The basal density of perennial grasses was 515. 99/4, 500 or . 1147. 37
Appendix Table B. Trees and Shrubs (in order of abundance).
. S p ecies F req u en cy " Line Intercept ' N o. P e r cent^ Quercus oblongifolia i 36 , 218.42 46.04 Quercus emoryi . 14 t 1 56.33 i 11.87 Dasylirion wheeleri i 31 i 46.50 i 9.80 Mimosa biuncifera , 37 i 40.92 1 i 8.62 Nolina microcarpa i 21 , 31.08 i 6. 55
►—1 !—• Gutter re zia sarothrae . 20 -4 3.67 Garrya wrightii i 7 . 15.83 « , , 3.34 Dalea wislizeni. i 15 i 7.87 1 1. 66 Yucca schottii . 4 , . 7.50 , 1.58 Baccharis pteronioides . 6 . ' 7.42 , 1. 56 Pinus cembroides , 2 . 5.92 . 1.25 Rhus choriophylla t 6 . ' 5.58 . 1.18 Opuntia engelmannii , 2...... , 4:83 , ' " 1.02 Mimosa grahami . , 8 •:i 4.75 i , , 1.00 Juniperus deppeana . 1 , 2.50 . . 53 Agave palmeri , 1 .8 3 , . 17 Bouvardia glaberrima ' 2 i . 75 i . 16 2 ' ' 4 7 4 .4 4 ' T otal ; 213 . 1 » ' . t 100. 00
Based on analysis of 90 50-foot line transects.
^ The crown density of trees and shrubs was 474. 44/4, 500 or . 1054. 38
Appendix Table C. Forbs (in order of abundance).
1 S p ecies ' Fre- Line Intercept 1 quency t ' N o. 1 F e e t P e r c e n t1 | | Chenopodium watsoni ' 133 1 44. 12 46. 12 t 1 Artemisia ludoviciana ' 90 1 24. 37 25. 48 i t Viguiera annua ' 56 1 19. 42 20.30 i t Sisymbrium linearifolium ' 9 1 3.67 3.84 i 1 Penstemon sp. ' 10 I 2.75 2.87 I Cheilanthes sp. ' 8 1 1.33 1.39 i 1 t 1 2 T otal , 306 ! 95. 66 100. 00 1
1 Based on analysis of 90 50-foot line transects.
^ The crown density of forbs was 95. 66/4, 500 or . 0213.