University of Montana ScholarWorks at University of Montana
Graduate Student Theses, Dissertations, & Professional Papers Graduate School
1997
Spatial patterns of woody plant regeneration in two California Central Valley floodplain forests
William Marquiss Jones The University of Montana
Follow this and additional works at: https://scholarworks.umt.edu/etd Let us know how access to this document benefits ou.y
Recommended Citation Jones, William Marquiss, "Spatial patterns of woody plant regeneration in two California Central Valley floodplain forests" (1997). Graduate Student Theses, Dissertations, & Professional Papers. 6839. https://scholarworks.umt.edu/etd/6839
This Thesis is brought to you for free and open access by the Graduate School at ScholarWorks at University of Montana. It has been accepted for inclusion in Graduate Student Theses, Dissertations, & Professional Papers by an authorized administrator of ScholarWorks at University of Montana. For more information, please contact [email protected]. Maureen and Mike MANSFIELD LIBRARY
The University ofIVIONTANA
Permission is granted by the author to reproduce this material in its entirety, provided that this material is used for scholarly purposes and is properly cited in published works and reports.
** Please check "Yes" or "No" and provide signature**
Yes, I grant permission X No, I do not grant permission___
Author's Signature
Date 6 Æ / f f
Any copying for commercial purposes or financial gain may be undertaken only with the author's explicit consent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SPATIAL PATTERNS OF WOODY PLANT REGENERATION IN TWO CALIFORNIA CENTRAL VALLEY FLOODPLAIN FORESTS
by
William Marquiss Jones
B.A. University of California at Berkeley, 1990
presented in partial fulfillment of the requirements
for the degree of
Master of Science
University of Montana
1997
Approved by:
Chanperson
Dean, Graduate School
Date
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number; EP3764Û
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.
UMI OisMTtation PuWishing
UMI EP37640 Published by ProQuest LLC (2013). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code
Pro.Q^st'
ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jones, William Marquiss, M.S., May 1997 Environmental Studies
Spatial patterns of woody plant regeneration in two California Central Valley floodplain forests (32 pp.)
Committee Chair: Len Broberg
Abstract Because flooding is a frequent and predictable disturbance in riparian plant communities, many studies have focused on the influence of abiotic factors on woody plant regeneration in these communities. However, biotic factors may also affect patterns of plant regeneration in riparian communities, especially in portions of the floodplain subject to less frequent and intense flood disturbances. I examined the correlation between woody seedlings and groundcover and overstory composition at two sites with low flood recurrence intervals in the floodplain of the Sacramento River, California. Seedlings were distributed nonrandomly among both groundcover and overstory classes at both study sites. Seedlings of all species examined were negatively spatially associated with forb and liana cover and positively spatially associated with litter-covered substrate. Additionally, seedling densities were lower in forb and liana cover than in more open groundcover types for several species as well as all species combined. Seedling distribution among overstory cover types was strongly species specific. My results imply that plant-plant interactions may constrain the regeneration of woody species at these sites.
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgments Many good and kind people helped me with this research: Peter Hujik and Steve Greco took time out of busy schedules to think hard about how I could refine my research methods; Tom Griggs always made time for my questions and allowed me to set up field camp on Nature Conservancy land; Dawit Zeleke offered hearth and home; Stacy Cepello, Julie Cunningham, and Joyce Lacey were unstintingly generous with their time and expertise; and Jim Snowden allowed me access to California Department of Fish and Game lands along the Sacramento River. This project would have been infinitely more difficult to implement and not nearly as much ftin without their help.
I also wish to thank my committee members, Drs. Len Broberg, Ray Callaway, and Vicki Watson, for their thoughtful comments and encouragement throughout this project. 1 am also grateful to Dr. Colin Henderson for help with statistical analyses.
For a late-season kick in the pants, I am grateful to Lyn Baldwin, editor extraordinaire. And finally, I wish to thank my parents, who have long known that magic informs this world and that wonderment is our natural inheritance.
Ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
A b st r a c t ...... ii A cknowledgments ...... iii
UNTRODUCnON ...... 1
S T U D Y A R E A ...... 3
METHODS ...... 5
S pe c ie s ...... 5 Sa m ple D e s ig n ...... 6 D a t a A n a l y s is ...... 7
RESULTS ...... 9
SriE Characteristics a n d Se e d l in g s ...... 9 SEEDLING Distribution a n d d e n s it y a m o n g groundcover Cl a s s e s ...... 9 Seed lin g distribution a n d D e n sit y A m o n g O ver sto r y Cl a s s e s ...... 14 S eed ling D istribution a n d D e n sit y U n d e r Ca n o p y a n d in Ca n o p y G a p s ...... 16
DISCUSSION ...... 18
Seed lin g -G r o u n d c o v e r A ssociations ...... 18 Seed lin g -O v e r sto r y A ssociations ...... 21 Seed lin g -C a n o p y Ga p A ssociations ...... 23 Co n c l u s io n ...... 24
REFERENCES ...... 26
A P P E N D IX A ...... 31
A P P E N D IX B ...... 32
LIST OF TABLES
Ta b l e 1. P ro po rtion of G r o u n d c o v e r a n d O v e r sto r y Cl a sse s at E a c h S t u d y S i t e ...... 9 Ta b l e 2. Seed ling n u m b e r a n d d e n s it y a t Ea c h St u d y S it e ...... lo Ta b l e 3. D is'ir ib u t io n of Seed ling s with R espect to G r o u n d c o v e r a t E a c h Stu d y Site...... 11 Ta ble 4. SEEDLING DENSITIES WITH RESPECT TO GROUNDCOVER AT EACH STUDY SITE ...... 14 Table 5. Distribution of J.caufornica, P. fr e m o n th a , n d ^ . n e g u n d oSeedlings with Respect TO OVERSTORY AT THE KOPTA SLOUGH S t u d y Site ...... 15 Table 6. Seedling Densities with Respect to O verstory forc.4jjfornica, J. P. fr e x i o n w , a n d A . NEGUNDO AT THE KOPTA SLOUGH STUDY SITE...... 16 Table 7. Distribution of A ll Seedlings,J. c a u f o r n ic a , and^ I. n e g u n d o w it h R e sp e c t t o C a n o p y G a p s a t the K o pt a Slo u g h St u d y Site...... 17 Table 8. seedling densities with respect to Canopy Gaps for a ll Seedlings,j . c a u f o r n ic a , P. FREMONTII, AND/i. NEGUNDO AT ITIE KOPTA SLOUGH STUDY SITE...... 17
IV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION
Flood disturbance can be important in explaining the distribution of riparian plant
species (Johnson et al. 1976, Nanson and Beach 1977, Robertson et al. 1978, Hupp and
Osterkamp 1985, Harris 1987). Flood disturbances do not occur uniformly across a
river’s floodplain, but vary along elevational and hydrological gradients. Flood events
are longer, more severe, and more frequent on younger fluvial landforms, such as point
bars, than on higher terrace deposits (Wissmar and Swanson 1990). On younger fluvial
landforms, abiotic factors, such as soil texture, soil moisture, and flow regime, have been
strongly correlated with successful regeneration of woody plants (McBride and Strahan
1984, Sacchi and Price 1992). It is possible, however, that the relative importance of
factors influencing woody plant regeneration vary along this flood disturbance gradient.
On older landforms subject to less frequent and intense flood disturbances, such as
floodplain and terrace deposits, biotic interactions may become important for successful
seedling establishment and survival.
The Sacramento River in California’s Central Valley is an ideal system to study
regeneration patterns within floodplain forest. The Sacramento is a large, meandering
river with a well developed floodplain. Channel-floodplain dynamics have been
substantially altered by flood control levees and reservoirs, but the unconfined river reach
in the upper Sacramento Valley is subject to active lateral channel migration and
supports remnant patches of floodplain forest. Plant assemblages along the Sacramento
River are distributed predictably along an elevational gradient (Conard et al. 1977,
Strahan 1984, Cepello 1991). Point bars are characterized by open substrate, and channel
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 shelf deposits tend to be dominated by a dense coverSalix of exigua Nutt, (sandbar
willow). Floodplain deposits are characterized by gallery forests ofPopulus fremontii S.
Watson ssp. fremontii (Fremont cottonwood)and Salix gooddingii C.Ball (Goodding’s
black willow). Higher floodplain and terrace deposits are characterized by a structurally
and floristically diverse mixed riparian forest, with an overstory P.of fremontii, Juglans
californica S. Watson var. hindsii Jepson (California black walnut), Quercus lobata Nee
(valley oak), and Platanus racemosa Nutt, (western sycamore). Finally, high terraces are
characterized by an openQ. lobata forest.
Conard et al. (1977) and Strahan (1984) hypothesized that differing hydrological
regimes caused these vegetation patterns. Cepello (1991) analyzed vegetation patterns
with respect to hydrology, soil texture, density of overstory, and density of groundcover.
He found that the distribution of woody vegetation had the greatest correlation with
hydrology, as defined by flood recurrence interval. He defined three distinct fluvial
landforms on the basis of flood recurrence: channel shelf (recurrence interval less than
1.5 yr), floodplain (recurrence interval between 1.5 yr and 10 yr), and terrace (recurrence
interval greater than 10 yr). He foundSalix spp. strongly associated with channel shelf;
Populus fremontii, adult S. gooddingii, Acer negundo L. var. californicum (Torrey & A.
Gray) Sarg. (box elder), andSambucus mexicana C. Presl. (blue elderberry) strongly
associated with floodplain;Ficus carica L. (edible fig), an exotic, strongly associated
with terrace; and Juglans californica associated equally with floodplain and terrace.
The correlation between species distributions and flood recurrence interval
suggests that flood disturbance exerts an important influence in structuring vegetation
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 assemblages among fluvial landforms. Flood disturbance may restrict certain species to
floodplain and terrace landforms; however, regeneration within these landforms might be
limited by other factors. Successful regeneration by seed occurs at three distinct phases;
seed dispersal, seed germination, and seedling survival (Jones et al. 1994). Successful
regeneration sites must satisfy a species’ life history requirements at all three phases. It
is possible that biotic interactions delimit the regeneration niche (sensu Grubb 1977) of
species associated with floodplains and terraces at one or all of these developmental
phases.
If biotic factors, such as plant-plant interactions, influence within-landform
regeneration success, it is likely that seedlings will be nonrandomly distributed with
respect to vegetative cover in either groundcover or the overstory. If certain plant species
or cover types facilitate seedling establishment or survival, then seedlings should be
positively spatially correlated with those species or cover types (McAuliffe 1988,
Callaway 1992). If certain plant species or cover types interfere with seedling
establishment or survival, then seedlings should be negatively spatially correlated with
those species or cover types (Maguire and Forman 1983, Clinton et al. 1994). I tested
this hypothesis by examining the spatial distributions of woody seedlings with respect to
groundcover and overstory composition at two sites with relatively low flood recurrences
on the middle Sacramento River.
STUDY AREA
The study area was on the middle Sacramento River, California, between the
town of Red Bluff and Chico Landing. This reach of the river contains most of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 largest and least disturbed remnant patches of riparian vegetation remaining in the
Sacramento River watershed (California Resources Agency 1989). Though banks have
been locally stabilized with rock revetment, this portion of the river is largely unlevied
(California Resources Agency 1989). Thus, the river channel and associated fluvial
landforms are still shaped by erosional-depositional processes.
Using vegetation atlases of the Sacramento River (McGill 1987, California
Department of Fish and Game 1988), I delineated remnant riparian vegetation patches on
the basis of overstory composition and fluvial landform. McGill (1987) defined two
landforms for established riparian vegetation; high terrace, which is generally free from
inundation except during high river flows, and low terrace, which is generally inundated
at only moderately high river flows. Additionally, potential study sites were limited to
locations in public ownership accessible by land. I used these criteria to select one high
and one low terrace site.
The first site, Kopta Slough, was a high terrace site located at river mile 219.0 in
Tehama County (55 m elevation, 39°55' N, 122°05’ W). The second site. Pine Creek,
was a low terrace site located at river mile 197.0 in Butte County (40 m elevation, 39°44'
N, 121°58' W). The Kopta Slough site is on the Woodson Bridge Nature Preserve
managed by the California Department of Parks and Recreation. The site is dominated
by Juglans californica and has a high proportion of liana cover as well as dense pockets
o f Ficus carica (Appendix A). The Pine Creek site is on the Pine Creek Wildlife Area
managed by the California Department of Fish and Game. The site is dominated by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 Populus fremontii and is characterized by a more open understory dominated by forbs
(Appendix B).
METHODS
Species
I evaluated seedling distributions of eight native woody species common in
California Central Valley riparian forests. I defined seedlings as stems < 1.0 cm diameter
at ground level. Species examined wereAcer negundo var. californicum, Juglans
californica var. hindsii, Quercus lobata, Sambucus mexicana, Populus fremontii ssp.
fremontii, Arisiolochia californica Torrey (California pipe vine),Platanus racemosa, and
Fraxinus latifolia Benth. (Oregon ash). Nomenclature follows Hickman (1993).
Because of low sample size, F. latifolia (n = 3 seedlings) andP. racemosa (n = 1
seedling) were not included in the analysis.
Though the six species examined vary in their life history traits, all are associated
with floodplain or terrace landforms at one or more life stages.Populus fremontii
produces large amounts of small wind dispersed seeds with short viability, is shade
intolerant, and usually colonizes open substrates (Holstein 1984, Ribiero 1991).Acer
negundo produces somewhat heavier seeds which are less well dispersed and is shade
tolerant (Holstein 1984, Ribiero 1991). BothJ. californica andQ. lobata produce heavy
seeds that are dispersed by gravity or animals. Though it is unclear how important
animal-mediated dispersal is for J.californica, frugivorous birds are an important
dispersal agent for many oak species (Darley-Hill and Johnson 1981, Harrison and
Werner 1984). Sambucus mexicana produces large crops of berries, and there is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 evidence that bird dispersal is important for other congeneric species (Debussche and
Isenmann 1994).
Sample Design
I defined the sampling area at each location as the portion of the site occupied by
either P. fremontii or mixed riparian forest. Portions of sites along active channel
margins or on landforms such as point bars were not included in the study. Using aerial
photographs (USAGE 1986), I measured the sampling area at the Kopta Slough and Pine
Creek sites at 76 ha and 32 ha, respectively. I divided each location into 50 50x m cells
and randomly selected individual cells until I had sampled approximately 4 percent of
each site. I completed all field work in August and September, 1995.
I used 50-m line-intercepts to measure the proportion of groundcover and
overstory cover types (Brower et al. 1990). Line-intercepts were laid out along randomly
chosen compass bearings using the centers of selected cells as their midpoints. I
measured both groundcover and overstory composition along 0.1-m intervals. I recorded
the dominant groundcover type along the line-intercept interval and the dominant
overstory species or canopy gap vertically over the line-intercept interval. Groundcover
types included bare mineral soil, leaf litter, woody debris, and vegetative cover, which
was recorded to species or life form. Overstory was recorded to species. Where multiple
species occurred in the canopy, only the tallest was recorded.
I measured seedling-cover type associations in 20 x 50 m plots bisected by line-
intercepts. I recorded the dominant groundcover immediately adjacent to and the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 presence and composition of the overstory vertically above each seedling. I recorded the
presence of all native woody seedlings.
Data Analysis
To determine patterns of seedling distribution with respect to groundcover, I
collapsed similar groundcover types into five cover classes; (1) litter, which consisted
primarily of vegetative litter (leaves, duff, organic debris), but also included bare mineral
soil and recent alluvial deposition (soil texture predominately sand or silt. Pine Creek site
only); (2) woody debris; (3) graminoid, which included both grasses and sedge species;
(4) forb, which was dominated byArtemisia douglasiana Besser (mugwort) andSolarium
americanum Miller (American black nightshade); and (5) liana, which was dominated by
Vitis californica Benth. (California wild grape) and the vinesRuhus discolor Weihe &
Nees (Himalayan blackberry) and R. ursinus Cham. & Schldl. (California blackberry),
but also included dense groundcover of woody species such Ficusas carica.
To determine patterns of seedling distributions with respect to overstory, I also
collapsed overstory types into five classes: the three dominant overstory species [(1)
Juglans californica, (2) Populus fremontii, and (3)Acer negundo], (4) all other overstory
species (predominantlyPlatanus racemosa andSalix gooddingii), and (5) canopy gaps.
For this analysis, I was primarily interested in intraspecific patterns between seedlings
and adults. I therefore limited my analysis to seedlings of the three dominant overstory
species, J. californica, P. fremontii, andA. negundo. Because seedlings at the Pine
Creek study site occurred exclusively under P. fremontii overstory, 1 only examined
seedling-overstory patterns at the Kopta Slough study site. Finally, to test for spatial
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 associations between seedling distributions and canopy gaps, I collapsed overstory data
into two cover classes: under canopy and canopy gap.
I used two approaches to quantify spatial associations between seedling
distributions and cover class proportions. To test if seedlings were positively or
negatively associated with cover classes, I used chi-square goodness-of-fit tests (Sokal
and Rohlf 1995). This analysis tests for departures from a random distribution over all
cover classes. To test if seedling distributions differed with respect to specific cover
types, I compared seedling dei^ities within cover classes. Seedling density within each
cover class was calculated for each plot. The extent of each plot occupied by a given
cover type was estimated by extrapolating from line-intercept proportions. On a few
plots, seedlings occurred within cover types that were not represented along the line-
intercept, and these seedlings were not considered in the analysis.
Data were tested for normality with Shapiro-Wilks tests and for homogeneity of
variance with Levene tests (Sokal and Rohlf 1995). As these data were both non-normal
and heteroscedastic, I used Kruskal-Wallace tests (Sokal and Rohlf 1995) to compare
seedling densities among cover classes. Where seedling densities were significantly
different, I used simultaneous unplanned comparisons (Gibbons 1976) to determine
which cover classes differed significantly from one another. Mann-Whitney U tests
(Sokal and Rohlf 1995) were used to compare seedling densities m gaps with densities
under canopy.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESULTS
Site Characteristics and Seedlings
Table 1 presents the proportion of sample plots occupied by each groundcover
and overstory class at the Kopta Slough and Pine Creek study sites. Seedlings of eight
woody species were recorded at the Kopta Slough study site, and seedlings of five woody
species were recorded at the Pine Creek study site (Table 2). The number of seedling
species was greatest at Kopta Slough; however, overall seedling density was greatest at
Pine Creek (Table 2).
T ab le 1— ^Proportion of groundcover and overstory classes at the Kopta Slough and Pine Creek study sites. Proportion was derived from proportionate coverage of 50-m line- intercepts {n = 34, Kopta Slough;n = 14, Pine Creek). Proportion Proportion Kopta Pine Kopta Pine Groundcover Classes Slough Creek Overstory Classes Slough Creek litter 0.271 0.262 open 0.072 0.028 woody debris 0.124 0.183 Juglans californica 0.490 0.025 graminoid 0.142 0.124 Populus fremontii 0.272 0.764 forb 0.092 0.311 Acer negundo 0.087 0.100 liana 0.371 0.120 other species 0.079 0.083
Seedling Distribution and Density Among Groundcover Classes
Seedling distributions were significantly nonrandom with respect to groundcover
composition at both study sites (chi-square goodness-of-fit tests; Table 3). While
seedling distributions among groundcover types varied by species, some patterns were
consistently found across all species. For all species, seedlings were positively associated
with litter and negatively associated with forb and liana cover (Table 3). Associations
between seedlings and woody debris and seedlings and graminoid cover varied by species
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10 Table 2— Seedling number and density per hectare of each species and all species
Seedling Density Location Species Seedling Number (# seedlings/ha) Kopta Slough All seedlings 695 204.41 Juglans californica var. hindsii 272 80.00 Acer negundo var. californicum 270 79.41 Sambucus mexicana 43 12.65 Populus fremontii ssp. Jremontii 44 12.94 Quercus lobata 39 11.47 Arisiolochia californica 24 7.06 Fraxinus latifolia 2 0.59 Platanus racemosa 1 0.29
Pine Creek All seedlings 339 242.14 Acer negundo var. californicum 307 219.29 Juglans californica var. hindsii 28 20.00 Populus fremontii ssp. fremontii 2 1.43 Quercus lobata 1 0.71 Fraxinus latifolia 1 0.71
but were generally positive (Table 3). Because of low expected frequencies of seedlings
of some species in one or more cover types, I did not includeArisiolochia californica
seedlings orJuglans californica seedlings at the Pine Creek study site in this analysis.
More seedlings occurred in litter than expected if seedlings were randomly
distributed with respect to groundcover type (Table 3). Litter covered 27 percent of the
Kopta Slough study site, but contained 45 percent of all seedlings (Table 3). Though all
species occurred in this cover type at greater than expected frequencies, the proportion
ranged from 36 percent ofQ. lobata seedlings to 70 percent ofS. mexicana seedlings
(Table 3). At Pine Creek, litter covered 26 percent of the study area but contained 44
percent of all seedlings (Table 3).
Seedling distributions within woody debris varied with both species and location
(Table 3). At Kopta Slough, more seedlings occurred in woody debris than expected
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 T ab le 3— Distribution of seedlings of each species and all species combined with respect to groundcover class at the Kopta Slough and Pine Creek study sites. Chi-square statistics refer to goodness-of-fit tests for number of seedlings associated with each cover type. ______Kopta Slough______Number of associated Number of associated Number of associated seedlings (all species Acer negundo Juglans californica Pro- Cover Type portion Observed Expected Observed Expected Observed Expected litter 0.271 315 188 118 73 117 74 woody debris 0.124 156 86 78 33 46 34 graminoid 0.142 152 99 44 38 76 38 forb 0.092 30 64 10 25 18 25 liana 0.371 42 258 20 100 15 101 « - 695 seedlings n = 270 seedlings n = 272 seedlings X^ = 370.04, df = 4 f = 162.44, df=4 X^ = 142.41, df=4 P<.0001 P< 0001 P< .0001
Number of associated Number of associated Number of associated Quercus lobata Sambucus mexicana Populus fremontii Pro seedlings seedlings seedlings Cover Type portion Observed Expected Observed Expected Observed Expected Utter 0.271 14 11 30 12 25 12 woody debris 0.124 18 5 5 5 2 6 graminoid 0.142 5 5 5 6 16 6 forb 0.092 1 4 0 4 1 4 Uana 0.371 1 14 3 16 0 16 n = 39 seedlings « = 43 seedlings « = 44 seedlings = 48.94, df= 4 = 40.93, df = 4 X^ = 50.73, df=4 f < .0001 f <0001 f <0001
Pine Creek Number of associated Number of associated seedlings (all species Acer negundo Pro combined) seedlings Cover Type portion Observed Expected Observed Expected litter 0.262 150 89 147 81 woody debris 0.183 68 62 48 56 graminoid 0.124 68 42 66 38 forb 0.311 46 105 43 95 liana 0.120 7 41 3 37 n = 339 seedlings « = 307 seedlings X^=119.83, df=4 %' = 135.26, df = 4 P< .0001 P< .0001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12 when all seedlings were combined: woody debris covered 12 percent of the study area
but contained 22 percent of all seedlings (Table 3). However, spatial associations
between seedlings and woody debris varied greatly by species: only 5 percentP. of
jremontii seedlings occurred in woody debris, compared to 46 percentQ. of lobata
seedlings (Table 3). At Pine Creek, spatial associations between seedlings and woody
debris were not pronounced (Table 3).
For most species, more seedlings occurred in graminoid cover than expected
(Table 3). At Kopta Slough, graminoid cover occupied 14 percent of the area but
contained 22 percent of all seedlings and 36 percentP. of fremontii seedlings (Table 3).
At Pine Creek, graminoid cover occupied 12 percent of the study area but contained 20
percent of all seedlings (Table 3).
Fewer seedlings occurred in both forb and liana cover than expected if seedlings
were randomly distributed with respect to groundcover type (Table 3). Negative
associations were consistent for all species examined at both study sites. At Kopta
Slough, forb cover occupied 9 percent of the study area but contained 4 percent of all
seedlings, and liana cover occupied 37 percent of the area but contained 6 percent of all
seedlings (Table 3). At Pine Creek, forb cover occupied 31 percent of the study area but
contained 14 percent of all seedlings, and liana cover occupied 12 percent of the study
area but contained only 2 percent of all seedlings (Table 3).
Seedling density was significantly different among cover types for several
species. At Kopta Slough, seedling density was significantly different among cover types
for all species combined (Kruskal-Wallace test: = 36.63, df = 4, P < .0001) as well as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13 for Acer negundo (Kruskal-Wallace test: = 16.06, df = 4, P = .0029), Juglans
californica (Kruskal-Wallace test: = 23.93, df = A,P^ .0001), andQuercus lobata
(Kruskal-Wallace test: = 12.08, df = 4, P = .0168) (Table 4). Seedling density was
not significantly different among cover types forSambucus mexicana (Kruskal-Wallace
test: = 5.25, df = 4, P = .2627), Populus fremontii (Kruskal-Wallace test: x^ ^ 3.18,
d f = 4, P = .5276), znà. Arisiolochia californica (Kruskal-Wallace test: x^ = 9.24, df = 4,
P = .0554) (Table 4). At Pine Creek, seedling density was significantly different among
cover types for all species combined (Kruskal-Wallace test: x^ " 21.13, df = 4, P =
.0003) and forA. negundo (Kruskal-Wallace test: x^ 16.09, df = 4, P = .0029) but was
not significantly different for/,californica (Kruskal-Wallace test: = 4.75, df = 4, P
= 3135) (Table 4).
At Kopta Slough, seedling density in litter, woody debris, and graminoid cover
was significantly higher than in forb and liana cover for all species combined, and
seedling density in litter, woody debris, and graminoid cover was significantly higher
than in liana cover for /californica (Table 4). At Pine Creek, seedling density in litter,
woody debris, and graminoid cover was significantly higher than in liana cover for all
species combined, and seedling density in litter was significantly higher than in liana
cover for A. negundo (Table 4). The simultaneous test procedure I used to make
unplanned comparisons was very conservative (experimentwise error rate of a = .0025)
and did not correct for ties (Gibbons 1976). Though Kruskal-Wallace tests found that
seedling densities were significantly different among groundcover classesA. for negundo
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14 T able 4—Seedling densities per 0.1 ha with respect to groundcover class at the Kopta Slough and Pine Creek study sites. Data are means + 1 SD. Different superscript letters within a row indicate significantly different values (P< .05, Kruskal-Wallace test followed by simultaneous multiple comparisons),n — number of plots occupied by a given cover type.
Cover Type litter woody debris graminoid forb bana Species «=31 « = 31 « = 25 « = 20 « = 31 all species combined 43.1 + 51.4* 58.3 ±92.2* 32.7 ±44.5* 8.5 ± 12.8” 4.1 ±8.2” Acer negundo* 16.3 ±30.2 26.3 ±48.2 10.2±21.5 3.5 ± 10.3 1.9 ±5.4 Juglans californica 12.4 ± 16.3* 21.0±43.1* 14.8 ±17.3* 4.8 ±7.7*” 1.7 ±3.7” Quercus lobata* 2.6 ±5.8 6.5 ±15.2 0.7 ±2.2 0.2 ±0.7 0.2± 1.1 Sambucus mexicana 5.5 ±21.4" 2.1 ±6.7* 0.7 ± 1.9" 0.0 ± 0.0" 0.2 ± 0.9" Populus fremontii 4.7 ± 19.7* 1.3 ±6.5* 5.7 ±28.6* 0.1 ±0.4* 0.0 ± 0.0* Arisiolochia californica 1.7 ±4.4* 1.0 ±2.5* 0.5 ± 1.8* 0.0 ±0.0* 0.1 ±0.2*
Pine Creek Cover Type litter woody debris graminoid forb liana Species n = 14 «= 12 n - 9 « = 14 « = 14 all species combined 58.7 ± 125.9* 35.0 ±57.0* 48.8 ±70.4* 8.4 ± 12.1*” 1.4 ±4.8” Acer negundo 57.2 ± 126.4* 27.3 ± 59.4*” 47.7 ±71.2*” 7.7 ± 12.3*” 0.5 ± 1.6” Juglans californica 1.5 ±3.7* 5.5 ± 12.9* 0.5 ± 1.6* 0.6 ±2.2* 0.8 ±3.0* * Though Kruskal-Wallace tests found seedling densities significantly different among groundcover classes, simultaneous multiple comparisons found no significant pairwise comparisons for these species.
andQ. lobata at Kopta Slough, no pairwise comparisons were significant for these
species.
Results for F. fremontii seedlings should be viewed with caution.Populus
fremontii seedlings were not common at either study site; only twoP. fremontii seedlings
were found at Pine Creek, and at Kopta Slough, 32 of the 44 seedlings found occurred on
one plot. Thus, results are based on very localized distribution patterns.
Seedling Distribution and Density Among Overstory Classes
Juglans californica seedlings were significantly nonrandomly distributed with
respect to overstory composition: more J. californica seedlings occurred under
conspecific overstory and fewer occurred under all other overstory types than expected
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15 (Table 5). Juglans californica covered 49 percent of the Kopta Slough study site but
contained 74 percent of J.californica seedlings, whereasP. fremontii covered 27 percent
of the area but contained 11 percent of seedlings, and canopy openings covered 7 percent
of the area but contained 4 percent of seedlings (Table 5). Additionally,J. californica
seedling density was significantly different among overstory classes (Kruskal-Wallace
test: = 15.16, df == 4, f = ,0044): seedling density underJ. californica canopy was
significantly higher than in canopy openings andMndtx F. fremontii canopy (Table 6).
Populus fremontii seedlings were also significantly nonrandomly distributed with
respect to overstory: more P. fremontii seedlings occurred under canopy openings and
fewer occurred under conspecific and/,californica overstory than expected if seedlings
were randomly distributed (Table 5). Canopy openings covered 7 percent of the study
area but contained 50 percent ofP, fremontii seedlings, whereasP. fremontii overstory
covered 27 percent of the area but contained 2 percent of seedlings (Table 5). However,
P, jremontii seedling density was not significantly different among overstory classes
(Kruskal-Wallace test: y} = 1.97, df=4,P = .7408) (Table 6).
Table 5—Distribution ofJ. californica, P. fremontii, and A. negundo seedlings with respect to overstory class at the Kopta Slough study site. Chi-square statistics refer to goodness-of-fit tests for number of seedlings associated with each cover type. Number of associated Number of associated Number of associated Juglans californica Populus fremontii Acer negundo Pro seedlings seedlings seedlings Cover Type portion Observed Expected Observed Expected Observed Expected open 0.072 11 20 22 3 29 19 J. californica 0.490 200 133 14 22 135 132 P. fremontii 0.272 30 74 1 12 91 74 A. negundo 0.087 15 24 4 4 1 24 other species 0.079 16 21 3 3 14 21 n ~ 272 seedlings « = 44 seedlings « = 270 seedlings = 68.53, df= 4 X^= 133.33, df= 4 %' = 32.93, df=4 P<.0001 f <0001 P< .0001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16 T able 6—Seedlings densities per 0.1 ha with respect to overstory class at the Kopta Slough study site. Data are means ± 1 SD. Different superscript letters within a row indicate significantly different values{P< .05, Kruskal-Wallace test followed by simultaneous multiple
Cover Type open Juglans Populus Acer O th er spp.
Species H - 10 n- 28 M = 18 «=12 « = 10 Juglans californica 4.5 ±11.6® 10.2 ± 11.7” 3.8 ±8.3® 5.3 ± 10.6®” 6.9± 11.1®” Populus jremontii 7.4 + 23.3® 0.8 ±2.9® 0.0 ± 0.0® 2.9 ±9.3® 2.1 ±7.2® Acer negundo 5.7 ±12.7®*’ 7,7 + 11.8® 17.6 ±44.3®” 0.0 ± o.o” 7.5 ±17,1®”
Finally, Acer negundo seedlings were also significantly nonrandomly distributed
with respect to overstory: A. negundo seedlings showed a strong negative spatial
association with conspecific overstory and were more common underJ. californica and
P. fremontii overstory (Table 5). Adult A. negundo occupied 8 percent of the Kopta
Slough study area but contained less than 1 percent ofA. negundo seedlings (Table 5).
Acer negundo seedling density was significantly different among overstory classes
(Kruskal-Wallace test: = 13.01, df = 4 ,P = .0112): seedling density under J.
californica was significantly higher than underA. negundo (Table 6).
Seedling Distribution and Density Under Canopy and in Canopy Gaps
Relatively few seedlings occurred in canopy gaps, and for some species, no
seedlings occurred in gaps. Because of insufficient sample size, I limited the chi-square
goodness-of-fit analysis to seedlings ofJ. californica, A. negundo, and all species
combined, and the Kruskal-Wallace analysis to seedlings J.of californica, A. negundo, P.
fremontii, and all species combined.
There was no general spatial association between all seedlings combined and
canopy gaps (Table 7). However, both A. negundo an d /,californica seedlings were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17 T a b l e 7—^Distribution of seedlings of all species combined as wellJ. as californica and A. negundo seedlings with respect to canopy gaps at the Kopta Slough study site. Chi-square statistics refer to goodness-of-fit tests for the number of seedlings associated with each cover ______Number of associated Number of associated Number of associated seedlings (all species Juglans californica Acer negundo Pro- combined)______seedlings______seedlings_____ Cover Type portion Observed Expected Observed Expected Observed Expected canopy gap 0.072 60 50 11 20 29 19 under canopy 0.928 635 645 261 252 241 251 n = 695 seedlings n = 272 seedlings n = 270 seedlings x' = 2.16,df= 1 %^ = 4.37,df= 1 X' = 5.66, df- 1 P=.1421 P=.0365 P=.0173
significantly nonrandomly distributed with respect to gaps: moreA. negundo seedlings
occurred in canopy gaps than expected, but fewerJ. californica seedlings occurred in
canopy gaps than expected (Table 7). Canopy gaps occupied 7 percent of the Kopta
Slough study site, but contained 11 percent ofA. negundo seedlings and 3 percent J.of
californica seedlings (Table 7).
Juglans californica seedling density was significantly higher under canopy than in
canopy gaps (Mann-Whitney U test; exactP = .0058) (Table 8). However, in contrast to
the chi-square goodness-of-fit test results, A. negundo seedling density was not
significantly different between gaps and canopy (Mann-Whitney U test: exactP = .0873)
(Table 8). Seedling densities forP. fremontii and all seedlings combined were also not
T a b l e 8— Seedling densities per 0.1 ha with respect to canopy gaps at the Kopta Slough study site. Data are means ± S1 O . Different superscript letters within a row indicate significantly different values (P < .05, Mann-Whitney Ü test),n = number of plots occupied by
Cover Type canopy gap under canopy Species n = 10 M = 33 all species combined 17.6 ±31.6" 20,2 ±22.7" Juglans californica 4.5 ±11.6" 8.4 ± 11.0” Populus fremontii 7.4 ±23.3" 0.8 ±3.0" Acer negundo 5.7 ± 12,7" 7.6 ± 13.8"
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18 significantly different between gaps and canopy (Mann-Whitney U tests: all species
combined, exactP = .0630; P. fremontii, exact P = .9662) (Table 8).
DISCUSSION
Seedling- Groundcover Associations
The results of this study support the hypothesis that plant-plant interactions
influence woody plant regeneration in portions of the floodplain with low flood
frequencies. Seedling distributions among groundcover classes were significantly
nonrandom: seedlings of all species examined were negatively spatially associated with
both forb and liana cover and were positively spatially associated with litter-covered
substrate. These results imply that at the Kopta Slough and Pine Creek study sites,
certain vegetative cover types might limit successful seedling establishment and survival.
Negative spatial associations between seedlings and both forb and liana cover
could reflect inhibitory plant-plant interactions. These patterns could be caused by
below-ground resource competition, light competition, or allelopathy (Maguire and
Forman 1983, Ponder 1987, DeSteven 1991, Clinton et al. 1994, Berkowitz et al. 1995).
Liana cover at the two study sites was usually extremely dense. Lianas can inhibit
sapling growth in tropical forests (Putz 1984), and liana removal has been linked to
increased survivorship of tropical trees (Brokaw 1985). Strahan (1984) hypothesized that
in Sacramento River riparian systems, within-floodplain regeneration was primarily
limited by dense liana cover. Forb cover varied in density but was dominated by tall
forbs such Artemisia douglasiana andSolanum americanum, both of which often
exceeded 1 m in height.Artemisia douglasiana was the most abundant forb species
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19 sampled at both locations, and other congeneric species have been found to
allelopathically inhibit herbaceous plant growth (Schlatterer and Tisdale 1969, Halligan
1976).
Given the strong negative association between seedlings and liana cover at both
study sites, it is worth noting that a substantial proportion of liana cover was comprised
of exotic species. Twenty-one percent of liana cover at the Pine Creek study site was
occupied by exotic species, predominately Ruhus discolor. At the Kopta Slough site, 60
percent of liana cover was occupied by two exotic species,K discolor andFicus carica.
Both R. discolor andF. carica are invasive, weedy species, and the proportion of
groundcover occupied these species at both study sites may increase over time.
Correlations between seedlings and vegetative cover were not uniformly negative.
Seedlings of most species were positively spatially associated with graminoid cover, and
for some species, seedling density was significantly higher in graminoid cover than in
other vegetative cover types. These findings are in contrast to other studies, which have
found that competition for soil moisture between grasses and woody seedlings can
significantly decrease seedling growth and survival (Gordon and Rice 1993, Kolb and
Robberecht 1996). Furthermore, Goldberg and Fleetwood (1987) and Gordon et al.
(1989) found that the competitive effects of grasses on woody seedlings were more
intense than the competitive effects of forbs. It is unlikely that the positive spatial
associations between graminoid cover and seedlings reflect facilitative interactions. In
contrast to the density of forb and liana cover, which was often high, the density of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 graminoid cover was usually very low. It is possible that the density of graminoid cover
was too low for competitive interactions between seedlings and graminoids to be strong.
It is also possible that seedling distributions among vegetative cover types are
independent of the cover type itself and instead reflect variation in the physical
environment. For example, forbs and lianas might grow preferentially in microsites not
amenable to woody seedlings, whereas graminoids might grow preferentially in
microsites amenable to woody seedlings.
Positive spatial associations between seedlings and litter-covered substrate could
reflect decreased biotic interference and/or increased resource availability in this cover
type. If species’ regeneration niches are limited by inhibitory interactions with certain
vegetative cover types, higher seedling density in litter could reflect the absence of these
interactions. In old-field communities, Facelli (1994) found that litter indirectly
facilitated woody seedling growth by reducing the density and biomass, and thus the
competitive effect, of herbaceous groundcover. The physical environment in litter cover
might also be more amenable to seedling establishment and survival because of higher
below-ground resource levels or higher light regimes. Similar mechanisms could also
explain positive associations between seedlings and woody debris.
Overall seedling distribution patterns at both study sites imply that seedling
distributions may be influenced by plant-plant interactions. Overall patterns (when
seedlings were combined across species) were similar between the Kopta Slough and
Pine Creek study sites. However, seedling distribution patterns were not always similar
between study sites for particular species. Juglam californica seedling densities were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21 significantly different among groundcover classes at the Kopta Slough study site but not
at Pine Creek. This difference is probably due to the low number ofJ. californica
seedlings (n = 28 seedlings) found at the Pine Creek site.
Seedling-Overstory Associations
Juglans californica, P. fremontii, andA. negundo seedlings were nonrandomly
distributed with respect to overstory composition. Seedling distributions were strongly
species specific and could reflect different phenologies or dispersal mechanisms.
Juglans californica seedlings showed a strong affinity for conspecific overstory. Other
studies have found similar patterns for heavy-seeded species (Andersson 1991). The
positive spatial association betweenJ. californica seedlings and conspecific overstory
could reflect gravity-mediated seed dispersal. Though the allelopathic potential J.of
californica has not been studied, many studies suggest that the congenericJuglans nigra
is allelopathic to many plant species (Rice 1984, Talley et al. 1996). However, this study
found no evidence that adult californicaJ. interferes with the establishment or survival
of J. californica seedlings. This study encompassed only one field season, and I have no
data on proportionate seedling mortality rates among overstory classes. However, if
seedling distributions observed in this study reflect long-term seedling survivorship
patterns, then it appears likely that J. californica recruitment occurs primarily under
established conspecific overstory, Juglans californica stands, therefore, may be self
replacing.
Populus fremontii seedlings were strongly associated with canopy openings.
Ribiero (1991) found that low light levels did not affectP. fremontii seed germination,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 but that light was a limiting factor for P, fremontii seedling growth. He hypothesized that
because of insufficient light regimes, P. fremontii could not regenerate under the canopy
of mature riparian forest. This study supports that hypothesis. Though P. fremontii was
a substantial component of the overstory at Kopta Slough and was the dominant
overstory species at Pine Creek, only 46 seedlings were recorded at both locations, and
32 were recorded at one plot. In contrast to their rarity under established riparian forest, I
observed thatP. fremontii seedlings were abundant on early successional landforms
adjacent to both study sites. Populus fremontii seedlings were strongly negatively
associated with conspecific overstory. However, this could be an artifact of the small
number o f P. fremontii seedlings found and their localized distribution.
The most striking pattern forA. negundo was the pronounced negative spatial
association between seedlings and conspecific overstory. Negative associations between
seedlings and adult conspecifics have been observed for other members of this genus,
includingAcer saccharum in northern temperate forest (Woods 1984) andAcer rubra in
southeastern floodplain forest (Streng et al. 1989, Jones et al. 1994). High density-
dependent seedling mortality, either through herbivory, pathogenic fungi, or intraspecific
competition, has been suggested as a possible explanation for similar patterns in tropical
tree species (Augsperger 1984, Clark and Clark 1985). However, Streng et al. (1989)
found no support for any of these mechanisms as causative agents, and it remains unclear
what causes this pattern. Beatty (1984) and Crozier and Boemer (1984) both found that
some overstory trees affected the distribution and composition of herbaceous cover by
altering soil cation exchange capacity and organic matter and nutrient content. It is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 possible that adult A. negundo trees alter microsites under their canopy, creating
conditions unfavorable for either seed germination or seedling survival.
Interpreting associations betweenA. negundo seedlings and all other overstory
species is limited by my sampling method. I only recorded the tallest overstory trees
along line-intercepts; however,A. negundo was often present in the subcanopy. Given
the pronounced negative spatial association between seedlings and conspecific adults,
patterns betweenA. negundo seedlings and non-conspecific overstory might be more a
reflection of the relative presence of adultA. negundo in the midstory.
SeedUng-Canopy Gap Associations
Canopy gaps have been found to be important to woody seedling growth and
survival in many forest systems (Augsperger 1984, Canham 1988, Holmes 1995).
However, at the Kopta Slough study site, seedling distributions with respect to canopy
gaps varied by species. Acer negundo seedlings were positively spatially associated with
canopy gaps while J. californica seedlings were negatively spatially associated with
canopy gaps and also occurred at lower densities in gaps . It is unclear what mechanisms
cause these species-specific responses.
It is possible that soil moisture is limiting at the Kopta Slough study site for some
woody riparian species. Ribiero (1991) found that soil moisture accounted for more
variability in germination ofA. negundo andJ. californica seeds than light in controlled
experiments. The Sacramento Valley experiences regular summer drought: virtually no
precipitation occurs between May and October, and summer temperatures often exceed
35®C. Drought effects in riparian communities may be ameliorated by proximity to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24 groundwater; however, due to limited root development, seedlings of some species may
be unable to access this resource, and soil moisture may be a critical factor for these
species. It is possible that J. californica seedling survivorship at the Kopta Slough study
site is soil moisture limited, and that for this species, the beneficial effects of higher light
regimes in canopy gaps might be outweighed by decreases in soil moisture in these
microsites (Holmgren et al. in press).
If soil moisture is limiting in canopy gaps, the positive correlation betweenA.
negundo seedlings and these microsites could reflect more efficient water relations for
this species. It is possible that for A negundo seedlings, decreases in soil moisture do not
outweigh the beneficial effects of higher light regimes in canopy gaps. Though soil
moisture was more important, higher light did increase seed germination ofW.negundo
seeds (Ribiero 1991). Other mechanisms, such as greater seed dispersal, could also
explain A. negundo distributions with regard to canopy gaps observed in this study.
Conclusion
This study supports the hypothesis that biotic interactions are important factors
for within-floodplain regeneration. Seedlings were nonrandomly distributed among both
groundcover and overstory classes, implying that regeneration for woody riparian species
in floodplain landforms might be influenced by overstory and groundcover composition.
Seedling distributions were strongly correlated with cover type. Whether mediated
directly by the composition of groundcover or overstory or by variation in physical
environment, microsite variability appears to delimit, at least in part, woody species’
regeneration niche at the two study sites. If long-term seedling survivorship is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 proportionate to distribution patterns observed in this study, microsites amenable to
seedling establishment and survival may be critical to successful regeneration by seed.
However, given the dominance of invasive exotic species such asRubus discolor
andFicus carica in portions of the floodplain, the proportion of groundcover classes may
change over time. Cepello (1991) foundF. carica was strongly correlated with terrace
landforms (flood recurrence interval >10 yr). The development of water resources,
especially the construction of Shasta Dam upstream of the study sites, has dampened the
intensity and frequency of flood events along the mainstem of the Sacramento River
(Kahrl 1979). This may create conditions more favorable toF. carica expansion. If
liana cover limits regeneration opportunities for native woody species, an increase in the
proportion of this cover type may alter long-term species composition in remnant
floodplain forest on the Sacramento River.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26 REFERENCES
Andersson, C. 1991. Distribution of seedlings and saplingsQuercus of robur in a grazed deciduous forest. J. Veg. Sci. 2: 279-282.
Augsperger, O.K. 1984. Seedling survival of tropical tree species: interactions of dispersal distance, light gaps, and pathogens.Ecology 65: 1705-1712.
Beatty, S.W. 1984. Influence of microtopography and canopy species on spatial patterns of forest understory plants. Ecology 65: 1406-1420.
Berkowitz, A.R., C D. Canham and V.R. Kelley. 1995. Competition vs. facilitation of tree seedling growth and survival in early successional communities.Ecology 76: 1156-1168.
Brokaw, N.V.L. 1985. Treefalls, regrowth, and community structure in tropical forests. Pp. 53-69 in S.T.A. Pickett and P.S. White (eds.). The ecology of natural disturbance and patch dynamics. Academic Press, Inc., London.
Brower, J.E., J.H. Zar and C.N. von Ende. 1990. Field and laboratory methods for general ecology. 3rd ed. Wm.C. Brown Publishers, Dubuque.
California Department of Fish and Game. 1988. Sacramento River riparian atlas. Wildlife Conservation Board, Nongame Heritage Program, Sacramento.
California Resources Agency. 1989. Upper Sacramento River fisheries and riparian habitat management plan. Sacramento.
Callaway, R.M. 1992. Effect of shrubs on recruitment of Quercus douglasii and Quercus lobata in California. Ecology 73: 2118-2128.
Canham, C D 1988. Growth and architecture of shade tolerant trees: response to canopy gaps. Ecology 69: 786-795.
Cepello, S. A. 1991. Riparian vegetation distribution along the middle Sacramento River in relation to flood frequency. MS Thesis, California State University, Chico.
Clark, D.B. and D A. Clark. 1985. Seedling dynamics of a tropical tree: impacts of herbivory and meristem damage.Ecology 66: 1884-1892.
Clinton, B.D., L.R. Boring and W.T. Swank. 1994. Regeneration patterns in canopy gaps of mixed-oak forests of the southern Appalachians: influence of topographic position and evergreen understory.Am. Midi Nat. 132: 308-319.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 Conard, S.G., R.L. MacDonald and R.F. Holland. 1977. Riparian vegetation and flora of the Sacramento Valley. Pp. 47-55 in A. Sands (ed.). Riparian forests in California. Institute of Ecology Publication No. 15, University of California, Davis.
Crozier, C.R. and R.E.J. Boemer. 1984. Correlations of understory herb distribution patterns with microhabitats under different tree species in a mixed mesophytic forest. Oecologia 62: 337-343.
Darley-Hill, S. and W.C. Johnson. 1981. Acorn dispersal by the blue jay{Cyanocitta cristata). Oecologia 50: 231-232.
Debussche, M. and P. Isenmann. 1994. Bird-dispersed seed rain and seedling establishment in patchy Mediterranean vegetation.Oikos 69; 414-426.
DeSteven, D 1991. Experiments on mechanisms of tree establishment in old-field succession: seedling survival and growth.Ecology 72; 1076-1088.
Facelli, J.M. 1994. Multiple indirect effects of plant litter affect the establishment of woody seedlings in old-flelds.Ecology 75: 1727-1735.
Gibbons, J.D. 1976. Nonparametric methods for quantitative analysis. Holt, Reinhart & Winston, New York.
Goldberg, D.E. and L. Fleetwood. 1987. Competitive effect and response in four annual plants. J. Ecol. 75; 1131-1143.
Gordon, D R. and K.J. Rice. 1993. Competitive effects of grassland annuals on soil water and blue oak {Quercus douglasii) seedlings. Ecology 74: 68-82.
Gordon, D R., J.M. Welker, J.W. Menke and K.J. Rice. 1989. Competition for soil water between annual plants and blue oak{Quercus douglasii) seedlings. Oecologia 79: 533-541.
Grubb, P.J. 1977, The maintenance of species-richness in plant communities: the importance of the regeneration niche.Bio. Rev. 52: 107-145.
Halligan, J.P. 1976. Toxicity o îArtemisia californica to four associated herb species. Am. Midi. Nat. 95; 406-421.
Harris, R.R. 1987. Occurrence of vegetation on geomorphic surfaces in the active floodplain of a California alluvial stream.Am. Midi. Nat. 118: 393-405.
Harrison, J.S and P.A. Werner. 1984. Colonization by oak seedlings into a heterogeneous successional habitat. Can. J. Bot. 62: 559-563.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 Hickman, J.C. (éd.). 1993. The Jepson manual: higher plants of California. University of California Press, Berkeley.
Holmes, T.H. 1995. Woodland canopy structure and the light response of juvenile Quercus lobata (Fagaceae). Am. J. Bot. 82; 1432-1442.
Holmgren, M., M. Scheffer and M.A. Huston. In press. The interplay of competition and facilitation in plant communities.Ecology.
Holstein, G. 1984. California riparian forests: deciduous islands in an evergreen sea. Pp. 2-22 in R.E. Warner and K.M. Hendrix (eds.). California riparian systems. University of California Press, Berkeley.
Hupp, C.R. and W.R. Osterkamp. 1985. Bottomland vegetation distribution along Passage Creek, Virginia, in relation to fluvial landforms.Ecology 66: 670-681.
Johnson, W.C., R.L. Burgess and W.R. Keammerer. 1976. Forest overstory and environment on the Missouri River floodplain in North Dakota.Ecol. Monogr. 46: 59-84.
Jones, R.H., R.R. Sharitz, P.M. Dixon, D.S. Segal and R.L. Schneider. 1994. Woody plant regeneration in four floodplain forests.Ecol. Monogr. 64: 345-367.
Kahrl, W.L. (ed.). 1979. The California water atlas. California Govemor’s Office of Planning and Research and California Department of Water Resources, Sacramento.
Kolb, P.F. and R. Robberecht. 1996. Pinus ponderosa seedling establishment and the influence of competition with the bunchgrassAgropyron spicatum. Int. J. Plant Sci. 157: 509-515.
Maguire, D.A. and R.T.T. Forman. 1983. Herb cover effects on tree seedling patterns in a mature hemlock-hardwood forest. Ecology 64: 1367-1380.
McAuliffe, J R. 1988. Markovian dynamics of simple and complex desert plant communities. Am. Nat. 131: 459-490.
McBride, J R. and J. Strahan. 1984. Establishment and survival of woody riparian species on gravel bars of an intermittent stream.Am. Midi. Nat. 112: 235-245.
McGill, R.R. 1987. Land use changes in the Sacramento river riparian zone. Redding to Colusa, third update, 1982 to 1987. California Department of Water Resources, Sacramento.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29 Nanson, G.C. and H.F. Beach. 1977. Forest succession and sedimentation on a meandering-river floodplain, northeast British Columbia, Canada.J. Biogeog. 4; 229-251.
Ponder, Jr., F. 1987. Effect of three weeds on the growth and mycorrhizal infection of black walnut {Juglans nigra) seedlings. Can. J. Bot. 64: 1888-1892.
Putz, F.E. 1984. The natural history of lianas on Barro Colorado Island, Panama. Ecology 65: 1713-1724.
Ribiero, J.F. 1991. Environmental heterogeneity in space and time and plant life history traits on zonation of five riparian woody species of the California Central Valley. Ph.D. Dissertation, University of California, Davis.
Rice, E.L. 1984. Allelopathy. 2nd ed. Academic Press, New York.
Robertson, P.A., G.T. Weaver and J.A. Cavanaugh. 1978. Vegetation and tree species patterns near the northern terminus of the southern floodplain forest.Ecol. Monogr. 48: 249-267.
Sacchi, C.F. and P.W. Price. 1992. The relative roles of abiotic and biotic factors in seedling demography of arroyo willow{Salix lasiolepis: Salicaceae). Am. J. Bot. 79: 395-405.
Schlatterer, E.F. and E.W. Tisdale. 1969. Effects of litter ofArtemisia, Chrysothamnus, andTortula on germination and growth of three perennial grasses.Ecology 50: 869- 873.
Sokal, R.R. and F.J. Rohlf. 1995. Biometry; the principles and practice of statistics in biological research. 3rd ed. W.H. Freeman and Co., New York.
Strahan, J. 1984. Regeneration o f riparian forests of the Central Valley. Pp. 58-67/n R.E. Warner and K.M. Hendrix (eds.). California riparian systems. University of California Press, Berkeley.
Streng, D R., J.S. Glitzenstein and P. A. Harcombe. 1989. Woody seedling dynamics in an east Texas floodplain forest.Ecol. Monogr. 59: 177-204.
Talley, S.M., R.O. Lawton and W.N. Setzer. 1996. Host preference ofRhus radicans (Anacardiaceae) in a southern deciduous hardwood forest.Ecology 77; 1271-1276.
USAGE [U.S. Army Corps of Engineers], 1986. Sacramento River Flood Control System: Sacramento River and tributaries, Collinsville to Red Bluff. 1 " = 400' aerial photographs. File No. 50-18-5805. Sacramento District, Sacramento.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 Wissmar, R.C. and F,J. Swanson. 1990. Landscape disturbances and lotie ecotones. Pp. 65-89 in R.J. Naiman and H. Décamps (eds.). The ecology and management of aquatic-terrestrial ecotones. UNESCO, Paris.
Woods, K.D. 1984. Patterns of tree replacement: canopy effects on understory pattern in hemlock-northern hardwood forests.Vegetatio 56: 87-107.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31
APPENDIX A
Table A-1— Proportion of groundcover at Kopta Slough (derived from proportionate cover of 50-m line-intercepts,n = ______Groundcover Proportion litter (duff, leaves) 0.259 graminoid (grasses andCarex spp.) 0.142 Rubus discolor (Himalayan blackberry) 0.129 Vitis californica (California wild grape) 0.096 Ficus carica (edible fig) 0.096 small woody debris (< 5 cm diameter) 0.071 large woody debris (> 5 cm diameter) 0.053 Artemisia douglasiana (mugwort) 0.047 Rubus ursinus (California blackberry) 0.043 Solanum americanum (American black nightshade) 0.019 bare (mineral soil) 0.012 Conyza spp. (horseweed) 0.007 Phytolacca americana (American pokeweed) 0.006 Urtica dioica ssp. holosericea (hoary creek nettle) 0.004 Bidens frondosa (sticktight) 0.004 Juglam californica var. hindsii (northern California black walnut) 0.003 forb (species unknown) 0.002 Xanthium strumarium (cocklebur) 0.002 Aristolochia californica (California pipevine) 0.001 Clematis ligusticifolia (virgin’s-bower) 0.001 Acer negundo var. californicum (box-elder) 0.001 Sambucus mexicana (blue elderberry) 0.001 Fraxinus latifolia (Oregon ash) trace Heliotropium curassavicum (wild heliotrope) trace Polygonum lapathifolium (wiUow-weed) trace Toxicodendron diversilobum (poison oak) trace Datura stramonium (jimsonweed) trace
T able A-2— Proportion of overstory at Kopta Slough (derived from proportionate cover of 50-m line-intercepts,n = 34) Overstory Proportion Juglam californica var. hindsii (northern California black walnut) 0.490 Populus fremontii ssp. fremontii (Fremont cottonwood) 0.272 Acer negundo var. californicum (box-elder) 0.087 open (canopy gap) 0.072 Platanus racemosa (western sycamore) 0.027 Ficus carica (edible fig) 0.019 Juglans regia (English walnut) 0.017 Salix gooddingii (Goodding’s black willow) 0.011 Ptelea crenulata (hoptree) 0,003 Sambucus mexicana (blue elderberry) 0.002 Salix exigua (sandbar willow) 0.001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 APPENDIX B
T ab le B-1— Proportion of groundcover at Pine Creek (derived from proportionate cover of 50-m line-intercepts, n = 14) Groundcover Proportion litter (duff, leaves) 0.232 Artemisia douglasiana (mugwort) 0.141 small woody debris (< 5 cm diameter) 0.115 graminoid (grasses andCarex spp.) 0.109 large woody debris (> 5 cm diameter) 0.068 Solanum americanum (American black nightshade) 0.063 Vitis californica (California wild grape) 0.033 Rubus ursinus (California blackberry) 0.032 forb (species imknown) 0.033 Conyza spp. (horseweed) 0.029 deposition (soil texture dominated by sand or silt) 0.027 Equisetum spp. (horsetail) 0.026 Rubus discolor (Himalayan blackberry) 0.023 Clematis ligusticifolia (virgin’s-bower) 0.015 Arundo donax (giant-reed) 0.015 Acer negundo var. californicum (box-elder) 0.007 Vinca major (periwinkle) 0.005 Xanthium strumarium (cocklebur) 0.005 Bidens frondosa (sticktight) 0.004 Chenopodium spp. (goosefoot) 0.004 bare (mineral soil) 0.003 Datura wrightii (thorn-apple) 0.003 Juglans californica var. hindsii (northern California black walnut) 0.002 Fraxinus latifolia (Oregon ash) 0.002 Salix lasiolepis (arroyo willow) 0.001 Salix exigua (sandbar willow) trace Nicotiana glauca (tree tobacco) trace Polygonum lapathifolium (willow-weed) trace Nicotiana quadrivalvis (Indian tobacco) trace
T able B-2— Percent cover of overstory at Pine Creek (derived from proportionate cover of 50-m line-intercepts,n = 14) Overstoiy Proportion Populus fremontii ssp. fremontii (Fremont cottonwood) 0.764 Acer negundo var. californicum (box-elder) 0.100 Salix gooddingii (Goodding’s black willow) 0,046 Platanus racemosa (western sycamore) 0.031 open (canopy gap) 0.028 Juglans californica var. hindsii (northern California black walnut) 0.025 Arundo donax (giant-reed) 0.005
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.