:7
<2.

. ,
. . .Af.
In“

3.

a :3
. L3
343“? in. u:

34.... -. a
.x .3va
:6

...I
1.1.. ..

 

 

 

\LI

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

thesis entitled

THE EFFECTS OF FALL BURNING ON TWO NORTHERN
CALIFORNIA PERENNIAL BUNCHGRASS COMMUNITIES

presented by

Charles C. Hughes IV

has been accepted towards fulfillment
of the requirements for

M.S. degree in Plant Biology

Om Magmsixfin’

Major professor

 

“fit/03

07639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8/01 cleIRGIDateDuo.p65-p.15

 

THE EFFECTS OF FALL BURNING ON TWO NORTHERN CALIFORNIA
PERENNIAL BUNCHGRASS COMIVIUNITIES

By

Charles C. Hughes IV

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Plant Biology

2003

ABSTRACT

THE EFFECTS OF FALL BURNING ON TWO NORTHERN CALIFORNIA
PERENNIAL BUNCHGRASS COMMUNITIES

By

Charles C. Hughes IV

California native perennial grasslands are declining due to pressure from exotic
species, changes in disturbance regimes and loss of habitat. Prescribed burning has been
used by land managers to promote native species, control invasive weeds, and increase
growth and reproductive output of perennial grasses in California grasslands. This study
examines the effect of fire on native species, exotic species, and perennial grasses in the
northern Sacramento valley. Communities containing two native perennial bunchgrasses,
Nassella pulchra and Aristida temipes var. hamulosa, were selected for burning. Data on
bunchgrass vegetative and reproductive growth, botanical composition, Shannon-Weiner
diversity, and seed bank composition was collected in late spring of 2001 and 2002;
burns were conducted in October of 2001. For N. pulchra, fire increased both basal area
and number of inflorescences, as well as seed weight. No effect on basal area or
inflorescences was found for A. ternipes, but fire sharply increased the mortality rate of
this species. Fire increased the Shannon-Weiner diversity index in the Nassella
community, but the heightened diversity was due to increases in exotic species. Annual
grasses decreased while annual forbs increased due to fire. Taeniatherum caput-medusae
and several other species, both native and exotic, experienced significant changes in
cover with burning. Annual grasses in the seed bank were reduced by burning. Fire was

helpful in restoring the Nassella community but not the Aristida community.

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Carolyn Malmstrom, for all her advice and
guidance throughout this project. I would also like to thank Dr. Peter Murphy, Dr. David
Rothstein, and the National Science Foundation for support (DEB 9983373). I am
grateful to Peter Hujik and The Nature Conservancy for the use of land and facilities at
the Dye Creek Ranch and Vina Plains Preserve. Scott Butterfield and Jessica Cook

provided assistance in the field and the greenhouse.

iii

PREFACE

The composition of California grasslands has changed dramatically over the last
200 years (Clements 1934, Heady 1988, Hamilton 1997, Schiffman 2000, Holstein 2001),
in part because natural disturbance regimes have been disrupted by human activities. In
the past, California grassland ecosystems experienced fires, floods, and grazing by
migratory elk and pronghom antelope (Heady 1988). Now these grasslands rarely
experience fires or floods, and instead are subjected to grazing by cattle and sheep,
pressure from exotic species, and a general loss of habitat due to urbanization and
agriculture. Perennial grasslands currently exist as widely scattered patches, sometimes
no larger than a tenth of a hectare. These remaining patches of bunchgrasses usually
exist within a matrix of invasive exotic species. Established bunchgrass individuals
under such pressure decline in size, and populations may not recruit well (Dyer and Rice
1997, Hamilton et a1. 1999, Brown and Rice 2000). Researchers have tried to mimic
natural disturbance regimes in order to restore bunchgrass populations, and have met with
mixed success (Dyer et al. 1996). Prescribed burning has become more common as land
managers try to simulate natural fire regimes. The Nature Conservancy alone burns
3,000-10,000 acres annually in California (P. Hujik pers. com.) The effects of
prescribed burning on natural communities may be quite variable due to variables such as
season of burn and annual and monthly patterns of precipitation. However, some trends

are apparent and are addressed here.

iv

Most prescribed burning in California is conducted during the spring, even though
natural burns are more likely to occur during the fall when the vegetation is drier and
there is more lightning as a source of ignition. Furthermore, most prescribed burning is
done solely for the purpose of invasive weed control. This project examines the effects
of a fall prescribed burn on several ecological attributes within two native perennial
grassland communities, one containing Nassella pulchra (Purple Needlegrass), and one
containing Aristida term'pes var. hamulosa (Prairie Three-Awn).

Chapter 1 reviews what is known about the historical extent of bunchgrasses in
California and the original fire regime; the effects of fire on the seed bank, the botanical
composition, and on N. pulchra and A. temipes; and the effects of fire on soil properties.
Chapter 2 describes a manipulative experiment which subjected two grasslands
communities to prescribed burning. A seedling key to some common California grasses
was developed as a result of work with the seed bank, and is presented in Chapter 3 as an

aid to anyone concerned with identifying grasses at an early stage.

Literature Cited

Brown, C. S., and K. J. Rice. 2000. The mark of Zorro: Effects of the exotic annual grass
Vulpia myuros on California native perennial grasses. Restoration Ecology 8: 10-
17.

Clements, F. E. 1934. The relict method in dynamic ecology. Journal of Ecology 22:39-
68.

Dyer, A. R., H. C. Fossum, and J. W. Menke. 1996. Emergence and survival of Nassella
pulchra in a California grassland. Madrono 43:316-333.

Dyer, A. R., and K. J. Rice. 1997. Intraspecific and diffuse competition: The response of
Nassella pulchra in a California grassland. Ecological Applications 7:484-492.

Hamilton, J. G. 1997. Changing perceptions of pre-European grasslands in California.
Madrono 44:311-333.

Hamilton, J. G., C. Holzapfel, and B. E. Mahall. 1999. Coexistence and interference
between a native perennial grass and non-native annual grasses in California.
Oecologia 121:518-526.

Heady, H. F. 1988. Valley grassland. Pages 491-514 in M. G. Barbour and J. Major,
editors. Terrestrial Vegetation of California. John Wiley & Sons, New York.

Holstein, G. 2001. Pre-agricultural grassland in central California. Madrofio 48:253-264.

Schiffman, P. M. 2000. Mammal burrowing, erratic rainfall and the annual lifestyle in the
California prairie: Is it time for a paradigm shift? in J. E. Keeley, M. Baer-
Keeley, and C. J. Fotheringham, editors. 2nd interface between ecology and land
development in California. US. Geological Survey Open-File Report 00-62.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................ viii
LIST OF FIGURES ........................................................................... ix
CHAPTER 1
FIRE AND CALIFORNIA GRASSLANDS .............................................. 1
1 California grasslands ....................................................... l
H. The history of fire in California grasslands .............................. 6
III. The effects of burning during different seasons ......................... 8
IV. Taeniatherum caput-medusae and the effects of thatch ............... 10
V. The seed bank ................................................................ 11
VI. The effects of burning on botanical composition ....................... 12
VD. The effects of burning on Nassella and Aristida ........................ 14
VD. The effects of burning on soil properties ................................. 20
IX. Conclusions .................................................................. 21
X. Literature Cited .............................................................. 22
CHAPTER 2
THE EFFECTS OF FALL BURNING ON TWO NORTHERN CALIFORNIA
PERENNIAL BUNCHGRASS COMMUNITIES ...................................... 27
I. Introduction .................................................................. 27
11. Methods ...................................................................... 33
1. Study Sites .......................................................... 33
2. Design ................................................................ 33
3. Analysis ............................................................. 38
III. Results ......................................................................... 39
IV. Discussion .................................................................... 52
V. Literature Cited ............................................................... 58
CHAPTER 3
A SEEDLING KEY TO SOME COMMON CALIFORNIA GRASSES ............. 62
I. Introduction .................................................................. 62
11. Methods ....................................................................... 64
II. The grass seedling key ...................................................... 65
III. Literature Cited .............................................................. 66

vii

LIST OF TABLES

Table 1. The responses of individual species and functional groups to burning.
Only species encountered during the experiment were listed. A “+” or “-“
indicate species cover increased or decreased in cover one year after burning ....... 15

Table 2. Response of functional group cover for both communities. Data is
Mean(SE). Significant (p < 0.05) differences for a given year are in bold ............ 36

Table 3. Response of species cover in the Nassella community. Data is

Mean(SE). Significant (p < 0.05) differences for a given year are in bold.
Functional group abbreviations are defined in Table 1. Members of the genera
Allium, T riteleia, and Dichelostemma were grouped together because they are
difficult to separate without flowers and they are all bulbaceous members of
Liliaceae ........................................................................................ 46

Table 4. Response of species cover in the Aristida community. Data is

Mean(SE). Significant (p < 0.05) differences for a given year are in bold.

California Native Plant Society list 4 species are in bold. Castillija attenuata

and Triphysaria eriantha ssp. eriantha were grouped together because they are
difficult to separate after senescence, and they are both root hemiparasites ........... 48

Table 5. Number of seeds per gram of soil in the Nassella community. Data

is Mean(SE). Significant (p < 0.05) differences are in bold. N = 8 for each
treatment ....................................................................................... 53

viii

LIST OF FIGURES

Figure 1. Map of northern California, indicating the two study sites in this project.
Map obtained from US. Geological Service, Flagstaff Field Center,
http://wwwflag.wr.usgs.g0v/USGSFlag/Data/maps/shadedRel.html .................. 2

Figure 2. The average annual pattern of precipitation and PET at Red Bluff, CA.
Data was obtained from the National Climatic Data Center, Red Bluff Municipal
Airport station (http://lwf.ncdc.noaa. gov). PET was calculated according to
Thomthwaite (1948). The approximate monthly growing periods of N. pulchra
and A. temipes are shown .................................................................... 3

Figure 3. The relationship between Taeniatherum caput-medusae cover and
thatch depth in plots from the Nassella community in 2001. The line is given by
the equation y = -5x2+8.65x+0.61 ........................................................... 30

Figure 4. Plot layouts in both communities. White areas are control, grey
areas were burned in October, 2001. Black squares represent 1 m2 plots ............ 34

Figure 5. Mean change in basal area per plant from Spring 2001 to Spring 2002.
The burn treatment was applied in October 2001. Standard error is shown.
Individuals that died were not included .................................................... 40

Figure 6. Mean change in number of inflorescences per plant from spring 2001
to spring 2002. The burn treatment was applied in October 2001. Standard
error is shown .................................................................................. 41

Figure 7. The mortality rate for individual bunchgrasses in 2002. Standard error
is shown. Significance was calculated with a Chi-square test .......................... 42

Figure 8. Effect of seed weight on N. pulchra germination rate. Standard error
is shown. Although burned plants produced heavier seeds, seed from both
burned and control plants fell within the outlined size class ............................ 43

Figure 9. The cover of four plant functional groups at the Gray Davis Dye Creek
Preserve from 1997 - 2002. Cover data were collected between 15 April and

15 May. There were no perennial grasses observed. Data were obtained from the
Nature Conservancies Dye Creek botanical monitoring program; data for 2001

and 2002 were collected by the author. Data used by permission of Peter Hujik,

Dye Creek Preserve manager ................................................................ 45

Figure 10. Shannon-Weiner diversity in the Nassella community. Standard
error is shown. Bums were conducted in October, 2001 ................................ 51

ix

Chapter 1
Fire and California Grasslands
1. California grasslands

Most of the California grasslands are located in the Central Valley, which lies
between the Coastal mountain ranges and the Sierra Nevada, and in the adjacent coastal
valleys. The Central Valley is completely surrounded by a ring of foothills, except for
the Sacramento and San Joaquin River deltas. The foothill areas are largely comprised of
an oak savannah habitat containing the California endemic Quercus douglasii (Blue
Oak). Q. douglasii usually forms an open canopy forest, leaving numerous large and
small patches available for grassland. The California oak savannah habitat fades into a
more pure grassland habitat in the lower foothills and the valley floor. The two study
sites in this project are located in the northeastern Central Valley (Figure 1).

Most of the California grasslands experience a Mediterranean climate (Figure 2)‘,
which is characterized by cool, wet winters, and hot, dry summers. The rainy season in
California usually begins in October or November and extends through the winter, with
the last significant rainfall occurring in April or May. Precipitation is almost non-existent
from June through August. The temperatures of the low elevation grasslands usually do
not dip far below freezing in the winter. In the summer, highs exceeding 37° C are
common. In a Mediterranean climate, the annual precipitation and the annual potential
evapotranspiration are perfectly out of phase with each other. Most California grassland
vegetation grows during the winter and spring. In the summer and fall, annual species

are senescent while perennials have become dormant until the next growing season.

 

1 Data obtained from National Climatic Data Center, http://lwf.ncdc.noaa. gov. Potential evapotranspiration
calculated according to (Thomthwaite 1948).

NEE.NMKEEQEQEEBQMEKMOQVEMdun:ukfiufiifiiEE: £850 205 teams—m .ooEom Rama—80
.m.D Set 85890 93.4 .80“on £5 5 mafia beam 95 2: wcuaofii .fiEotEU E055: .«o 9%; A 9.53%

 

5:500 83m
.033on €me
92> 2385888
SEEK £3,262

3550

«Senor—t £25m
M080 PAD EEO
>30 ”5858800
BEES efitfisv‘

.5505. 2a BEES .< 23 63923 .2 ho mnotom wEBEw 35:9: Bunnies“; 23. .vaac 283562; 8
$6808 Baa—:28 we? Ema .A>ow.mmo=.089035935 nouns tomb/x 393.52 #35 BM £850 San 338:0
Ecouaz of So: 3:850 33 3mm .<0 .tEm com a Ema new cosmzfiooa the 50:8 13:5 omega 05. .N 95w:—

 

 

_ BEES antennax _

. . . 323.32 _ _ 63.63% . . .

8o 52 60 new on< .2. :2. >32 .9. 32 no“. can

 

 

 

 

B

1

9

l

\r/

3

W
-3, .x. cozmcamcmzoagm r or
-olo‘-.- _m_wcmw01.\:\\‘ F wF
85385 III i om

 

 

 

 

 

 

Most California grasslands are comprised almost entirely of exotic species. The
most successful invaders are annual grass species from the Old World Mediterranean
regions of southern Europe and Asia Minor. Particularly successful annual grass genera
such as Avena and Bromus have come to be dominant species in much of the California
grassland. The invasion of California grasslands by exotics may have begun even before
the first permanent European settlements, the Spanish missions, were established
(Mensing and Byme 1998). The present preponderance of exotic species in California
obscures the original grassland composition, but even still researchers have drawn
conclusions about the composition of the grasslands before European contact.

Clements (1934) postulated that a vast perennial grassland once covered the entire
valley, but his evidence was limited to observations made in often-disturbed railroad
rights of way. Heady (1988) offers the opinion that perennial bunchgrasses were the
climax community on most well-drained upland areas within the valley. Others suggest
that a variety of habitat types may have occurred within the valley, including perennial
grassland, annual grassland, Chaparral, and even desert in the South (Hamilton 1997,
Schiffman 2000). Holstein (2001) gives evidence that lowland areas within the valley
were dominated by rhizomatous grasses and sedges. One study has demonstrated the
contraction of a perennial bunchgrass population on a small spatial scale (Bartolome et al.
1986), but experimental evidence of changes in cover of perennial grassland at the
landscape level is lacking. It is reasonable to assume on the basis of competition
experiments between native perennials and exotic annuals that more of the California
grassland was once occupied by perennial bunchgrasses (Bartolome and Gemrrrill 1981,

Dyer and Rice 1997, Clausnitzer et al. 1999, Dyer and Rice 1999, Hamilton et a1. 1999).

Current opinion holds that native rhizomatous graminoids prevailed in the lower,
flood-prone areas of the valley (Holstein 2001), and that native annual grasses and forbs
were common and possibly dominant in well-drained areas and the surrounding foothills
(Hamilton 1997, Schiffman 2000). Patches of native perennial bunchgrasses were likely
common in the surrounding foothills and some upland sites within the central valley
(Heady 1988, Holstein 2001).

Since European contact, some of the ecological forces shaping California
grasslands have changed. The state has experienced invasion by numerous exotic
species, leading to large changes in botanical composition. Patterns of grazing have also
changed in the last several hundred years (Heady 198 8). Migratory herds of elk and
pronghom antelope used to move freely through the grasslands. Today, much of the
range is fenced and used for cattle and sheep grazing. Grazing by domestic livestock
may be much more intense if the stocking rates are high. The pattern of another
disturbance agent, fire, has been altered with the development of the state (D'Antonio and
Vitousek 1992). In the prehistoric grasslands, fire in the summer and fall would have
been able to spread through the dry grasslands unimpeded by modern developments such
as roads and irri gated pastures. Currently, active fire suppression has largely removed
fire from the grasslands.

This project examines two of the many species of perennial bunchgrasses native
to California. Nassella pulchra (Purple Needlegrass), a C3 perennial bunchgrass, is one
of the more common perennial bunchgrasses in California. This species begins growth
after the first significant precipitation, usually in the late fall or early winter. Vegetative

growth continues through the winter and early spring. In late spring, inflorescences are

produced, and the plant senesces in early summer. This seasonal growth pattern is
common among perennial bunchgrasses of the state. Aristida temipes var. hamulosa
(Hook Three-Awn), a C4 perennial bunchgrass, is less common and usually grows in
drier conditions than N. pulchra. Growth begins in the spring and the plant remains
green throughout much of the summer. Inflorescences are produced continually, and may
be found from late spring through the end of the summer. By fall, A. temipes has
senesced, and remains dormant until the next growing season. The seasonal growth
pattern of A. ternipes is less common among California perennial bunchgrasses and

provides a contrast to N. pulchra.

II. The history of fire in California grasslands

Fires were likely common in the grasslands of the Central Valley prior to the
settlement of the region by Europeans (Biswell 1989, D'Antonio and Vitousek 1992).
The grasslands are dry for several months of the year and ignition sources could have
started fires that may have spread for several square kilometers. Ignition of fire in most
natural cases is attributable to lightning (Barbour et al. 1999), which is more common at
higher altitudes, but still occurs with sufficient frequency to potentially cause hundreds of
wildfires annually in the valley and surrounding foothills (Biswell 1989). Biswell (1989)
presents evidence that in some years, when climatic conditions are favorable, there may
have been many more fires in the lower elevations. In addition, Native Americans are
believed to have deliberately burned parts of valley grasslands for several possible
reasons (Biswell 1989). The fires may have been lit to improve hunting prospects by

creating a flush of new grass growth to attract grazers. Alternatively, the fires may have

been lit to promote the abundance of acorns, another food source. A short fire return
interval is known to promote the dominance of oaks under some conditions in other parts
of North America (Abrams 1992, Van Lear and Watt 1992). Since the purpose is
unclear, the season of Native American fires is not known. Natural fires may have been
more common in the fall when there is a greater likelihood of lightning (Biswell 1989),
but the true pre-historic fire regime is not known with accuracy.

The fire regime in the Central Valley was likely altered as settlers arrived. Not
only was fire actively suppressed, but the native natural communities were also
disappearing as land was plowed for agricultural use. An official government policy of
fire suppression began with the creation of the Forest Service in 1905 (Biswell 1989), and
it has not been until recent decades that the contributions of fire to natural ecosystem
functioning have been widely recognized. Increasingly, prescribed burning has been
used to simulate a natural fire regime and to restore and maintain natural ecosystem
conditions and functioning in California grasslands (Menke 1992, DiTomaso et al. 1999,
Meyer and Schiffman 1999). However, these goals are usually secondary to the main
goal of noxious weed control.

The best times for prescribed burning in California are during the late spring and
early fall, when vegetation is relatively dry and temperatures are lower than their summer
extremes. Fall burns are often hotter and more complete than spring burns, since the
fuels have cured throughout the summer drought. Natural burns are more likely to occur
in the fall, since lightning strikes are more frequent at that time of year. Today,

prescribed burning in the fall is often prohibited due to air quality constraints.

III. The effects of burning during different seasons

The season in which a burn occurs may have large effects on grassland botanical
composition the year following fire. In Mediterranean climates efforts have focused on
examining differences between cool-season (winter and spring) and warm-season
(summer and fall) burns (Meyer and Schiffman 1999). There are two primary
explanations offered to explain the different results of warm and cool-season burns. The
first explanation is that target species differ in phenologies and may show differential
vulnerability to burns in different seasons. For instance, some species mature and drop
seed early in the spring, while others may not drop seed until mid-summer or later.
Differences in phenologies have been exploited to control late maturing invasive weeds
such as Taeniatherum caput-medusae (Medusa Head) and C entaurea solstitialis (Yellow
Star-Thistle) in stands of earlier maturing natives (DiTomaso et al. 1999).

The second explanation is that fire intensities can differ between the two seasons
due to weather conditions and fuel availability. Meyer and Schiffman (1999) found
differences in the relative proportions of plant functional groups, notably increased forbs
and decreased grasses, as a result of both cool and warm season burns. They also found
that both late spring and fall burns had a larger impact on functional groups than winter
burning. They concluded that both hypotheses were supported. Parsons and Stohlgren
(1989) observed similar results from spring and fall burns but had no winter bums for
comparison.

Species response to fire depends upon the point at which the fire occurs within its
annual growth cycle. In the case of annual plants, a fire that occurs after a species has

begun growth during the wet season but before the species has dropped seed may reduce

 

the cover of that species the following year by reducing that species’ annual contribution
to the seed bank. A fire that occurs during the period an annual species is present only as
seed may have little effect on that species’ population if the seeds are protected from the
burn. Similarly, perennial species are most likely to be affected by burns during periods
of active growth, not when the species is dormant. For example, N. pulchra begins
growing early in the wet season and has completed growth by late spring, when little
moisture is left in the soil (Figure 2). The perennial bunchgrass A. ternipes begins
growth in the spring and continues through the dry summer months, finally senescing in
late summer or fall. A late spring burn may have very different effects on these two
species, since it would occur just as N. pulchra is senescing but just as A. temipes is
beginning its annual growth, possibly resulting in a greatly reduced annual biomass
output for A. ternipes. A late fall burn is less likely to have different effects on the two
species since both would have senesced by that time.

Much of the prescribed burning currently performed in California is spring
burning aimed at controlling late-maturing exotic weeds. Three species noted as major
rangeland weeds by DiTomaso (2000) occur within the two communities I have chosen
to study: Centaurea melitensis (Malta Star-Thistle), C. solstitialis, and T. caput-medusae.
C. melitensis has not become an especially pernicious pest in the Central valley and
is a minor component of the Aristida community. C. solstitialis is a major pest in
California and has been controlled with repeated annual burning (DiTomaso et al. 1999)
but only occurs in the two communities examined here in low concentrations. T. caput-
medusae is a major weed occurring throughout much of the west and accounts for

approximately a third of the cover in the Nassella community.

 

IV. T aeniatherum caput-medusae and the effects of thatch

T. caput-medusae is an annual grass that is unpalatable to livestock in the later
stages of growth and decomposes slowly, due to the high silica content of its tissues
(Young 1992). Due to the slow decomposition rate, grasslands with large amounts of T.
caput-medusae build up a layer of senesced thatch on the soil surface. The seedlings of
T. caput-medusae have higher gemrination and survival rates in a high thatch
environment due to their ability to germinate above the ground and grow roots down into
the soil (Evans and Young 1970, Young 1992). A deep thatch layer inhibits the seedlings
of most other grassland species. This effectively creates a positive feedback loop in
which the presence of T. caput-medusae produces thatch which promotes even more T.
caput medusae.

The accumulation of thatch leads to moist, cool, and dark conditions near the soil
surface, as well as the reduction of environmental extremes (Evans and Young 1970,
Reynolds et al. 2001). Precipitation promotes thatch accumulation by stimulating plant
growth, but also inhibits thatch accumulation by increasing decomposition rates. In
California, the highest thatch accumulation occurs in grasslands with about 635 mm of
annual rainfall (J anes 1967). 635 mm is a medium amount of precipitation for Califomia
grasslands, and close to the average received by both communities in this study. The
removal of thatch opens up the surface to sunlight, creating a drier, warmer rrricroclimate.
The differing soil surface conditions created by thatch or the lack of it can control which
species germinate and survive.

The seed of T. caput—medusae matures 2-4 weeks later than other common annual

grasses, both native and exotic, and so control efforts have focused on burning in that

10

window of time. McKell et al. (1962) found excellent control using this method in the
Central valley, and even a late summer burn after most T. caput-medusae seed had
dropped resulted in a large decrease in T. caput-medusae cover the following year. In
contrast, fire on the Modoc Plateau in the Great Basin floristic province, slightly
increased the cover of T. caput-medusae, even after 3 consecutive annual burns (Young
et al. 1972). Differences in the seed bank may account for the varied results. The first
study started with a 23% cover of forbs, while the second started with only 1%. It may
be that the second study failed to control T. caput-medusae because there were virtually

no other species present in the seed bank.

V. The seed bank

The seed bank of Mediterranean grasslands exhibits a predictable annual cycle,
reaching a peak in late spring and summer following seed maturation by most species,
and reaching a low point in the following early spring when most species have
germinated (Russi et a1. 1992). Legumes are the only prominent group with substantial
seed carryover from year-to-year, and annual grasses show nearly nonexistent carryover,
since nearly all grass seedlings germinate during the winter rains (Young et al. 1981,
Russi et al. 1992). Perennial grasses comprise little if any of the seed bank, even in areas
where they are a significant part of aboveground vegetation (Major and Pyott 1966,
Kotanen 1996). The composition of the seed bank often does not wholly correspond with
the composition aboveground, with species appearing aboveground absent from the seed

bank and vice versa (Major and Pyott 1966). Most seeds in grasslands are in litter on the

11

soil surface or top 2 cm of the soil (Kotanen 1996). Viable seeds are unlikely to be found
in large numbers at greater depths due to the small year-to-year carryover.

Fire destroys most of the seeds on the soil surface, but even a 1-cm layer of soil
will protect most seeds from the heat of the burn (Morgan 1999). Any seeds entangled in
thatch will almost certainly be destroyed by fire. Burning is likely to shift the balance of
the seed bank in the direction of long-term dormant seeds buried deeper in the soil. I
propose that the seeds of forb species are more likely to fall through thatch to the soil
because of their often smaller size relative to grass seed and because the seeds of many
forb species do not posses awns. Thus, fire may destroy more grass seed than forb seed
and alter the seed bank in favor of forbs. The seed bank may display significant annual
variation, depending on which species are the most successful depositors in a given year
(Russi et al. 1992). Fire thus indirectly alters the seed bank by affecting aboveground

botanical composition.

VI. The effects of burning on botanical composition

Fire may affect the cover of different plant functional groups the following
growing season, but the results have been found to vary depending on the season of the
burn. This project breaks down vegetation into five commonly used functional groups;
annual grasses, perennial grasses, annual forbs, perennial forbs, and legumes. Burning
has been shown to decrease annual grass cover in some cases, especially when done
during the spring when seeds have not all fallen to the ground (Meyer and Schiffman
1999). DiTomaso et al. (1999) observed no cover change in this group with a spring

burn, but that may be because the study was done at a site heavily infested with C.

12

solstitialis. Larson and Duncan (1982) and Parsons and Stohlgren (1989) both saw little
change in annual grasses the year after a fall burn. Detecting changes in the relative
abundance of native and exotic annual grasses is more difficult because native annual
grasses are relatively uncommon in California valley grasslands.

The cover of perennial grasses, including N. pulchra, has not changed
significantly due to fire (DiTomaso et al. 1999, Hatch et al. 1999). Ahmed (1983)
observed a significant increase in the basal area of burned N. pulchra individuals, but this
would not necessarily have resulted in a measurable increase in vegetative cover for the
species, and Langstroth (1991) observed the exact opposite effect. Vegetative cover is a
broad measurement, and is unlikely to detect finer scale changes in perennial grasses.

The cover of forbs generally increases following prescribed burning (Parsons and
Stohlgren 1989, Fossum 1990), particularly native species (DiTomaso et al. 1999). In
particular, perennial forbs from bulbs appear to increase in cover following fire (York
1997). DiTomaso et al. (1999) and McKell et al. (1962) have reported an increase in
legumes, but other studies have not always treated them as a separate functional group,
although some have shown significant increases in individual legume species (Parsons
and Stohlgren 1989, York 1997). Overall, native species generally increase relative to
exotics following burning, but the effect is less pronounced for fall burns (York 1997,
Meyer and Schiffman 1999).

The studies from California grasslands show strong agreement that species
richness and diversity tend to increase following fire, with the Shannon-Weiner Diversity
Index increasing 25-45% (Parsons and Stohlgren 1989, York 1997, DiTomaso et al.

1999, Meyer and Schiffman 1999). The cover of species that dominate in the absence of

13

fire tends to be reduced, opening up space for other species, and thus increasing diversity.
However, this effect may be partially offset by increases in species that compete better in
soils with increased nutrients (Huenneke et al. 1990, J anssens et al. 1998). Most studies
use both native and exotic species when calculating a diversity index. However, Meyer
and Schiffman (1999) used a native-only Shannon-Weiner Diversity Index and found that
it increased significantly from .04 to .34 with fall burning.

The vegetative cover of individual species may change greatly following fire, and
again the response of a particular species may depend on the season of the burn and also
the weather during the following growing season. In some cases, species that were
completely undetected in the plots before burning appear afterward (Meyer and
Schiffman 1999), indicating either the presence of the species in the seed bank or
colonization following the fire. The general trend in the literature shows that native
species occurring within the study sites of this project are more likely to increase in cover

following fire than exotics (Table 1).

VII. The effects of burning on Nassella and Aristida

Fire has the potential to stimulate perennial bunchgrass growth as well as to
decrease survivorship. Burning may stimulate growth by reducing competition,
removing thatch, or increasing soil nutrient availability. Burning may kill or reduce the

size of bunchgrasses by destroying shoot meristems, thus removing the plant’s ability to

14

0.553 .050 00> 0:0 .050
:_ 00009000 .0 00000.05 .0>0o 020000 20205 .0 ..+.. < 00.0: 0.03 EmE:00x0 05 0550
00.95080 020000 2:0 .0553 0. 00:05 0:285: 0:0 020000 032205 .0 0005000.. 05 .F 030...

 

 

 

 

 

 

 

 

- - O< 00:05:: 0.555

+ + ”2 00:00 0:050

- “2 .00 00500.:

+ - “2. 0503 0:00:03»:

+ u< 00:00.0: 53.500

+ u< 0.0:030E 00:25:00

- 0< 00:05:05: 02:05

+ - 0< 03000000: 02:05

- . 0< 0330.6 025$

- + 0< 050: 0:0><

00.025 ozoxm

+ u< 05:0.:0 .000 05:0:0 000555

+ ._ E50305 53505.:
+ u< «3:058 0.058203

+ n_< 050::050 03800055

+ u_< 0:50.005 0.5000> 02

+ ._ 03:0: 03:53

+ + ._ 3:053:83 053

+ u< 0550.300 0.50503

+ u< 0.0::00 030005

+ n_< 0503:0550 005.5000
+ Mn .00 000.60.5l

0:80 00.025 02.02
.ocam
=0”. :0“.
a actaw =0”. a 950m =0". 950m :35 Fe 08:.
mam F mom F mam F mam F mam F
.:0E:Eom £0050 £05205 50> .._0 .6

0 .035. 0 50.0.. 0 0:00.00 000E0._._o

15

 

 

 

+ .. 00:500..
+ "E 050”. _0_::0:0m
on. 00000.0 _0_::0:0n.
+ + “E 050“. 0.89:
- 6.4. 000005 _0:::<
00:20 3:020:3—
000 F «00 F 000 F 000 F 000 F
.:0E:_:0m 60055 605.55 50> .._0 .00
w .0082 a. :00:0._ .0 0:00:00 000E920

.8083 F 0.an

l6

re-sprout. Most of what is known about the response of California perennial
bunchgrasses to burning comes from the study of a single species, Nassella pulchra.

There exists a general belief among managers that burning is good for N. pulchra.
In fact, the reported effects of fire on the vegetative growth of N. pulchra are mixed, and
may depend in part on variation in local conditions and genetic ecotypes, which may be
considerable (Knapp and Rice 1998). Ahmed (1983) found that both summer and fall
burning dramatically increased tiller length and tillers per plant for the next two growing
seasons, and caused a smaller but significant increase in basal area. In contrast, the basal
areas of plants burned in late summer have been found to be as much as 40% smaller than
unburned plants in the following growing season (Langstroth 1991). Spring but not
summer grazing reduced the observed differences. The same study found that burned
plants were 71% shorter.

One explanation for the variety of burn responses is soil differences. The
Langstroth (1991) study took place on poor soil with a shallow hardpan horizon. Dyer
and Rice (1997) observed the effects of burning on N. pulchra basal area on the same
poor soil and found no difference between burned and unburned plants. Huntsinger et al.
(1996) applied clipping treatments to N. pulchra plants in three different regimes: winter,
winter and early spring, and late spring. N. pulchra seeds were collected from two
p0pulations but were grown in a common garden. The soil of one population was a
typical grassland soil but the soil of the other was a serpentine soil, which has nutrient
levels unsuitable for the growth of many plants. The winter clipping had no effect on
tillers or inflorescences in either population, but winter and early spring clipping reduced

tillers and inflorescences only in the serpentine population. Late spring clipping severely

l7

reduced tillers for both populations. Huntsinger et al. (1996) suggests the differing
responses from the two ecotypes may be due to adaptations to different overall plant
densities in the communities where they originate. These studies suggest that soil fertility
and ecotype may help predict the response of N. pulchra vegetative growth to defoliation.
N. pulchra ecotypes adapted to poor soils may not recover from defoliation as readily as
ecotypes from more fertile soils.

The effects of defoliation on reproductive output are also mixed, and may depend
in part on soil fertility. Huntsinger et al. (1996) report varied reductions in inflorescences
between serpentine and non—serpentine populations with spring and winter clipping, but
not with winter clipping alone. In fall-bumed N. pulchra, Langstroth (1991) observed a
reduction in both inflorescences and seeds per inflorescence, resulting in a 45% reduction
in seed per plant. However, average seed weight increased by 14%. Among seeds from
the same population stored for 10 years, those from burned plants were 20% larger and
had a 72% higher germination rate (Dyer 2002). Ahmed (1983) found that mowing
increased inflorescence and seed number per plant only sometimes, but that summer or
fall burning always increased those numbers. The soil in the Langstroth (1991) study had
a shallow, impermeable hardpan layer which would have prevented the plants from
utilizing deep water in the late spring. This may have interfered with reproductive output
and thus contributed to the different findings of the studies.

Fire may increase the germination and early survival of N. pulchra seedlings.
Dyer et al. (1996) found that fall burning increased germination the following spring, but
that mortality was high and virtually no individuals had survived four years after the start

of the experiment. The authors hypothesized that a year of heavy rainfall may also be

18

necessary for recruitment. In a greenhouse experiment, Ahmed (1983) found that N.
pulchra seedlings had a higher survival rate in bare pots than in pots covered with a
mulch layer, suggesting that the removal of thatch by fire creates conditions more
favorable for seedlings. But perhaps more importantly, the same study found that
burning reduced the amount of exotic annual grass seed, thus reducing the interspecific
competition experienced by the seedlings. The deepest soils in the Langstroth (1991)
study were observed to have more N. pulchra seedlings when burned and grazed than
control plots.

There are many bunchgrass species other than N. pulchra native to California.
The perennial bunchgrass group represents a range of life history strategies including
both cool and warm season species, and species adapted to a range of habitats. These
species may respond to burning differently than N. pulchra. In addition to N. pulchra,
this study examined Aristida ternipes, a warm season bunchgrass whose response to fire
is unknown. The effects of fire on other Aristida species are mixed. In some cases,
burning has no effect, as Brockway et al. (2002) found for A. purpurea cover in a
shortgrass prairie. In A. stricta, fire has been found to increase both vegetative and
reproductive growth in established plants, as well as to increase tissue concentrations of
nitrogen and phosphorous (Parrott 1967, Anderson and Menges 1997), perhaps because
of increased nutrient uptake immediately following the burn. In A. beyrichiana, fire
produces a pulse of recruitment, but can also increase mortality of young plants
(Mulligan and Kirkman 2002). Winter burns also increased mortality in seedlings in
annual A. oligantha populations in the Midwest (Owensby and Launchbaugh 1977), but

did not reduce the total annual biomass of the species (Engle et a]. 1990). Competition

19

 

experiments with A. stricta in Florida show that removing surrounding vegetation also
increases seedling mortality (Kindell et al. 1996), but this effect has not been seen in A.
longiseta (Fowler 1990). On the other hand, removal of thatch by burning may promote
Aristida populations, as thatch has been shown to inhibit A. longiseta seedlings (Fowler

1988), as well as preventing the seeds from contacting the soil (Fowler 1986).

VIII. The effects of burning on soil properties

Changes in the soil environment induced by fire are likely to affect the growth of
both species in the following growing season. The charring effect of fire creates a
darkened soil surface that may result in temperature changes both above and below the
soil. A month after burning, Ewing and Engle (1988) found significantly lower
temperatures 10 cm above the soil, and higher temperatures 1 cm below the soil surface.
DiTomaso et al. (1999) found that increased belowground temperatures due to burning
may persist for at least 3 years and extend to 10 cm. Thatch acts as an insulator, and its
removal increases temperature extremes in both directions, but the effects are greater
above than belowground (Evans and Young 1970). Higher temperatures in burned sites
may in turn lead to lower soil moisture due to increased evaporation (Smith and Owensby
1973). Significant differences in soil water content were not found by Ewing and Engle
(1988) or Larson and Duncan (1982), but the former did find lower leaf water potentials
in burned perennial bunchgrasses in the Midwest. In some cases, fire may decrease soil
moisture by creating a hydrophobic layer that resists absorption of rain, but this effect is
most pronounced in high-intensity fires or habitats containing vegetation with large

amounts of waxes and oils, such as Chaparral (Biswell 1989).

20

Grassland fires generally have a small positive effect on the amount of soil
nutrients available to plants (Smith and Owensby 1973, Vogl 1974), as long as the heat
generated is not intense (Biswell 1989). Nitrogen can be volatilized and lost to the
atmosphere due to the heat of the fire (Seastedt and Ramundo 1990), but this loss is often
offset by nitrogen released in the conversion of litter to ash, or by the conversion of
nitrogen in the soil to a usable form from the heat of the fire (White and Loftin 2000).
Romanya et al. (2001) suggest that the heating of soils by fire converts organic nitrogen
to ammonia, a form readily used by plants. Grasslands that have not burned recently
build up a layer of thatch containing minerals in an unusable form (Harrison 1985) that
are released by fire (Vogl 1974). The nutrients released by burning the thatch may be
quickly utilized by vegetation re-growing after the fire from root or seed (Brockway et al.
2002). The lingering effects of burning may also increase nitrogen mineralization rates in
the soil due to warmer soil temperatures from a darkened surface (Vogl 1974, Biswell
1989), but this phenomenon has not been corroborated in every study (Blair 1997). The
difference in findings may be because mineralizing microorganisms require water as well
as heat. Romanya et al. (2001) suggest that the effects of burning on nitrogen
mineralization may be seasonal in Mediterranean grasslands, with fire increasing the rate

in late winter and spring.

IX. Conclusions
N. pulchra has been observed to respond positively to fire in several cases in
terms of both basal area and inflorescences. In instances where fire inhibited the growth

of N. pulchra, poor soil may be a contributing factor. I know of no burn studies on A.

21

 

temipes, but other bunchgrasses in the Aristida genus do well with burning in the
southeastern United States. T. caput-medusae and other annual grasses are usually
reduced by burning, while annual forbs usually increase. The effects of fire on soil
properties may contribute to changes in perennial bunchgrass growth and the cover of

different plant functional groups.

X. Literature Cited

Abrams, M. D. 1992. Fire and the development of oak forests. BioScience 42:346-353.

Ahmed, E. O. 1983. Fire ecology of Stipa pulchra in California annual grassland. Doctor
of Philosophy. University of California, Davis, U.S.A.

Anderson, R. C., and E. S. Menges. 1997. Effects of fire on sandhill herbs: nutrients,
mycorrhizae, and biomass allocation. American Journal of Botany 84:938-948.

Barbour, M. G., J. H. Burk, W. D. Pitts, F. S. Gilliam, and M. W. Schwartz. 1999.
Terrestrial Plant Ecology, Third edition. Benjamin/Cummings, Menlo Park, CA.

Bartolome, J. W., and B. Gemmill. 1981. The ecological status of Stipa pulchra
(Poaceae) in California. Madrono 28: 172-184.

Bartolome, J. W., S. E. Klukkert, and W. J. Barry. 1986. Opal phytoliths as evidence for
displacement of native Californian grassland. Madrono 33:217-222.

Biswell, H. H. 1989. Prescribed Burning In California Wildlands Vegetation
Management. University of California Press, Berkeley.

Blair, J. M. 1997. Fire, N availability, and plant response in grasslands: A test of the
transient maxima hypothesis. Ecology 78:2359-2368.

Brockway, D. G., R. G. Gatewood, and R. B. Paris. 2002. Restoring fire as an ecological
process in shortgrass prairie ecosystems: initial effects of prescribed burning

during the dormant and growing seasons. Journal of Environmental Management
65: 135-152.

Clausnitzer, D. W., M. M. Borman, and D. E. Johnson. 1999. Competition between
Elymus elymoides and Taeniatherum caput medusae. Weed Science 47:720-728.

Clements, F. E. 1934. The relict method in dynamic ecology. Journal of Ecology 22:39-
68.

22

D'Antonio, C. M., and P. M. Vitousek. 1992. Biological invasions by exotic grasses, the
grass/fire cycle, and global change. Annual Review of Ecology and Systematics
23:63-87.

DiTomaso, J. M. 2000. Invasive weeds in rangelands: Species, impacts, and management.
Weed Science 48:255-265.

DiTomaso, J. M., G. B. Kyser, and M. S. Hastings. 1999. Prescribed burning for control
of yellow starthistle (Centaurea solstitialis) and enhanced native plant diversity.
Weed Science 47:233-242.

Dyer, A. R. 2002. Burning and grazing management in a California grassland: Effect on
bunchgrass seed viability. Restoration Ecology 10:107-111.

Dyer, A. R., H. C. Fossum, and J. W. Menke. 1996. Emergence and survival of Nassella
pulchra in a California grassland. Madrono 43:316-333.

Dyer, A. R., and K. J. Rice. 1997. Intraspecific and diffuse competition: The response of
Nassella pulchra in a California grassland. Ecological Applications 7 :484-492.

Dyer, A. R., and K. J. Rice. 1999. Effects of competition on resource availability and
growth of a California bunchgrass. Ecology 80:2697-2710.

Engle, D. M., T. G. Bidwell, J. F. Stritzke, and D. Rollins. 1990. Atrazine and burning in
tallgrass prairie infested with praire threeawn. Journal of Range Management
43:424-427.

Evans, R. A., and J. A. Young. 1970. Plant Litter and Establishment of Alien Annual
Weed Species in Rangeland Communities. Weed Science 18:697-703.

Ewing, A. L., and D. M. Engle. 1988. Effects of late summer fire on tallgrass prairie
microclimate and community composition. American Midland Naturalist
120:212-223.

Fossum, H. C. 1990. Effects of prescribed burning and grazing on Stipa pulchra (I-Iitchc.)
seedling emergence and survival. Master of Science. University of California,
Davis, U.S.A.

Fowler, N. L. 1986. The role of competition in plant communities in arid and semiarid
regions. Annual Review of Ecology and Systematics 17:89-110.

Fowler, N. L. 1988. What is a safe site? Neighbor, litter, germination date, and patch
effects. Ecology 69:947-961.

Fowler, N. L. 1990. The effects of competition and environmental heterogeneity on three
coexisting grasses. Journal of Ecology 78:389-402.

23

Hamilton, J. G. 1997. Changing perceptions of pre-European grasslands in California.
Madrono 44:311-333.

Hamilton, J. G., C. Holzapfel, and B. E. Mahall. 1999. Coexistence and interference
between a native perennial grass and non-native annual grasses in California.
Oecologia 121:518-526.

Harrison, A. F. 1985. Effects of environment and management on phosphorous cycling in
terrestrial ecosystems. Journal of Environmental Management 20: 163-179.

Hatch, D. A., J. W. Bartolome, J. S. Fehmi, and D. S. Hillyard. 1999. Effects of burning
and grazing on a coastal California grassland. Restoration Ecology 7 2376-381.

Heady, H. F. 1988. Valley grassland. Pages 491-514 in M. G. Barbour and J. Major,
editors. Terrestrial Vegetation of California. John Wiley & Sons, New York.

Holstein, G. 2001. Pre-agricultural grassland in central California. Madror‘io 48:253-264.

Huenneke, L. F., S. P. Hamburg, R. Koide, H. A. Mooney, and P. M. Vitousek. 1990.
Effects of soil resources on plant invasion and community structure in Californian
serpentine grassland. Ecology 71:478-491.

Huntsinger, L., M. P. McClaran, A. Dennis, and J. W. Bartolome. 1996. Defoliation
response and growth of Nassella pulchra (A. Hitchc.) Barkworth from serpentine
and non-serpentine populations. Madrono 43:46-57.

Janes, E. B. 1967. Botanical composition and productivity in the California annual
grassland in relation to rainfall. M. S. University of California, Berkeley,
Berkeley.

Janssens, F., A. Peeters, J. R. B. Tallowin, J. P. Bakker, R. M. Bekker, F. Fillat, and M. J.
M. Oomes. 1998. Relationship between soil chemical factors and grassland
diversity. Plant and Soil 202269-78.

Kindell, C. E., A. A. Winn, and T. E. Miller. 1996. The effects of surrounding vegetation
and transplant age on the detection of local adaptation in the perennial grass
Aristida stricta. Journal of Ecology 84:745-754.

Knapp, E. E., and K. J. Rice. 1998. Comparison of isozymes and quantitative traits for
evaluating patterns of genetic variation in purple needlegrass (Nassella pulchra).
Conservation Biology 12: 1031-1041.

Kotanen, P. M. 1996. Revegetation following soil disturbance in a California meadow:
the role of propagule supply. Oecologia 108:652-662.

Langstroth, R. P. 1991. Fire and grazing ecology of Stipa pulchra grassland: A field
study at jepson prairie, California. Master of Science. University of California,
Davis, USA.

24

Larson, J. R., and D. A. Duncan. 1982. Annual grassland response to fire retardant and
wildfire. Journal of Range Management 35:700-703.

Major, J ., and W. T. Pyott. 1966. Buried, viable seeds in two California bunchgrass sites
and their bearing on the definition of a flora. Vegetatio 13:253-282.

McKell, C. M., A. M. Wilson, and B. L. Kay. 1962. Effective burning of rangelands
infested with Medusahead. Weeds 10: 125-131.

Menke, J. W. 1992. Grazing and fire management for native perennial grass restoration in
California grasslands. Fremontia 20:22-25.

Mensing, S. A., and R. Byme. 1998. Pre-mission invasion of Erodium cicutarium in .
California. Journal of Biogeography 25:757-762. F

Meyer, M. D., and P. M. Schiffman. 1999. Fire season and mulch reduction in a
California grassland: A comparison of restoration strategies. Madrono 46:25-37. M

 

Morgan, J. W. 1999. Defining grassland fire events and the response of perennial plants
to annual fire in temperate grasslands of south-eastem Australia. Plant Ecology
144: 127-144.

Mulligan, M. K., and L. K. Kirkman. 2002. Burning influences on wiregrass (Aristida
beyrichiana) restoration plantings: Natural seedling recruitment and survival.
Restoration Ecology 10:334-339.

Owensby, C. E., and J. L. Launchbaugh. 1977. Controlling prairie threeawn (Aristida
oligantha Michx.) in central and eastern Kansas with fall burning. Journal of
Range Management 30:337-339.

Parrott, R. T. 1967. A study of wiregrass (Aristida stricta Michx.) with particular
reference to fire. M. A. Duke University.

Parsons, D. J ., and T. J. Stohlgren. 1989. Effects of varying fire regimes on annual
grasslandsin the southern Sierra Nevada of California. Madrono 36: 154-168.

Reynolds, 8. A., J. D. Corbin, and C. M. D'Antonio. 2001. The effects of litter and
temperature on the germination of native and exotic grasses in a coastal California
grassland. Madrofio 48:230-235.

Romanya, J., P. Casals, and V. R. Vallejo. 2001. Short-term effects of fire on soil

nitrogen availability in Mediterranean grasslands and shrublands growing in old
fields. Forest Ecology and Management 147239-53.

Russi, L., P. S. Cocks, and E. H. Roberts. 1992. Seed bank dynamics in a mediterranean
grassland. Journal of Applied Ecology 29:763-771.

25

Schiffman, P. M. 2000. Mammal burrowing, erratic rainfall and the annual lifestyle in the
California prairie: Is it time for a paradigm shift? in J. E. Keeley, M. Baer-
Keeley, and C. J. Fotheringham, editors. 2nd interface between ecology and land
development in California. US. Geological Survey Open-File Report 00-62.

Seastedt, T. R., and R. A. Ramundo. 1990. The influence of fire on below-ground
processes of tallgrass prairie. Pages 99-177 in S. L. Collins and L. L. Wallace,
editors. Fire in North American grasslands. University of Oklahoma Press,
Norman, Oklahoma, USA.

Smith, E. F., and C. E. Owensby. 1973. Effects of fire on true prairie grasslands. Pages 9-
22 in Tall Timbers Fire Ecology Conference No. 12. Tall Timbers Research
Station, Tallahassee, Florida.

Thomthwaite, C. W. 1948. An approach toward a rational classification of climate. The
Geographical Review 38:55-94.

Van Lear, D. H., and J. M. Watt. 1992. The role of fire in oak regeneration. Pages 66-78
in Regeneration: serious problems practical recommendations. USDA General
Technical Report SE-84.

Vogl, R. J. 1974. Effects of fire on grasslands. Pages 139-194 in T. T. Kozlowski and C.
E. Ahlgren, editors. Fire and Ecosystems. Academic Press, New York, New York,
USA.

White, C. S., and S. R. Loftin. 2000. Response of 2 semiarid grasslands to cool-season
prescribed fire. Journal of Range Management 53:52-61.

York, D. 1997. A fire ecology study of a Sierra Nevada foothill basaltic mesa grassland.
Madrono 44:374-383.

Young, J. A. 1992. Ecology and management of Medusahead (Taeniatherum caput-
medusae spp. Asperum [SIMK.] melderis). Great Basin Naturalist 52:245-253.

Young, J. A., R. A. Evans, C. A. Raguse, and J. R. Larson. 1981. Germinable seeds and
periodicity of germination in annual grasslands. I-Iilgardia 49: 1-37.

Young, J. A., R. A. Evans, and J. Robison. 1972. Influence of repeated annual burning on
a Medusahead community. Journal of Range Management 25:372-375.

26

 

Chapter 2
The Effects of Fall Burning on Two Northern
California Perennial Bunchgrass Communities
1. Introduction
Grasslands are a focus of restoration in California due to declining habitat and
reduced abundance of native species, including perennial bunchgrasses. Threats to native
grasslands include invasive species, altered disturbance regimes, and urban ff

encroachment. Since most researchers agree that native perennial grasslands in the state

 

have contracted over the last 200 years and continue to decline (Bartolome et al. 1986, l
Heady 1988), communities of native bunchgrasses that have been invaded by large
numbers of exotics have become prime targets for restoration. In California grasslands,
fire is a component of the historical disturbance regime (Biswell 1989), and so prescribed
burning has been increasingly used in restoration programs that aim to simulate natural
conditions. Several studies have used fire in an attempt to decrease cover of noxious
weeds (McKell et al. 1962, Young et al. 1972, DiTomaso et al. 1999), increase cover of
native species (Meyer and Schiffman 1999), or encourage the growth of perennial grasses
(Ahmed 1983, Fossum 1990, Langstroth 1991). All three of those attributes are
examined in this study within two native perennial bunchgrass communities.

Most research concerning California bunchgrasses has focused on one species,
Nassella pulchra (Purple Needlegrass). Although currently N. pulchra is one of the most
common, many species of bunchgrasses are native to California, and they may not all
respond similarly to burning. This study compares the effects of a fall burn on N. pulchra

and Aristida temipes var. hamulosa (Hook Three-Awn), a warm season bunchgrass. In

27

addition to aboveground vegetation, the seed bank is also examined, since it provides the
propagules for much of the re-vegetation following burning.

California grasslands have been successfully invaded by a wide variety of exotic
plants, most notably annual grasses, that compete with native species for resources
(Heady 1988). Biological invasions are a global problem and California is no exception
(Vitousek et al. 1996), as many areas are nearly completely dominated by exotics. Prior
to European settlement, fires may have burned significant areas in these grasslands, P

which are dry from late spring through fall (Biswell 1989). The general exclusion of fire

 

since that time may have inhibited native species adapted to a short fire return interval. l
Thus, an increase in native species following prescribed burning is expected, and has
been observed in California grasslands (York 1997, Meyer and Schiffman 1999). Fire
has also been shown to increase Shannon-Weiner diversity 25-45% (York 1997,
DiTomaso et al. 1999) and alter the relative cover of different plant functional groups
within one year of burning.

Here, I have divided vegetation into five commonly used functional groups:
annual grasses, perennial grasses, annual forbs, perennial forbs, and legumes. Spring
burning has been found to decrease the cover of annual grasses (Meyer and Schiffman
1999), but fall burns have shown little effect on this functional group (Larson and
Duncan 1982, Parsons and Stohlgren 1989). Perennial grass percent cover has generally
not been observed to change following fire (DiTomaso et al. 1999, Hatch et al. 1999).
However, cover is usually measured with broad classes, and so may not identify small yet
significant changes. Indeed, perennial grass basal area has been found to change

significantly with burning (Ahmed 1983, Langstroth 1991). Forbs and legumes have

28

been shown to increase after fires in both spring and fall (McKell et al. 1962, Parsons and
Stohlgren 1989, Fossum 1990, York 1997, DiTomaso et al. 1999). Most prescribed
burning has been done in the spring to destroy late-maturing target weed species, such as
Centaurea solstitialis (Yellow Star-Thistle), before their seeds mature. However, fall
burns are a more likely time for natural fires since fuels are drier and lightning is more
common (Biswell 1989). Thus prescribed burning in the fall may be a better mimic of
the natural fire regime. Mechanisms for the change observed in the relative covers of
functional groups include removal of the thatch layer and alteration of the seed bank.

Thatch alters surface microclimate conditions, reducing temperature extremes and
light levels, and increasing moisture (Evans and Young 1970, Reynolds et al. 2001).
Thatch accumulates in California grasslands quickly if fire and grazing are excluded.
The presence of the exotic grass Taeniatherum caput-medusae (Medusa Head), increases
thatch accumulation rates substantially because its high silica content renders it slow to
decompose (Figure 3) (Young 1992). The thatch produces conditions favorable for T.
caput-medusae seedlings, thus creating a positive feedback loop that enhances T. caput-
medusae pressure on communities (Young 1992).

The Nassella community in this study has been heavily infested with the exotic
and has a thick thatch layer, but fire removes thatch so that other species may compete
more effectively with T. caput-medusae. McKell et al. (1962) found that burning T.
caput-medusae even after it had dropped seed resulted in a greatly reduced cover of the
exotic the next year, but Young et al. (1972) observed exactly the opposite phenomenon.
The explanation of these differing results may be in the seed bank. The first study had a

significant forb cover before the burn while the second study had almost none. Thus, the

29

36360.??? n 5 0000000 05 .3 03% 0F 0:: 05. .38 5 5508800
20000 Z 05 80.: 303 E 5000 5305 0:0 .8>00 00030055350 EE05030§ 0003009 9:000:20: 0.3. .m 0.59..—

 

 

00>00 0003005¢0m00 H

 

 

n5000 _. axoOw $00 o\oo.v o\oON o\oO
F b _ _ O
H
l
r _. qur
mod N N0 m.
r N U:
W a
9
1 m 1m".
\0}
T v
W
i m

 

 

 

 

30

 

second study may have failed to control T. caput-medusae because there was nothing in
the seed bank to take its place.

Burning reduces the seed bank by destroying seeds. Grassland fires are rarely hot
enough to destroy seeds below the surface (Vogl 1974), but seeds at or above the surface
may be killed. Grasses with papery glumes are especially likely to burn and this may be
why annual grass cover is often reduced following fire. By fall, the seeds of most species
have dispersed and lie on the surface or are suspended in the thatch layer (Young et al.
1981). Consequently, fall burning is likely to destroy the seeds of many species, but
especially those seeds that are suspended in thatch. Soil is a good insulator, and so seeds
that are covered by even a thin layer of soil may escape destruction by fire (DeBano et al.
1998). Thus, species with seeds that persist for years underground or exhibit hygroscopic
boring may increase in cover the year following burning due to reduced competition.
Legumes have a high year-to-year seed bank carryover in Mediterranean grasslands
(Young et al. 1981, Russi et al. 1992) and Erodium sp. (filaree) are hygroscopic borers
(Stamp 1989), so these vegetation types would be likely to increase following a fall burn.
A seed possessing a long awn such as T. caput-medusae, but no hygroscopic boring
capabilities, may be more likely to become ensnared in the thatch and be destroyed by
fire.

Warm season burning has been reported to both increase (Ahmed 1983) and
decrease (Langstroth 1991) the basal area and inflorescences of N. pulchra. The soil in
the latter study had a shallow hardpan layer that decreased soil fertility and may have
contributed to the different results. Burned plants have produced seeds that are 14-20%

heavier (Langstroth 1991, Dyer 2002). Investigators have demonstrated that burning

31

promotes N. pulchra germination (Dyer et al. 1996) and also early seedling growth by
removing thatch (Ahmed 1983). However, the pulse of seedlings documented by Dyer et
al. (1996) experienced high mortality and virtually no seedlings were still alive 4 years
after the beginning of the experiment. N. pulchra is a species that can thrive with a short
fire return interval (Hamilton 1997).

I know of no studies that have examined the effects of fire on A. temipes, but A.
stricta and A. longiseta in longleaf pine communities in the southeast benefit from fire
(Parrott 1967, Anderson and Menges 1997) and thatch removal (Fowler 1986, 1988). In
California, A. temipes is a C4 grass that continues to grow during the dry summer months
after N. pulchra has senesced. However, both species have senesced by autumn when
burning occurred in this study.

The deep thatch in the Nassella community of this project traps numerous T.
caput-medusae seed, which is likely to be consumed by the fire. As a result, I expect T.
caput-medusae to be reduced in the seed bank and in the following seasons botanical
cover. Most previous studies agree that annual grasses decrease and annual forbs
increase as a result of fire, and there is no reason not to expect those results in this case as
well. Likewise, Shannon-Weiner diversity is likely to increase, and native species are
expected to increase as percent of botanical cover.

The N. pulchra population in this project occurs on soil underlain by a hard
duripan similar to the Langstroth (1991) study, which found that fire decreased
bunchgrass basal area and inflorescence output the following year. Thus, I expect a
decrease in N. pulchra basal area and inflorescences. Although other bunchgrasses in the

Aristida genus do well with frequent burning, the A. term'pes population in this project

32

 

may not experience the same results. The soil under the population is underlain by a
shallow layer of cemented cobble and gravel, which may inhibit the bunchgrasses

similarly to N. pulchra under the same conditions.

11. Methods
1. Study Sites

Both study sites are in the northeastern Central Valley (Figure 1). The Nassella
community is situated on Anita Clay Loam at 39° 18’ N and 121° 59’ W, 20 km
northwest of Chico, where mean annual precipitation is 670 mm. Precipitation in the
period from July to June was 82% and 83% of average for the periods ending in 2001 and
2002, respectively (Chico University Farm, National Climatic Data Center,
http://lwf.ncdc.noaa. gov). The Nassella community has not been grazed by livestock
since 1986 or earlier, but is open to grazing by wild animals. The Aristida community is
on a 26% grade of Tuscan cobbly loam at 40° 06’ N and 122° 02’ W, 19 km southeast of
Red Bluff, where mean annual precipitation is 566 mm. Precipitation in the
aforementioned periods was 94% and 80% of average (Red Bluff Municipal Airport,
National Climatic Data Center). The Aristida community was grazed by cattle during the
experiment and for many years previously, but the stocking rate during the experiment
was low, just 0.07 head per acre. Cattle were put on the range around mid-December,
and removed in late April.
2. Design

In March of 2001, 1-m2 plots were established in both communities on four

parallel transects within populations of the two bunchgrass species (Figure 4). In the

33

 

 

.‘l
i. 'h.

Nassella community plot layout

 

        
  

     

  

 
      

    
   

  
 
 
 

   

  
 

  

      
  
    
    
    
  
     
 
  
   

    
      

     

      

\....11 ..1. ..-..1 1.1I.1‘1..1..1“ .1‘“ .1“ .1"‘ ..11 ..11“ ‘11 1.1 11“
11:11:“. ““‘“ .1.11““.““““ .11‘.‘“‘ ..1“: 11““. “‘ 1““ 1““ .1“ “‘ ““ .11“ . “ “
‘11‘0MNNV‘WW“M‘WWM~“
11111“ .‘1 .‘1‘1 .1 " 1‘
.11“‘\‘ ..11‘ . .1““1 ‘1‘1 1‘“ “ 11‘“ 1‘“ “ ‘
. “..1“... ..1. ..1..1.‘... .1 ..1'1. 1.
..\\\\\1“t”..11‘1““....1111“....11“.‘...1.1““.‘ “““‘.‘.1 1..1“““.1““..1““‘.‘.“‘“ “ .
\1‘““‘ “‘1“... .... ."1‘..1“““‘. 111‘“ ..1... .. .11.“ ”..1...“ ..1.” 11“\ ...1 ..1 . .1‘ .
.1. . .l“ ..1 11‘ “
‘11‘11‘11"“““‘111 ““‘““‘ “““‘ “‘ “11‘1“ “““ -‘1‘"““ “

 

.. .‘1““ 1.11‘ ..1

““‘“‘“..‘.‘.1“‘.‘““‘..‘11“.“.‘.‘
11“‘ .11“‘ . .

..1‘...111“1.‘.1“ ‘1“ ‘1
.. 11“ ‘ ‘

.1“‘1

1
MNWWW
.11“ “.11“
1“ .1111 1“ ..
.1l11“““

    
 

  

  
            

 

1““..‘111‘1‘ “‘

““.‘:‘.:1‘1‘l‘..11““:.11‘ ....:“-‘ “““‘ 1111‘“ ....1" ..1‘ “ “.‘ .11“ .‘1 .11
1 ‘ ..‘1 .
11““ .\.111““‘ 111‘" ‘ 1““: .1111" .11" 1“ ..1...” “‘.11“‘

11
1111 .11“ ..1‘1““““.1 1.11;“. MN“ 1“““
~11.‘1‘.1 ‘1“‘

.1 ‘1‘. .1. -11 . 1‘ “.1“

‘1‘.“. 1“ .

 

 

.“..1

     
 
 

““11.““““.1‘
....11 ..11 .11““““““ :.\

      
   

m
m
ml]

    

    

.1

.‘1111 1‘ *1
11“ .1111 1“ ..111““ ..1 1
.... .“..1 11..1 ..

...11 ..111‘
1.1:“.1 . ..‘ “
.1

 

1‘1‘ \‘ ..1.» 1“ 11
1‘““ 1“ .1.1‘ 11“
11 ..1.‘

.11‘“
1‘ ““.‘.11“‘ .1.
“‘ ““‘-‘ 11“I“‘ 11‘

...:1 .13.. .1 11 11

 

 

 

 

Aristida community plot layout

1111‘" .. 111.1‘11111‘11‘ 1111‘ 11 ..1‘11 1111‘
.‘.‘.‘1.‘.‘.‘1‘““““‘ ‘2“. .. ..111‘11“ ...1.111““......1.1\“““‘ ““11”“: ....\\“““{..1111‘1“.\.\““ N111“

“11.111111“ .1.1.1‘1\“....111‘.1‘.1...“““......

 

1111111“

  
  

      

  

       
  
       

   
 
   

..11111““‘ .“‘1.1‘

...1.11‘1“‘
1.1.1.1“

    
 

   
 

 

  
          
 

“ ““““ 1111‘“ “ 1.1.1111
..11111111‘3“ 1111‘ 1‘
.1‘“ ““1 .1\ \1‘1‘3‘1

. ...
|...“.‘.1‘1l“.‘. \11“‘ ..1.
.‘.11 ... .‘..
....‘. 1 ““‘

11111‘1‘1‘“““‘ 11
....1.1.1.1‘ ““‘..‘.....11“““
.1111“1““‘.....‘.1111

11“
........1111\‘ “‘“
1

11111“
I_l_l_l_l_l_11111 ..1... .1111 ....1111.111111 11“.

1““ ‘ .11111“““‘.1111“““

\1\
..1\‘ “‘““

...1...111

11
1““ ““‘

111‘“‘.\.‘1 ., ... . ..11‘.
.1..1‘11\11‘......‘..1 ‘ .... ..111 .
111“‘.‘ 111.11 ‘..““‘“ ..1}. \.. ‘
. 1"....11‘ . ‘.1.“‘ ..1' .11: ....1.1 .1111 ..11. ..1 ”1.1.1 W111“ ‘ “‘1‘...‘.....‘1“.11 ....‘."“‘|111\l\11 ‘. M... h... A “m...“ .1“ M. ‘\
.“1‘..1“‘ 11:.‘ 1.11.111...“ ..1‘.‘ ......11 ..1.‘ 1. . ‘ ‘

 

     

 

  
 
 

11‘
11. “.1“

 
      

 

.1“ ..l‘ ..‘1 "' .1111“‘ ‘ . “1‘“ “‘ 1.1“‘I ' . ‘
..‘\““ \1““““.1 “‘ ...1““ ‘..1.1““‘.“.. 1"“.‘.1.‘..“1‘ ...1““‘ ‘ ..... “

‘ 1‘.“ ““1“ 1‘“ “‘l‘“““ 1““ 11“ “‘ 11‘1“““ 1111‘ ““‘ 1111““ ‘1“
1““ 11“ \l“ 1‘“ ll‘ 11 1‘ \ll‘ . 1 .1‘1' 1.11 1‘1““ 1 11““‘
11‘.“ “..1 .111“‘ .11.“: ..1.......1......111 1‘10““ 1.11 ...1. 11 1111“.“.“ .1‘ “M““‘ .....1.111 1. “11““

I‘ \‘ll ‘1 .
1.1““..1.‘ “..‘ ““ .....1

     

.11‘1‘1“.‘

        
 
   
 

 
   

         

     
   
    
  

 

 

    
      
 

. 1
._..

 
 

 

..1‘1 . .1‘
1‘ 1.11.1.1
..1..1‘1‘.““.“ . ..11111

   

 
      

     

.1‘11'111
...1. ..11111‘

«1111‘ .. 111‘
....1111-
..11‘.. 1.1.\“ “‘111.1“‘11“ . .1...

..1‘
“‘“‘. 11‘

11“
..‘..11 \
..11
..111 1‘“ ...11

.‘1‘ 1“ 11““
1‘.‘“.‘1‘l“1“. ..1‘1‘"“-. 111““ ..11“‘ .11“
“ l‘

“1“1‘1‘
1. “‘.1 1.12. .

 

. 1

 

 

 

Figure 4. Plot layouts in both communities. White areas are control, grey areas were
burned in October, 2001. Black squares represent 1 m2 plots.

34

Aristida community, two rectangles covering the ends of two transects and each
containing a quarter of the 48 total plots were designated as the prescribed burn area and
two similar rectangles served as controls. One control plot was later discarded due to
excessive trampling during the burn. In the Nassella community, a square covering one
end of all four transects and containing half of the 80 total plots was designated as the
prescribed burn area; the rest served as a control. Randomization of the burn treatment
with regard to the plots was not an option due to constraints on the number of fires that .'r

could be lit. However, burn and control areas were chosen for their similarity, which is

 

evident in their pre-burn botanical composition (Table 2). ..

Data on 39 Aristida plants were taken in April of 2001 and 2002, and 210
Nassella plants were measured in May of 2001 and 2002. In each plot the basal
circumference and number of inflorescences for individuals of the two bunchgrass
species was recorded. Individuals that were highly fragmented or were near the comers
of the plot were not used. Inflorescence counts for A. ternipes were taken early in the
annual reproductive cycle of the species, and do not represent total annual inflorescences.
The cover of all species present in the plots and the cover of bare ground was measured
using the Daubenrnire (1968) cover class system. Botanical nomenclature follows
Hickman (1993). The botanical cover data were used to calculate the Shannon-Weiner
diversity, an index that is comparable across different spatial scales due to the fact that
the area sampled is not used in the equation. The depth of senesced thatch was measured
4 times in each plot, 20 cm from each corner with a ruler.

In both communities, cover of different plant functional groups can vary annually,

as is the case with all California grasslands. Variability may be driven by differences in

35

.Ammvcmms. m_ 98

.23 5 9m. 28> :95 m .2 «858:6 God v 3 .52:ch

.355658 58 .2 .98 956 .3285... .o mmcoammm .u 03:...

 

 

 

 

 

 

 

 

36

A83 8.: at: 8.? at: 8.8 8.: 8.8 9526 98
a; 24. 60.: 88 8.8 5.8 8.3 9.». 2 8:584
A89 3.? See cod 8.9 8; 8.9 on; an: 8896 .9598
6:8 28. 8.3 8; 8.9 8.8 $98 2.» EV 8.8 .2298
and. 8.... sad «ha? 899 8.: 8.: «Zr 63 8890322
8188.2 3.8 8.». 63¢ K8 8.3 8.8 as «eon. 15:5
5.3 2.2 8.3 ts 5.3 8.8 8.8 8.8 8688 268
3.8 84... 25.8 a: 8.3 3.8 $th 2.8 898% 9,82

3.5558 unamt<
5.: 8.3 85% 8.8. Ewe 5.8 8.9 a; 1 2:90 2mm
48.9 8; 8.9 8.~. 8.3 8.. 5.: 8.... 2 8:584
88.: 8s- E: 8.? 68 8.8 88.8 8.8 5% 889655.28
5.3 mud 8138.0. awe mam 28.9 :8 mi 85“. .9229".
:89 5.8- 8.3 88 8.3 8.8 8.9 9.3. as 8839522
5.3 8.8 6.8 8.8. 5.3 2.8 5.3 2.8 :3 mean. .852
8.8 m». T 3.8 mm. F P- 3.9 8.8 3.3 8.8 928mm 268
A8. 5 8s- :53 8.8. cm. s 8. 8 893 8. 5 8688 9,82

3.55500 .5832
Sam .82 Sam 02 Sam 28am Eam oz

g 8:28 .28 g .250 .25..

both total annual precipitation and its monthly distribution. In order to quantify this
natural annual variation, botanical monitoring data were obtained from the Dye Creek
Ranch from 1997 to 2002. The author collected the Dye Creek data for the two most
recent years.

To measure the seed bank in the Nassella community, 8 soil samples were taken
pre-burn in both treatment and control areas an hour before the prescribed fire was lit. In
the treatment area, 8 additional samples were taken 30 min after the fire was extinguished
within 30 cm of the spots that had been sampled pre-burn. Soil samples were taken

outside of the quadrats used for botanical composition data. The top 5 cm of soil, as well

 

as any over-lying thatch, was removed with a 7.14 cm diameter soil auger. The samples
were weighed with the largest rocks removed, and the data are expressed as seeds per
gram of soil. The seed bank was estimated by a direct germination method similar to
Gross (1990). Soil samples were stored at room temperature for not more than three
months following collection. The samples were spread on top of potting soil to a depth
of 0.5-1.0 cm and kept moist in a greenhouse set to 70° F at night and 80° F during the
day. As seedlings emerged they were identified, recorded, and removed. Seedlings that
were not identifiable were allowed to grow until flowering and were then identified.
After 48 (1 very few new seeds were germinating and the seed bank experiment was
terminated. Seed bank data from the Aristida community were compromised by
approximately 1.3 cm of fall rain that caused seeds to germinate before samples were
taken. Thus, seed bank data for the Aristida community are not presented.

The Nassella community was burned on 10 October 2001 when temperature was

83° F, relative humidity was 21%, and the wind was 5.5 km per h. The Aristida

37

community was burned on 22 October 2001 when temperature was 62° F, relative
humidity was 53%, and the wind was calm. Peak fire temperature was measured at the
soil surface with temperature pellets (Omega Engineering Inc., Stamford, CT), and
ranged between ZOO-300° C for both burns.

In the spring of 2002, seeds were collected from Nassella plants for a weight and
germination experiment. Seeds were individually weighed and then kept moist in potting
soil in a greenhouse set to 70° F at night and 80° during the day. Gerrninated seeds were
recorded and removed. After 3 weeks no new seeds were germinating and the experiment
was terminated. Seed weights and germination rates per plant were calculated before
statistical analysis. Seed data are not presented for Aristida plants, since their seeds
matured too late for collection.

3. Analysis

All statistical analyses were performed with SYSTAT Version 10 (SPSS Inc.
2000). Inflorescence counts, species and functional group cover data, seed bank data,
and germination rates were not normally distributed, and could not be successfully
transformed, so Mann-Whitney U-tests were used to evaluate the changes in these
variables. Seed weights and Shannon-Weiner diversity were analyzed with t-tests. Chi-
square tests were used for bunchgrass mortality. Preliminary data analysis showed that
average N. pulchra basal area in the treatment area was somewhat larger than in the
control area prior to burning. To remedy this, individuals of similar size in control and
burned areas were paired, and a paired t-test was used on the change in basal area. A t-

test was used for analysis of A. temipes change in basal area. A modified Bonferroni

38

correction was used to adjust p-values in the species and functional groups sets of

comparisons in both communities.

III. Results

The two bunchgrass species responded differently to fire. Burned N. pulchra
increased in basal area by an average of 7.7 cm2, while unburned plants increased by only
1.5 cm2 (Figure 5). In contrast, both burned and unburned A. temipes experienced a
decrease in basal area. Burning also increased the number of inflorescences on N.

pulchra plants, while it had no effect on the inflorescences of A. temipes (Figure 6).

 

Mortality was much higher among A. temipes, even for unburned controls. 38% of
burned A. temipes died, compared to only 4.5% of unburned controls (p < 0.01) (Figure
7). There was no significant difference in mortality among N. pulchra, where 0.6% of
burned plants died, while mortality of unburned controls was 2.4%. Seed from burned N.
pulchra plants was found to be significantly heavier. Seed from unburned controls had a
mean weight of 0.01 g per seed, while seed from burned plants was 8% heavier (p <
0.01). Figure 8 shows that although seed weight affects germination rate, the effect
above =0.009 g per seed is minimal and the germination rate levels off at about 95%. As
a result, there was no significant difference in germination rate between burned and
control N. pulchra plants.

Significant cover changes were detected in several functional groups (Table 2),
but burning did not significantly affect the total cover of native and exotic species in
either community. The cover of annual forbs increased in burn plots from 19.1-25.8%

relative to controls. The largest variation in annual forbs in the Dye Creek monitoring

39

 

down—2: 8: 203 3% 85 226365 .535 m_ 8.5 c8985 .58 89080
E 3:98 83 8258: EB 05. Noon $ch 8 Son wctmm Set :83 Ba 3.8 .83 E omSEo :32 .m «._.—awn

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.8858 Quanta». 33883 58832
cm.
.. m7 3
85.5 E W
-. 63:00 D I or- fi
m.
r m- Won
W.
-. _ _ o W
9
Du
I m \nw/
wz
- or

 

mZ

 

 

 

 

 

E?
L

40

 

.5505 mm Hobo Eucafim .38 82200 E 3:38
83 8253.: EB 2F .88 meta. 8 SON wanna Bot Ema 8m 80:888GE mo “38:: E owaso :32 .9 95»:—

 

8853 Sonatax 8.333% 58832

 

 

N
I

mz :58 v a

I
F
I

 

H
O

 

 

nmesm a
6:80 B

 

 

 

 

 

iue1d 19d soouoosoiogul ut 9811qu)

 

 

l
(0 L0 V CO N

 

41

 

.32 833-20 « FEB noun—3030 33
cacao—baa .5505 2 hobo Egg—Sum doom 5 mommfiwnogn 36365 8“ 23 3:358 2t. .5 «._.-ur—

 

 

 

 

 

 

 

 

 

 

wmmfitfi $33.24* GEQNVS. 53362
85 v a _ mz

H w 1 m N
T or W

I 9
W.
I ON am.
1 mN H
, om m
- mm 9
85.5 I )
055 n. I ow {ko

_ o - mv

 

 

 

 

om

 

42

 

 

.320 36 @2550 95 55m? :8 €83 35:8 ES 353 Eon Eat 38 £33 833: woo—68m
353 BED :w:o::< .9505 mm Hobo Edvcfim .23 :oumcgow QEQEQ 3:332 :0 Emma? 38 we 80th .w 0.53,.—

 

 

Aw ii 330% Bum
«.7 «.7: 2-3 0.3.0 m.o-m.o mdfid n.o-o.o odv
‘ x - o

I S
1 cm
- om
, ow
,. om
I oo
1 on
- om
[ om

uopeuyuueg) mean-3d

 

 

 

 

cor

 

 

43

data was about 14% (Figure 9). Burning significantly decreased annual grasses in the
Aristida community by 13.6%. The largest variation in the Dye Creek monitoring data
was about 21%. The cover of annual grasses in burned plots also dropped sharply
relative to controls in the Nassella community, but large variability resulted in non-
significance. Burning also tended to increase the cover of legumes and decrease the
cover of perennial grasses in both communities, but neither result was statistically
significant. The reason that cover of perennial grasses in the burned Nassella community
decreased, while the average basal area of burned individuals increased is that the burned
plants had shorter leaves. Cover was higher in the unburned plots where the longer
leaves of the plants covered more area but did not contribute to basal area any more than
short leaves.

Several species from both communities experienced significant changes in cover
following the burn (Table 3, Table 4). Erodium brachycarpum and E. cicutarium
increased due to burning in both sites, but the result was only significant in the Nassella
community. Some other exotic annual forbs also increased significantly in the Nassella
community, leading to the overall increase in the annual forb functional group. Two
species declined in cover significantly due to burning, Selaginella hansenii and T. caput-
medusae. Fire had a positive effect on overall Shannon-Weiner diversity in the Nassella
community, but it had no effect on the native component of diversity alone (Figure 10).
Thus, the Shannon-Weiner index was increased by greater diversity among exotic
species, not natives. The Aristida community exhibited the same pattern, except burning

had only a near-significant (p = 0.07) effect on overall diversity. In both communities,

44

 

 

commune 3335 x080 can can: boom mo :23?qu

.3 cum: San— dor—Sm 05 .3 380:8 203 NOON 98 Bow H8 83 ”88on wccozcoE 32:33 x85 08 86.83850
22a Z 2.: Bot 3:850 203 «an .3283 33me 3.823 o: 803 82$. $32 2 use :2? 2 :85?— 382—00

803 3% .830 .88 I 32 Bob chomp—m x080 can £25 .320 05 a 8:on 3.55:8 :83 .58 “0 ~98 2F 5 Bantu

 

NOON room OOON mam? mam? hmmw

 

 

 

 

 

 

 

o

.. m
- 2 mm
mezmmjlll m
was". .mEcmEQ ......< I 9 m
1
3.0“. _m:::<\.u.\ .. ON 3
$3.80 .m:::<|6|| nAU
T mm m

. om

 

 

 

 

 

 

45

 

60000:... .0 9850:. «50003.3 ._0

Eu >05 new 2026.. 39...? 290000 2 0.30.050 Sm >05 00:82. $50.00. 80:95 0.52. mEEmEoEcQQ new

5.0.0.0..- .E:....< 05:3 0.... .o 23:55. ._. 030... 0. amazon Em m:o_.m_>2nnm 039m .2285“. .23 c. 20 .09. :96
m .0. 08:20:... God v 0. Emgzcgw ..mwvcmos. 0. San. 5.02.508 0.000002 05 c. .98 00.800 .0 00:003.. .9 03¢...

 

 

 

 

 

 

.000. 00.0 .000. 00.9 .000. 00.0 60.0. 00.0 0< 0500.00 .08 00.00.00 0.083009
.000. 00.0 .000. 9.0 80.0. 00.0 .000. 00.0 -_ 5202090. 500.0000
.90. 00.0 . F 9.0. 00.0 .000. 9.0 .000. 9.0 4 8200.050 00> 5280000000 5000009
.90. 9.0- .000. 90.9 5.0. 00.0 .000. 8.0 w< 20008.06 «8.006
.000. 00.0 .000. 00.0 80.0. 00.0 .000. 00.0 “2 0.0.000 00.0mm
.80. 00.0 .000. 00.0- 60.0. 00.0 80.0. 00.0 0< 0.09.... 00300.0
.000. 00.0- .000. 00.0 .000. 9.0 .000. 00.0 0< 0:00.050: 00580.0.fl
.00.: 00.0- E... 00...- .90. 00.00 60.0. 00.00 .00 800.00 0000002
.000. 00.0 .90. 00.0 .000. 00.0 .90. 3.0 -_ 00:00 000.03
.000. 00.0 .000. 00.0- .000. 00.0 .000. 00.0 0< .5859. 0:3
.90. 00.0 80.0. 00.0 80.0. 00.0 .000. 00.0 0< Es... 0.50050:
80.0. 00.0 80.0. 00.0- .90. 00.0 01.0. 00.0 ".0 080008 500000.”.
.90. 9.0- .90. 00.0- .9.0. 00.0 5.0. 00.0 0< 0300.08 00000859.“.
.000. 9.0 .90. 3.0 .000. 00.0 60.0. 00.0 0< 508.3 50.8.3
.000. 00.0 .000. 5.0- 80.0. 00.0 .000. 3.0 00 0000.900 08580.80
.000. 9.0- c 9.0. 9.0- .90. 00.0 .90. 00.0 00 55000.09 .08 5500000.. 50.50030
.90. 5.0- .000. 00.0 .000. 5.0 .000. 00.0 0< Emcsfiomao .000 $3050 0000.0
80.0. 00.0- .000. 00.0 60.0. 00.0 60.0. 00.0 0< 0000000000 00.500000
.90. 9.0 .30. 9.0 c 9.0. 9.0 5.0. 00.0 0< 0.00500 00008000
.90. 9.0 5.0. 9.0.. .90. 00.0 5.0. 000 00 mEEm.8.0co.0\m.0.0EtE§<
.30. 00.0- 50. 000 .90. 00.0 .90. 00.0 0< 0.0.05 0000000520
$22.31 .2502

Sam .33 Sam oz Sam 20qu 5.5 oz mac-.0 3.0on 02:02

30. 09.2.0 .960 $8 .030 .03.... .05.“.

 

46

 

 

 

 

 

 

5.9 8.8- .89 5.0 60.9 8.8 85.9 8.00 9. 008005.330 59050303
02.030: 9.8%

:39 8.0 80.9 00.0 60.9 8; 80.9 8: “E 088 800000
8.9 8.0 8.9 8.0 80.9 8.0 8.9 8.0 4 59800895 593$
:09 8.0 30.9 8.0.- 60.9 00.. at: 00.0 ._ 50:9 50009»
8.9 8.0 30.9 2.0 80.9 00.0 8.9 00.0 ._ 50.5% 50.505
$9 3.0 5.9 2.0 8.9 8.0 5.9 5.0 “E 0.08.3. 0.600%
80.9 00.0 60.9 9.0- 8.9 8.0 89 8.0 .. 882508 0089002
5.9 8.0 80.9 8.0 83.9 00; 80.9 5.0 0< 8900208 $0.03
8.9 8.0 5.9 5.0 8.9 8.0 8.9 8.0 o< 009093 0.098:
8.9 3.0 80.9 2.0 5.9 8.0 5.9 8.0 ”2 $050 0.200000%:
80.9 90 8.9 00.0 80.9 00.0 000.9 00.0 “_< $980890 50.0860
:09 3.0- 83.9 8.0 80.9 8.0 80.9 2.0 “2 0.90.0098 003908
:09 5.0- 60.9 00.0 60.9 8.0 8.9 8.0 o< 8.8% d8 0.82.805 850$
90.: 8.0 60.: 30.7 8.9 3.0 20.: 3.0 .04. 8800900 859m
5.9 00.0- 8.9 8.0- 5.9 00.0 8.9 8.0 w< 09389;.
3:233. .8282

80.: 8.0 30.9 3.0. $09 8.0 60. s 8.0 w< use“: .5. 85? 99§
30.9 8; 80.9 2.0 3.9 2.0 8.9 8.0 “2 0.500 maggot
60.9 8." 30.9 8.0 8.9 8.0 60.9 8.0 “E 8.900. 000E
3.9 3.0 83.9 2.0. 80.9 8., 80.9 00.0 “2 50.20390 8909.0
5.: 00..- an: 00.0.. E: 8.3 E: 6...; ”2 50889005 Sufism
3:032“. «>230.

3.025 ozoxLMI

6.0009 0 0.00»

47

 

02.00.0980; .09 £00 0.0 >05 0:0 .00:0000:00 00:0 0.0.0000 o. 0.00.50
0.0 >05 0030000 005000. 000005 0.03 0520.5 .000 05:00.0 30039:... 0:00 0.00:0:0 0.0.0.000
.200 c. 0.0 00.00% .0 .0: 3208 E05 02.02 0.50500 .20: c. 0.0 .00.. :020 0 .0. 000:0.0E0
.modvav 200.935 $950.2 0. 0.00 03:05:80 00.00.:< 05 c. .050 00.0000 .0 00:800.“. .0 0.00...

 

 

 

 

 

 

 

80.9 8.0 9:9 00.0 9:9 :00 .009 00.0 ”9. 000000000: 00000002
.009 00.0 .0 :9 00.0 .009 8.0 .009 00.0 “2 000.0000 0.000005
.009 0:0 .00.: 00.0 8:9 0:0 8:9 0:0 "2 0.00.5000 000000.:
90.9 00.0 .009 0:0 6:9 0:0 309 00.0 02 000.050.0000 000000.:
9:9 .00 90.9 00.0 .009 00.. .009 00.0 ._ 00000000000 00.0..
.009 8. 7 .89 00.0. 20.9 8.0 9.0.9 00.. “9. 000.0050: 0.00.0000
5.9 00.0 90.9 00.. 9:9 000 9:9 000 “E 500.00 80.0.0000
.009 00.0 .009 00.0 9:9 .00 9:9 .00 "2 0.000500 0.30..
00:9 0:0 80.9 00.0 .009 00.0 60.9 8.0 02 000000.00 0.005000
90.9 90.0 c :9 :0 .009 8.0 80.9 8.0 ...< 00.0000 000000
90.9 03 9:9 00.0 9:9 .00 90.9 00.0 0< .0090 0000.050:
.89 00.0 90. c 8. .- 909 0:0 9: 5 R9 ...< 0.00.5 0.00
60.9 8.0 .89 8.0 .0 :9 0:0 .009 00.0 00 E000: 500000.00
20.9 8.0 90.9 90.0- .009 00.. .009 00.. 0< 00009000 0080008000
.009 8.0 .0 :9 F :0- .009 00.0 90.9 00.0 02 000000000000 .000 00000000 0000.0
9:9 .00 60.9 8.0 6:9 0:0 .009 8.0 02 00000000000 050000000
.009 00.0 .009 0:0 .009 00.0 .009 00.0 “9. 0500.00 .900000000 0000000
9:9 00. 0. 9:9 00. .- 909 00.0 .009 00.. ”2 0.00000 0000000000
400.9 8.0 .89 8.0 .89 00.. .009 8.0 02 000500 0000000000
3:9 0:0 809 00.0 .009 8.0 .89 8.0 ._ 0000000800 00.00000:
90.9 90.0- .009 00.0 .009 00.. 90.9 00.0 on. 009.00. 00.00::
.0 : c 00.0 .009 00.0 90.9 00.. 90.9 R. 0.0 0EE0.8.000..000..000§00...v.
9:9 .00- 80.9 8.0 9:9 .00 80.9 8.0 02 0.00.0 0000000090...
00:00.00: .8302
90.: 0.0.0 .009 :0. 9:9 .00 . 00.9 00.0 0< 0000000 0.000000
00:00—00: 03:000.
53m 005‘ 5.5 02 Sam 0.20m Sam 02 0:20 macaw 02.0."...
30. 09.2.0 .030 $0. .030 .0.=:. .0500.

48

 

 

 

 

 

 

 

.009 00.0- 000.9 8.0 .009 00.0 .00.9 8.0 .9. 0 .000 50.505 5000.0:
..09 .00 .0 :9 ...0 90.9 00.0 .0..9 09.0 n... 0000.00.00. 50.000
.009 00.. .009 00.0 .009 0 ... .009 00.0 0... 000.00 000......
.009 00.. 90.9 09.0 .009 00.0 .009 00.0 02 5000.000 50.09.”.
9 ..9 00.0. .0 .. .. 8.. ..0. .. 00.9 .0: .. 00.9 02 5050000050 50.050
.00.9 00.0 .0 :9 00.0- .009 8.0 .0 :9 00.0 0.... 0.0000000 00.00500
.00.9 00.0 . . :9 ...0- .009 00.0 9:9 .00 “.< 0.000...05 00.00500
.009 00.0 .009 00.. .009 00.0 .009 90.. 0.. .. .000 0.0005005 0050.0
.00... .0.0- .00... 0:0 .009 00.0 ..0... 00.9 9.. 0000000000 0050.0
.009 00.0- .009 8.0 .009 00.. .009 00.. 9. 0.0500 0000:
.0..9 09.0- .. ..9 ...0- .0..9 09.0 .0..9 0:0 0.9. 0.000.:0 0.0.0005.
9.9 .00 .009 8.0 .009 8.0 .00.9 00.0 0.. 00.00.000.00 0...:
00.50.00”. .8502

00.00%. 0:03

.000. 00.0. .009 00.. .000. .00. ..00. 0:0 0.0 000000.. 0000.00.00
00.5.0002 02.0%

9.. .. .00- .00. .. 00.0 .0: .. .00 ..0. c 00.9 0.. 0500.00.05 0.0.00
.009 0:0- .0 ..9 ...0. .009 00.0 .0..9 .00 0 50.00.008.05 5000.00
..09 00.0 .009 8.0 .0..9 .00 .0..9 09.0 -_ 5950000000 500050
..09 .00 .0..9 00.0 .0..9 09.0 .009 00.0 0 5000500090 5000......
.0..9 0:0 .. ..9 ...0 .89 8.0 .00.9 8.0 02 0.0.000 00.000
.009 00.0 .0..9 00.0 .009 8.0 .009 00.0 n... 0000.00.00. 00000000...
.0..9 0 ..0- .0 ..9 00.0- . .09 00.0 .0..9 09.0 0< 00.0030... 500000.00
.00. .. 09.9- ..0. .. 0:0- .80. 00... .0: .. 00. .. 0:. 0.00.0 000.00....
.0..9 0:0. .009 8.0 9:9 .00 .89 8.0 0... 0.00.0.0 00050.0.
.0..9 0:0 .009 8.0 9.9 .00 .009 8.0 "2. 005.000 005000.000.
.0: .. .00. .00. .. 00..- .00. .. 0:0 .0..9 0:9 0< 00.0000. 0.0050002
.00. .. 00.0 .009 00. .. .009 3.0 .009 09.9 “.9. 000000000. 0.0050002

.0509 9 0500

 

.00. .. 00.0 .80. 8.0 .80. 8.0 .80. 8.0 o< 000005 0.0.00
.80. 8.0 .0 ..0. 8.0 .80. 8.0 .0 ..0. 0 ..0 0 500... 5000.00
.0..9 0:0- .80. 8.0 .0..9 0:0 .80. 8.0 ._ 50.000 5000000
.0 ..0. 8.0 .80. 8.0 .0..9 0:0 .0..9 0:0 0... 000000 0.....00
.30. .00- .. ..0. ..0 .30. .00 .80. 8.0 0.0 00....00 0000.0
.80. 8.0 .0..9 00.0 .80. 8.0 .80. 8.0 09. 0000.00 0.00000
0. ..0. .00- .00. .. .90- .00. .. 8.0. .90... 00.0. 0< 0.000 0.0050000
.0 .. .. 00.0 .00. .. 00.0 .000. 00.. .000. 00. . ._ 00000.0 0000.003.
.00. .. 00.0- .000. 00. .. .000. 0 ..0 .000. 00. . 9. 5.000.005 5000-.
.009 00.0 .0 ..0. 00.0 .000. 00.0 .000. 8.0 0< 0000.050 0.00.000.
..00. 00.0 .009 00.0 ..00. 00.. .000. 09.0 “2 0.00.0 0.0000800...

.0008. 9 0000

50

.58 009800 E 00.00.0000 0.03 0.55 030:0 0. 00..-.0 0.00005 5.05588 3.00002 0... E 5.0.03.0 000.005.08.85 .3 0.5»:—

 

 

 

 

 

 

 

 

 

 

 

 

bg 00300 Z 00.0000 =<

meow .oom Noom .oom
. _1 , _ , cod
- om...
m2 m7. .. 00..
- om.

005.5 0

00:50:: U -. ooN

mZ

 

 

 

 

 

 

 

omN

 

 

51

 

increases in the Shannon-Weiner diversity were mainly attributable to an increase in the
evenness of species, rather than an increase in species richness.

Estimates of the seed bank are presented in Table 5. The annual grass functional
group was reduced significantly by burning. This was due to combined reductions in the
seed bank of three exotic annual grasses, T. caput-medusae, Vulpia myuros (Zorro
Fescue), and Bromus hordeaceus (Soft Chess). Thus, burning was more likely to reduce
the seed bank of annual grasses than any other functional group. Other intriguing but
non-significant changes due to burning in the seed bank were a reduction in exotic

species, and an increase in the native annual forb Crassula connata (Pygmyweed).

IV. Discussion

Most research done on fire and N. pulchra has found that burning is beneficial for
that species. The greatest restoration benefit to burning the N. pulchra populations is
likely increased reproduction potential. Increased inflorescence numbers were found in
this study and others (Ahmed 1983), although in one case inflorescences were found to
produce fewer seeds (Langstroth 1991). It was surprising that N. pulchra plants growing
in a soil with a shallow hardpan responded positively to burning, as that result was not
the case for Langstroth (1991). Other studies have found heavier seeds (Dyer 2002),
although not a higher germination rate. However, Fossum (1990) found that heavier
seeds grew more quickly as seedlings and were more likely to survive their first year.
Fire may also increase the probability of N. pulchra seedling survival by removing the

thatch layer (Ahmed 1983). In this study basal area increased and mortality was flat at

52

30,500: :000 :2 0 u : .200 :_ 00 00800030 Amodvav 0:005:05
$950.2 0_ 0000 3:050:00 0000002 0:: :_ :00 .0 E06 :00 00000 .0 03:52 .m 030...

 

 

 

 

 

88.3 80.0 80.3 80.0 :003 000.0 85:03
:83 000.0 80.3 80.0 :83 08.0 000090 028.00
80.3 8.8 80.3 80.0 88.3 8.0V 00.00 .0590:
800.3 000.0 83.3 000.0 800.3 000.0 00000.0 .085
800.3 20.0 800.3 000.0 :33 P000 030... .0002
83.3 000.0 :83 3.0.0 8003 03.0 02800 0098
800.3 000.0 800.3 60.0 83.3 00.0 02800 9,002
00:95 3:28:30

:53 50.0 :003 0:0 :003 03.0 3 090.00 0.33.
803 08.0 80.3 80.0 :003 000.0 ._ Eats E0035
800.3 03.0 8 3.3 30.0 :003 30.0 0< 000805.508 520500003
88.3 08.0 :83 8.0 80.3 08.0 ”2 00003.. 0.00000
83.3 000.0 8003 30.0 :33 30.0 “2 0008 050%
:83 000.0 88.3 0 6.0 000.3 30.0 “E 03830 08:3.
:83 08.0 88.3 08.0 88.3 08.0 "2 008 00000900
:83 08.0 88.3 80.0 :83 000.0 00 80030 0000002
:83 8.0 :83 08.0 83.3 :00 9.. 8300035 $0.03
803 08.0 :83 8.0 :83 000.0 "2 8000 00000803
803 08.0 80.3 08.0 80.3 8.0v "2 002.0 080.50:
800.3 000.0 83.3 80.0 83.3 000.0 ”2 503008005 53ch
8 3.3 000.0 88.3 03.0 88.3 08.0 “2 00:08 030090
88.3 08.0 800.3 03.0 :83 000.0 “2 8.000 000E
803 80v 803 80.0 88.3 8.0V ”E 5300000. 50.03300
800.3 8.0 800.3 30.0 83.3 00.0 ”2 5330530. 5000000
80.3 80v 883 30.0 800.3 08.0 ”2 02088000 00.080300
30.0 08.0 80.3 80.0 80.3 80.0 “2 .0858 00008000
: 3.3 0 3.0 800.3 000.0 833 3:0 9. 00000038 00590.0.
Sam .002 5:6 280m .2200 0:26 00.00am

.ocau

S3

the same time that these reproductive benefits were realized. Thus, I conclude that
burning is helpful for sustaining the N. pulchra population.

I found the A. ternipes population to be in decline, and burning only appeared to
hasten the process. Both burned and control A. term'pes decreased in basal area, but their
number of inflorescences were unchanged. Control plants had a mortality rate of 4.5%,
but the rate for burned plants was more than eight times greater. No new seedlings were
observed in either control or burn plots. The plants were grazed, although not heavily,
and this may be contributing to their decline. The crowns of the Aristida plants appear
more exposed and less embedded in the soil than Nassella, perhaps due to the rocky soil
in the Aristida community. The greater exposure may have added to the heat experienced
by the plants during the fire, thus increasing mortality. Overall, burning was detrimental
to the A. term'pes population. Since so much of the research on perennial bunchgrasses in
California has focused on N. pulchra, special care should be taken when extrapolating
those results to other bunchgrass species.

In neither case did burning increase the percentage of native species cover or
decrease the exotic species cover. In the Nassella community, however, fire decreased
cover of the noxious weed T. caput-medusae. Most efforts at controlling T. caput-
medusae with fire have utilized spring burns. Although we were able to significantly
reduce T. caput-medusae with a fall burn, the species still had a cover of about 10%, and
so is likely to quickly return to pre-burn levels within 1-2 years (Young 1992).
Controlling T. caput-medusae with fall burning may require burning in at least 2
consecutive years. The space created by the reduction of this weed was exploited more

by exotic species, particularly Erodium and Vulpia myuros, than by natives. This result

54

was to be expected considering exotic species dominate the seed bank. Meyer and
Schiffman (1999) found an increase in native species cover with fall burning, but they
used plots that had much less native cover to begin with. The failure to increase the
cover of native species was reflected in the native component of the Shannon-Weiner
diversity index, which was unchanged in both communities. Although reducing specific
weeds is possible with burning, decreasing the overall cover of exotics is unlikely in
grasslands with such a large number of naturalized exotic species. In grasslands that
have few native species in the seed bank, managers should consider adding native species

seed post-bum.

 

As expected, annual forbs were increased in both communities. Annual grasses
decreased in both, although the effect was significant only in the Aristida community.
Annual grasses were also reduced in the seed bank, while annual forb seeds were not
affected by the burn. The changes observed in both these groups were similar to natural
annual fluctuations driven by weather variation (Figure 9), indicating that fire is at least
as important as annual weather in determining the botanical composition of these
communities. If an objective is to increase native species, however, management
practices that produce trade-offs between annual grasses and annual forbs are unlikely to
achieve that goal, since both groups include many exotic species. Even the legumes are
comprised of several exotic species in the Trifolium and Medicago genera. However, in
most California grasslands the other two functional groups, perennial grasses and
perennial forbs, are almost entirely composed of native species. Management practices
that benefit these two groups are more likely to result in an increase in native species.

Perennial grasses, with the exception of N. pulchra, have been little studied. Grassland

55

studies rarely focus on perennial forbs, even though they are often a significant
component of cover.

Holstein (2001) has recently suggested that tarweeds may have originally
dominated the part of the Central Valley where the study sites are located. In the
Nassella community, it is interesting to note that Hemizonia fitchii, a tarweed, was found
only in burned plots. H. fitchii also tended to increase with burning in the Aristida
community, although not to significant levels. It may be that a lack of fire, or an
accumulation of thatch, has prevented this and similar species from being as prevalent as
they may once have been. Tarweeds are deep-rooted warm season annuals, just like the
invasive Centaurea solstitialis, which is present in the Nassella community. Tarweeds
and C. solstitialis are more likely to experience competition from each other than from
cool season species. Promoting tarweeds, or other late season native annuals, may
decrease the odds of a C. solstitialis invasion (Dukes 2001).

The reduction of Selaginella hansenii in the Aristida community due to burning
may have consequences that are of concern to restoration ecologists. This spike-moss
species grows as a tightly packed groundcover and excludes most other plants from the
area it occupies. In the spring following the burn the areas that had been covered by S.
hansenii were readily apparent because they had very few seedlings and were mostly bare
soil. This pattern is evidence that S. hansem'i largely prevents seeds from penetrating the
thick vegetation mat it forms. The bare soil that persists after its removal, even through
the wet season, may experience higher erosion rates (Heady 1956), especially in steep
areas such as the Aristida community. Consideration of this effect should be given

before administering prescribed burning on land with significant slope and S. hansenii

56

 

cover. The same effect is likely to be observed in other areas that contain groundcovers,
especially ones that reproduce primarin by vegetative means.

The physical characteristics of a seed may determine whether it survives a burn in
an area with a deep thatch layer. Seeds with long awns are more likely to become
trapped in the thatch, where they are sure to be destroyed in a burn. T. caput-medusae has
long awns and is adapted to germinating while suspended in its own thatch (Young
1992). This explains the observed drop in T. caput-medusae in the seed bank and in
cover due to burning. N. pulchra seed also has long awns, and although it is a weak
hygroscopic borer, is likely to be caught in the thatch layer and destroyed in a fire.
However, unlike T. caput-medusae, there was very little N. pulchra seed in the seed bank
to begin with. The bare ground created by fire may benefit N. pulchra recruitment more
the second year after burning, when mature plants have had time to disperse more seed.

Fall would have been the most likely time for natural fires in California
grasslands since that is when they are driest. However, fall burns do not appear to have
any more beneficial effects than late spring burns when the literature is examined. The
cover of native species and the performance of native bunchgrasses are sometimes
increased with burns in either season, but neither season appears to be particularly
advantageous. A spring burn is preferred when a late-maturing invasive species is
present, because it will offer better control than a fall burn. Other than that scenario,
there does not appear to be an obvious advantage to either late—spring or fall burning. In
the Nassella community, two common management goals were achieved with our fall
burn: a decrease in noxious weeds and an increase in native bunchgrass population

viability. No management goals were achieved in the Aristida community with fall

57

burning. I would not suggest a spring burn either in that community because A. temipes
is still green and growing at that time.

N. pulchra is possibly the most widespread and common of California’s native
bunchgrasses, and certainly the most studied. Researchers dating back to (Clements
1934) have noted its ability to persist and thrive in areas that experience frequent fire.
This study found that fall burning can enhance the vigor of N. pulchra plants, an outcome
that largely agrees with previous work. However, there are dozens of native bunchgrass
species in California, most of which are potential targets for conservation or restoration,
and which may respond differently to fire. Some of those species have significantly
different life histories than N. pulchra. The negative effect of burning on A. temipes in
this study should serve as a warning that other species of bunchgrasses may have
different responses to fire than N. pulchra. That possibility should be taken into account

when designing management plans in California grasslands.

V. Literature Cited

Ahmed, E. O. 1983. Fire ecology of Stipa pulchra in California annual grassland. Doctor
of Philosophy. University of California, Davis, U.S.A.

Anderson, R. C., and E. S. Menges. 1997. Effects of fire on sandhill herbs: nutrients,
mycorrhizae, and biomass allocation. American Journal of Botany 84:938-948.

Bartolome, J. W., S. E. Klukkert, and W. J. Barry. 1986. Opal phytoliths as evidence for
displacement of native Californian grassland. Madrofio 33:217-222.

Biswell, H. H. 1989. Prescribed Burning In California Wildlands Vegetation
Management. University of California Press, Berkeley.

Clements, F. E. 1934. The Relict Method in Dynamic Ecology. Journal of Ecology 22:39-
68.

Daubenmire, R. F. 1968. Plant Communities: A Textbook of Plant Synecology. Harper &
Row, New York.

58

 

DeBano, L. F., D. G. Neary, and P. F. Folliott. 1998. Fire's Effects on Ecosystems. John
Wiley & Sons, Inc., New York, New York, U.S.A.

DiTomaso, J. M., G. B. Kyser, and M. S. Hastings. 1999. Prescribed Burning for control
of yellow starthistle (Centaurea solstitialis) and enhanced native plant diversity.
Weed Science 47:233-242.

Dukes, J. S. 2001. Biodiversity and invasibility in grassland microcosms. Oecologia
126:563-568.

Dyer, A. R. 2002. Burning and grazing management in a California grassland: Effect on
bunchgrass seed viability. Restoration Ecology 10:107-111.

Dyer, A. R., H. C. Fossum, and J. W. Menke. 1996. Emergence and Survival of Nassella
pulchra in a California Grassland. Madrono 43:316-333.

Evans, R. A., and J. A. Young. 1970. Plant litter and establishment of alien annual weed
species in rangeland communities. Weed Science 18:697-703.

Fossum, H. C. 1990. Effects of prescribed burning and grazing on Stipa pulchra (Hitchc.)
seedling emergence and survival. Master of Science. University of California,
Davis, Davis, CA, U.S.A.

Fowler, N. L. 1986. The role of competition in plant communities in arid and semiarid
regions. Annual Review of Ecology and Systematics 17:89-110.

Fowler, N. L. 1988. What is a safe site? Neighbor, litter, germination date, and patch
effects. Ecology 69:947-961.

Gross, K. L. 1990. A Comparison of Methods for Estimating Seed Numbers in the Soil.
Journal of Ecology 78: 1079-1093.

Hamilton, J. G. 1997. Changing perceptions of pre-European grasslands in California.
Madrono 44:311-333.

Hatch, D. A., J. W. Bartolome, J. S. Fehmi, and D. S. Hillyard. 1999. Effects of Burning
and Grazing on a Coastal California Grassland. Restoration Ecology 7:376-381.

Heady, H. F. 1956. Changes in a California annual plant community induced by
manipulation of natural mulch. Ecology 37:798-812.

Heady, H. F. 1988. Valley Grassland. Pages 491-514 in J. Major, editor. Terrestrial
Vegetation of California. John Wiley & Sons, New York.

Hickman, J. C., editor. 1993. The Jepson Manual. University of California Press,
Berkeley.

Holstein, G. 2001. Pre-agricultural grassland in central California. Madrofio 48:253-264.

59

Langstroth, R. P. 1991. Fire and Grazing Ecology of Stipa pulchra Grassland: A Field
Study at Jepson Prairie, California. Master of Science. University of California,
Davis, Davis, California, U.S.A.

Larson, J. R., and D. A. Duncan. 1982. Annual Grassland Response to Fire Retardant and
Wildfire. Journal of Range Management 35:700-703.

McKell, C. M., A. M. Wilson, and B. L. Kay. 1962. Effective burning of rangelands
infested with Medusahead. Weeds 10: 125-131.

Meyer, M. D., and P. M. Schiffman. 1999. Fire Season and Mulch Reduction in a
California Grassland: A Comparison of Restoration Strategies. Madrono 46:25-
37.

Parrott, R. T. 1967. A study of wiregrass (Aristida stricta Michx.) with particular
reference to fire. M. A. Duke University.

Parsons, D. J ., and T. J. Stohlgren. 1989. Effects of varying fire regimes on annual
grasslandsin the Southern Sierra Nevada of California. Madrono 36:154-168.

Reynolds, S. A., J. D. Corbin, and C. M. D'Antonio. 2001. The effects of litter and
temperature on the germination of native and exotic grasses in a coastal California
grassland. Madrono 48:230-235.

Russi, L., P. S. Cocks, and E. H. Roberts. 1992. Seed bank dynamics in a mediterranean
grassland. Journal of Applied Ecology 29:763-771.

SPSS Inc. 2000. SYSTAT 10. SPSS, Inc., Chicago, Illinois.

Stamp, N. E. 1989. Seed dispersal of four sympatric grassland annual species of Erodium.
Journal of Ecology 77: 105-1020.

Vitousek, P. M., C. M. D'Antonio, L. L. Loope, and R. Westbrooks. 1996. Biological
invasions as global environmental change. American Scientist 84:468-478.

Vogl, R. J. 1974. Effects of fire on grasslands. Pages 139-194 in T. T. Kozlowski and C.
E. Ahlgren, editors. Fire and Ecosystems. Academic Press, New York, New York,
U.S.A.

York, D. 1997. A Fire Ecology Study of a Sierra Nevada Foothill Basaltic Mesa
Grassland. Madrono 44:374-383.

Young, J. A. 1992. Ecology and management of Medusahead (Taeniatherum caput-
medusae spp. Asperum [SIMK.] melderis). Great Basin Naturalist 52:245-253.

Young, J. A., R. A. Evans, C. A. Raguse, and J. R. Larson. 1981. Germinable seeds and
periodicity of germination in annual grasslands. Hilgardia 49:1-37.

6O

Young, J. A., R. A. Evans, and J. Robison. 1972. Influence of repeated annual burning on
a Medusahead community. J oumal of Range Management 25:372-375.

61

 

Chapter 3
A Seedling Key to Some Common California Grasses

I. Introduction

The ability to reliably identify plant species at the seedling stage has challenged
botanists around the world. In some cases, waiting until the plant flowers to allow for
easier identification is an option, but in others reliable identification is needed earlier.
Dicots, even those within the same family, display an astounding variety of leaf shapes
and characteristics. Even though leaves are known for large environmental plasticity,
many defining characteristics can be found on the first leaves to develop after the
cotyledons. If the investigator knows the geographic area from which the seed came, the
species can often be narrowed down to a few dozen possible candidates based on the
primary leaves alone. Grasses present an especially difficult problem in this case because
of the similarity of their leaves. Grass keys rely on floral characteristics even more than
keys to dicot families.

Land management decisions sometimes require a botanical survey early in the
growing season. For example, burning permits must be obtained well in advance, so the
decision of which areas to burn must be made long before plants in those areas have
matured and become easily identifiable. Few plants, and possibly no grasses, flower
early in the season. However, invasive, late-maturing grasses such as Taeniatherum
caput-medusae (Medusahead) are often the targets of management decisions. The
Poaceae in California is treated exhaustively by Munz and Keck (1959) and Hickman
(1993), and partially by Crampton (1974). However, these keys rely primarily on floral

characteristics. Having a grass seedling key will increase the information available to

62

land managers and thus increase the value of their decisions. This key includes some of
the most common California grass species, including some species that are often specific
targets of management plans, such as T. caput-medusae and Nassella pulchra (purple
needlegrass).

Focusing on seedlings greatly limits the number of useful characteristics available
for writing a key. There are relatively few morphological traits with significant
interspecific variation among grass seedlings. I found the ligule to be one of the most
useful. The length of the ligule may be used to differentiate between species in several
places throughout the key. In the case of Aristida temipes var. hamulosa, bristly hairs in
place of a ligule provide a unique trait for identification.

The overall hairiness, or in some cases the positions or relative lengths of hairs, is
sometimes useful. Some seedlings have hairs easily seen with the naked eye, such as
Bromus carinatus (California Brome), while others such as Lolium multiflorum (Italian
ryegrass) are completely glabrous even under magnification. In some cases, species have
noticeably different length hairs on blades and sheaths. In the case of sheath hairs,
observations should be concentrated on the upper parts of the sheath nearest the collar.

Some traits are displayed by only a few species, and so are helpful in
identification. When present, auricles are an unmistakable clue to two species in the key.
Two other species have bow-tipped leaves. Their leaves are pointed in a shape
reminiscent of the bow of a canoe. Leaf width is an occasionally useful trait, as some
species have much wider or narrower leaves than most other species.

Since only a select set of California species are included in this key, it should be

used only when something is known about the botanical composition of the area in which

63

the specimen at hand came from. It will be most useful when species included in the key
are known to comprise the bulk of the grasses found in the area from which the specimen
came. This is often the case on actively managed lands. Specimens should have at least
three primary leaves in order to be at a stage at which they will key reliably. The key
may then be used until the grass begins to flower, at which point other keys, such as
Hickman (1993), should be used. In the case of grass seedlings, the leftover seed hull can
often still be found where root and shoot meet. This leftover hull can be invaluable for

identification, even in cases where the seedling is too young even for this key.

11. Methods

Plants used to create this key were grown from known seed. Seed was obtained
commercially as well as gathered from wild communities. Most of the seed came from
plants originating in the Sacramento Valley. Seeds were germinated and grown in a
greenhouse until they had produced 3-4 leaves, and then examined for useful
characteristics.

The key was tested with plants grown from the same seed. Unknown seedlings
were keyed out by myself, and then checked against a previously sealed list. Under these
conditions the key worked greater than 90% of the time. The most common error
occurred in couplet 17, where the key sometimes failed to accurately separate Vulpia

myuros and V. microstachys.

64

III. The grass seedling key

1. Li gule composed of bristly hairs
1’ Li gule membranous
2. Ligules _>. 0.8 mm
3. Auricles present
3’ Auricles absent
4. Sheath puberulent
5. Sheath hairs < 0.5 mm

5’ Some sheath hairs 2 0.8 mm
4’ Sheath glabrous
6. Blade flat-tipped
6’ Blade bow-tippedl
7. Ligule pointed at tip, > 2 mm
7’ Li gule rounded at tip, 1-2 mm
2’ Ligules < 0.8 mm
8. Base of stem green/white
9. Many sheath hairs > 0.2 mm
10. Some hairs > 1.5 mm
10’ All hairs < 1.5 mm
9’ Sheath hairs < 0.2 mm
11. Blade hairs < 0.4 mm
11’ Blade hairs 2 0.4 mm
12. Ligule < 0.4 mm
12’ Ligule 2 0.4 mm
8’ Base of stem with some red
13. Some sheath hairs _>_ 1 mm
13’ Sheath glabrous or hairs < 1 mm
14. Sheath glabrous
15. Auricles present
15’ Auricles absent
16. Leaf width 2 3mm
16’ Leaf width < 3 mm

17. Lines of scabrous protrusions2 on

sheath at 15X

17’ No lines of scabrous protrusions on

sheath at 15X

 

. Aristida temipes

var. hamulosa

. Hordeum vulgare

. Hordeum b. ssp.

brachyantherum

. Bromus carinatus
. Melica califomica

. Scribneria bolanderi
. Poa s. ssp. secunda

. Koeleria phleoides
. . Koeleria macrantha

. Nassella lepida

. Nassella cemua
. Nassella pulchra

. Elymus multisetus

. Lolium multiflorum

. Gastridium ventricosum

. Vulpia myuros

. Vulpia microstachys

‘ In bow-tipped leaves, the halves of the leaf on either side of the midrib come together like the bow of a

canoe.

2 Small protrusions of tissue arranged vertically along the length of the sheath are the scabrous protrusions

referred to in this couplet.

65

14’ Sheath puberulent
18. Blades puberulent
19. Lower sheath with prominent red veins

on translucent background T aeniatherum

caput-medusae

19’ Lower sheath evenly red Hordeum
brachyantheru ssp.
californicum

18’ Blades glabrous
20. Blade width > 2.5 mm Elymus glaucus
20’ Blade width < 2.5 mm
21. Ligule 2 0.3 mm F estuca califomica

21’ Ligule < 0.3 mm
22. Reddish color extending up to ligule on
outermost leaf F estuca rubra
22’ Reddish color not extending to ligule Festuca occidentalis or
F estuca idahoensis

III. Literature Cited

Crampton, B. 1974. Grasses in California. University of California Press, Berkeley.

Hickman, J. C., editor. 1993. The Jepson Manual. University of California Press,
Berkeley.

Munz, P. A., and D. D. Keck. 1959. A California Flora. UC Press, Berkeley.

66