Citation
Riparian tree associations and storage, transport, and processing of particulate organic matter in a subtropical stream

Material Information

Title:
Riparian tree associations and storage, transport, and processing of particulate organic matter in a subtropical stream
Creator:
Roberts, Christopher Richard, 1972-
Publication Date:
Language:
English
Physical Description:
x, 98 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Creeks ( jstor )
Floodplains ( jstor )
Leaf abscission ( jstor )
Leaves ( jstor )
Macroinvertebrates ( jstor )
Opossums ( jstor )
Plant litter ( jstor )
Species ( jstor )
Streams ( jstor )
Taxa ( jstor )
Dissertations, Academic -- Environmental Engineering Sciences -- UF ( lcsh )
Environmental Engineering Sciences thesis, Ph. D ( lcsh )
City of Tallahassee ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references (leaves 84-93).
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Christopher Richard Roberts.

Record Information

Source Institution:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
028834881 ( ALEPH )
50419032 ( OCLC )

Downloads

This item has the following downloads:


Full Text










RIPARIAN TREE ASSOCIATIONS AND STORAGE, TRANSPORT, AND PROCESSING OF PARTICULATE ORGANIC MATTER IN A SUBTROPICAL STREAM












By

CHRISTOPHER RICHARD ROBERTS


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2002













ACKNOWLEDGMENTS

I greatly appreciate all the efforts of my major professor, Dr. Thomas Crisman. His advice and guidance, both dealing with this project and in more personal areas, through this arduous process was considerable. I am also grateful to my other committee members, Drs. Lauren Chapman, Frank Nordlie, and William Wise for their counsel.

My wife was inspirational through this entire process, never letting me get down on my setbacks and always pushing me to work my hardest and best. She has worked equally hard in her studies and although our time together has been limited, her constant inspiration has led me to the point where I am now.

Special thanks go to Todd Osborne for helping with processing of leaf materials in the laboratory. All chemical analyses were performed by the Forage Evaluation Support Laboratory at the University of Florida, with specific thanks to Richard Fethiere. Without all their help, the completion of this project and thesis would never have been accomplished. Thanks also go to all the student workers who slaved away in anonymity to allow others to reap the benefits. I would also like to thank the Center for Wetlands staff, specifically Sherl Brinkley, for their behind the scene efforts in making this research a reality.


ii
















TABLE OF CONTENTS

page
ACKNOWLEDGMENTS.......................................................... ii

LIST OF TABLES.................................................................... v

LIST OF FIGURES.................................................................. vi

A B STR A C T ........................................................................... ix

CHAPTERS

1 INTRODUCTION............................................................ 1

2 LEAF ASSOCIATIONS.................................................. 4

Introduction.................................................................. 4
Methods...................................................................... 5
Results....................................................................... 7
Discussion.................................................................... 12

3 STORAGE AND TRANSPORT......................................... 21

Introduction.................................................................. 21
Methods....................................................................... 22
Study Site............................................................. 22
Experimental Approach............................................ 23
Data Analysis........................................................ 27
Results and Discussion..................................................... 27
Latitudinal Comparisons........................................... 27
Organic Matter Budget............................................ 31
Inputs.................................................................. 34
Storage................................................................ 4 1
Transpoui............................................................. 50

4 LITTER PROCESSING..................................................... 56

Introduction.................................................................. 56
Methods....................................................................... 57


iii









Study Site ............................................................ 58
Data Analysis....................................................... 61
R esults......................................................................... 6 1
D iscussion..................................................................... 73
Leaf Chemistry...................................................... 73
Timing of Litter Fall................................................ 77
Macroinvertebrates................................................. 78

5 CONCLUSIONS ............................................................ 82

REFERENCES....................................................................... 84

APPENDICES......................................................................... 94

BIOGRAPHICAL SKETCH....................................................... 97


iv















LIST OF FIGURES


Figure page
2-1 Dendrogram showing associations of 31 common riparian tree
species from north-central Florida based on phosphorus, nitrogen,
recalcitrant and inhibitory compounds, cuticle development, and
timing of leaf fall. Numbers on left indicate four discrete groupings......... 10

2-2 Period of leaf fall of 31 common riparian tree species from
north-central Florida from September 2000 to August 2001.................... 11

2-3 Percentage of dry weight ( 1 standard error) consisting of
phosphorus for 31 common riparian tree species from north-central
F lorid a ................................................................................... 13

2-4 Percentage of dry weight ( 1 standard error) consisting of nitrogen
for 31 common riparian tree species from north-central Florida................ 14

2-5 In vitro organic matter digestibility (IVOMD) (+ 1 standard error)
as a percentage of dry weight for 31 common riparian tree species
from north-central Florida............................................................ 15

3-1 Map of Possum Creek, Florida, study site and placement of sampling
equipm ent in floodplain............................................................... 24

3-2 Approximate timing of leaf abscission from selected forests of the
world. Solid line indicates timing of leaf fall..................................... 28

3-3 Organic matter budget for the Possum Creek system, Florida, from
October 2000-November 2001. Solid arrows indicate measured
movements, and dashed arrows indicate hypothesized movements of
organic matter between input, storage, and export compartments. Inputs
(left column of boxes) and exports (right column) are expressed as
annual summation of each transport vehicle. Storage (middle column)
is expressed as the annual mean. All data are expressed as g AFDM
m y r .................................................................................. 3 3

3-4 Organic matter inputs via litterfall (circles) and lateral deposition
(triangles) from October 2000 through November 2001 into Possum
Creek, Florida. Error bars represent one standard error............................ 35


V









3-5 Percent by weight of constituents of organic matter in litterfall
from October 2000 through November 2001 at Possum Creek, Florida.
Reproductive structures include nuts, seeds, blossoms, and other
structures associated with floral reproduction..................................... 37

3-6 Percent by weight of constituents of organic matter in lateral
deposition from October 2000 through November 2001 at Possum
Creek, Florida. Gap in data represents a flooding event that disabled lateral deposition traps. Reproductive structures include nuts, seeds,
blossoms, and other structures associated with floral reproduction............... 39

3-7 Percent by weight of constituents of organic matter stored on the
Possum Creek floodplain, Florida, from October 2000 through November 2001. Reproductive structures include nuts, seeds,
blossoms, and other structures associated with floral reproduction............ 42

3-8 Organic matter storage on the Possum Creek floodplain, Florida, from
October 200 through November 2001. Error bars represent one
standard error.......................................................................... 43

3-9 In-stream organic matter storage in leaf packs (white) and debris
dams (black) per 100 m of stream length from October 2000
through November 2001 in Possum Creek, Florida .............................. 45

3-10 Percent by weight of constituents of in leaf packs (upper) and
debris dams (lower) from October 2000 through November 2001 in
Possum Creek, Florida. Gaps in data represent periods where no leaf
packs were present. Reproductive structures include nuts, seeds,
blossoms, and other structures associated with floral reproduction ............ 46

3-11 In-stream benthic organic matter storage as CPOM (circles) and FPOM
(triangles) from October 2000 through November 2001 in Possum
Creek, Florida. Error bars represent : one standard error....................... 49

3-12 In-stream organic matter transport as FPOM (white) and CPOM
(black) from October 2000 through November 2001 in Possum
C reek, Florida .......................................................................... 51

3-13 Gage height at Possum Creek, Florida, from September 2000 to
October 2001. Asterisks (*) represent instances when water height
exceeded bankfull stage............................................................... 52

4-1 Percent of dry weight of phosphorus, nitrogen, and refractory
compounds in the senescent leaves of five tree species collected
from October 2000-May 2001. Error bars represent one


vi








standard error . ......................................... .. ---.................... 63

4-2 Macroinvertebrate abundance and leaf carbon content over time
collected from Possum Creek from December 2000 to May 2001.
Gray bars represent macroinvertebratesdleaf pack-' and white bars
represent macroinvertebrates-g C' -leaf pack-'. Both sampling date
and cumulative degree days are shown. Error bars represent
one standard error ...................................................... 64

4-3 Dominant shredders and collector-gatherers collected from Possum
Creek from December 2000 to May 2001. Gray bars represent
macroinvertebrates- leaf pack'I and white bars represent
macroinvertebrates-g C- -leaf pack-. Both sampling date and cumulative degree days are shown. Error bars represent one
standard error .......................................................... ...... 67

4-4 Dominant filter-feeders collected from Possum Creek from
December 2000 to May 2001. Gray bars represent
macroinvertebrates- leaf pack- and white bars represent
macroinvertebrates-g C' leaf pack-'. Both sampling date and cumulative degree days are shown Error bars represent one
standard error ....................................................................................... 68

4-5 Dominant predators collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf
pack-' and white bars represent macroinvertebrates-g C -leaf
pack-'. Both sampling date and cumulative degree days are
shown Error bars represent one standard error ............................... 69

4-6 Chironomidae (Diptera) collected from Possum Creek from December 2000 to May 2001. Gray bars are macroinvertebratesleaf pack-' and white bars are macroinvertebrates-g C' -leaf pack'.
Both sampling date and cumulative degree days are shown.
Error bars represent one standard error ................................................. 71

4-7 Chironomidae (Diptera) relative abundance over time collected from Possum Creek from December 2000 to May 2001. Both
sampling date and cumulative degree days are shown. ............................. 72


vii














LIST OF TABLES

Table page
2-1 Scientific and common names of the 31 common riparian species
from north central Florida used in this study and the degree of
cuticle development. Cuticle development is on a scale of 1-5, with
1 representing the most heavily developed cuticle and 5 the least
developed .............................................................................. 8

3-1 Litterfall from representative forests along a latitudinal gradient.............. 29

3-2 Lateral deposition from representative forests along a latitudinal
G radient................................................................................ 32

3-3 Organic matter storage in representative first and second-order streams
along a latitudinal gradient.......................................................... 40

3-4 Organic matter export from representative first and second-order streams
along a latitudinal gradient.......................................................... 53

4-1 Macroinvertebrates collected from Possum Creek, Florida, on leaf
packs from December 2000 to May 2001......................................... 65


viii














Abstract of Dissertation Presented to the Graduate School in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

RIPARIAN TREE ASSOCIATIONS AND STORAGE, TRANSPORT,
AND PROCESSING OF PARTICULATE ORGANIC MATTER IN A SUBTROPICAL STREAM

By

Christopher Richard Roberts

May 2002

Chairperson: Thomas L. Crisman
Major Department: Environmental Engineering Sciences

Riparian tree associations, leaf-litter storage, transport, and in-stream processing were investigated in subtropical north-central Florida. Chemical constituents, cuticle development, and timing of leaf fall were quantified for 31 species and analyzed with hierarchical cluster analysis. Four groups and two outlying species were classified. Pinus elliottii, uncommon in natural riparian systems, but widely planted regionally in plantations, and Sapium sebiferum, an exotic, were statistically associated with other species, but monocultures could deleteriously affect invertebrates that process leaf litter. These associations can be used to better understand vegetative community management and restoration of riparian areas.

Storage and transport of leaf litter were analyzed biweekly over a year in both a stream and floodplain system. Peak leaf fall occurred from September-December with a smaller peak during January-february and were reflected in peaks of both lateral and longitudinal organic matter transport. Longitudinal transport was also correlated to stream discharge. Floodplain storage was highest during winter 2000-2001 during and after the period of peak leaf fall. Variability in floodplain organic matter increased ix








markedly after flooding during July and September 2001. These results suggest that leaf fall is temporally extended and that fresh litter is available for much of the year.

Leaves of five common riparian tree species were compared for in-stream

processing rates. Early-abscising deciduous trees were processed twice as rapidly as late-abscising evergreen trees. In contrast to most temperate streams, shredding macroinvertebrates were rarely collected and appear to be of limited importance in leaf processing. This suggests that fungal and bacterial processing is of greater importance in subtropical streams. Macroinvertebrate abundance on leaf packs decreased from December through May, rather than over the course of processing, suggesting that factors other than leaf availability influence macroinvertebrate abundance. The fact that few shredders were present suggests that invertebrates were using leaf packs as habitat and, in the case of scrapers and collector-gatherers, for feeding fungal communities and zone of fine particulate organic matter accumulation, respectively.

This research fills a void in understanding stream ecosystem function along a latitudinal gradient. Subtropical streams, in terms of both timing of leaf fall and a macroinvertebrate community only loosely synchronized to seasonal leaf inputs, behave in a way intermediate between conditions in temperate streams and those of the Southern Hemisphere.


x













CHAPTER 1
INTRODUCTION

Small streams in forested watersheds typically rely on materials derived from outside the stream channel as their primary energy base (e.g., Cummins et al. 1973; Fisher and Likens 1973; Hynes 1975). Allochthonous materials that enter a stream are primarily in the form of coarse particulate organic matter (CPOM), which is predominantly leaves and wood, but also reproductive structures such as blossoms, fruits, and nuts, and dead animals. Once in the stream, the most CPOM is trapped by in stream structure such as fallen tree limbs, roots, and rocks. Here, it is processed by a suite of organisms, ranging from bacteria and fungi to macroinvertebrates and broken down into smaller particles (fine particulate organic matter (FPOM) and dissolved organic matter (DOM)) that are exported downstream and utilized by other organisms. This process has been synthesized into the "River Continuum Concept" (Vannote et al. 1980), which describes the structure and function of streams from their headwaters to where they enter a larger stream or lentic body of water.

These processes are well known in Northen Hemisphere temperate systems

(e.g., Cummins et al. 1973; Fisher and Likens 1973; Hynes 1975), but poorly studied in the subtropics. The temperate zone and subtropics differ in many fundamental respects, including climate, pliutoperiod, and timing of litter accession (Lugo and Zacca 1983; Watt and Golladay 1999). Applying concepts based on temperate systems to those in other regions is potentially inappropriate. For example, lowland tropical systems


I






2


operate differently than what is outlined in the "River Continuum Concept" (Vannote et al. 1980), but rather is more properly expressed by Junk (1984) in the "Flood-Pulse Concept". Furthermore, Southern Hemisphere temperate systems operate differently than those of the Northern Hemisphere (King et al. 1987, Lake and Barmuta 1986), with leaf fall more closely associated with precipitation (dry season) than with temperature and photoperiod as is the case in the Northern Hemisphere. Because systems outside the Northern Hemisphere temperate zone operate differently than what has been accepted by scientists in those regions, it stands to reason that Northern Hemisphere subtropics may differ from temperate systems as well.

Subtropical systems, such as those in north-central Florida, experience an

extended period of leaf fall as compared to the brief autumnal period in the Northern Hemisphere temperate zone (Lugo and Zacca 1983; Watt and Golladay 1999). Not only will availability of leaves be governed by the extended subtropical leaf fall as compared to the Northern Hemisphere temperate zone, but also leaf quality will differ among trees. Initial leaf quality is often the most important factor determining processing rates of leaves (Kaushik and Hynes 1971; Merritt et al. 1984). Different tree species have leaves that contain different concentrations of chemical and physical deterrents as well as nutrients, which influence their processing rates in streams. Chemical defenses are of greatest importance in many species, but thick, waxy cuticles may also retard processing in others (Bunn 1988; Canhoto and Graga 1995).

This research addresses these issues in three parts. First, riparian trees taxa from north-central Florida were segregated into discrete groupings based on the commonalities of when taxa are likely to become available to be processed, using






3
phosphorus, nitrogen, and recalcitrant compound content of leaves, as well as cuticle development and timing of leaf fall. Second, temporal patterns of organic matter inputs, storage, and transport were quantified in a subtropical stream, Possum Creek. Finally, rates at which leaves of five common riparian trees on the Possum Creek floodplain are processed was determined, and the temporal patterns of abundance for stream macroinvertebrates was examined in relation to sequential leaf organic matter additions. Addressing these issues allows us to better understand the importance of organic matter in the subtropics and how processes differ from the Northern Hemisphere temperate zone.













CHAPTER 2
LEAF ASSOCIATIONS

Introduction

Leaf fall is often the dominant carbon source for forested streams (Hynes 1975), forested wetlands, (Brinson et al. 1981) and upland-forest floors (Maity and Joy 1999). Utilization of this energy is governed by a number of factors, including moisture (Singh 1984); initial leaf nutrient content (Petersen and Cummins 1974); presence of inhibitory and recalcitrant compounds (Haslam 1989); biota adapted to processing leaves (Merritt and Lawson 1992); and in aquatic systems, water chemistry (Griffith and Perry 1993, Howarth and Fisher 1976). When leaves are introduced to the system over a short period of time, as is the case in the Northern Hemisphere temperate zone, the above factors can typically be of greatest importance in leaf processing. However, in the subtropics (Elder and Cairns 1982, Liu et al. 1996, Lugo and Zucca 1983), the tropics (Songwe et al. 1988, Spain 1984), and Southern Hemisphere temperate zone (King et al. 1987, Lake and Barmuta 1986), leaf fall extends over a longer period of time, and leaves of different species are added to the system sequentially.

Differences in nutritional quality and inhibitory and recalcitrant compounds, along with timing of leaf fall, are plant-specific factors that influence temporal utilization of these energy sources. By understanding leaf quality and timing of abscission, one can better predict the relative importance of particular species to those


4






5
organisms directly feeding on leaf litter and those grazing on microbes involved in leaf processing.

Environmental changes and disturbance can affect the plant community of an

ecosystem and, consequently, those organisms depending on energy derived from leaves. For example, afforestation by exotic species such as Eucalyptus can alter in stream macroinvertebrate communities (Abelho and Graga 1996) because of much higher levels of inhibitory and recalcitrant compounds and differences in timing of leaf fall. The influence of nutritional content of leaves of tree species along a temporal successional gradient can also influence stream macroinvertebrates (Haefner and Wallace 1981, Lyons et al. 2000) and terrestrial communities (McClaugherty et al. 1985).

This study addresses these issues by measuring the amount of phosphorus, nitrogen, and recalcitrant compounds in leaves, as well as cuticle development, and timing of leaf fall for common tree species in north-central Florida riparian forests. Based on these data, relations between these trees will be established that can be used to identify when taxa are likely to become available for processing. The relations between two invasive species and native riparian trees will also be addressed by discerning how they cluster.

Methods

Leaves were collected from 31 common tree species growing in riparian areas in multiple locations in north central Florida, including the Hogtown Creek, Blues Creek, Santa Fe River, and Oklawaha River watersheds (approximately 290 50' N, 820 20' W). Freshly abscised leaves were collected during the period of leaf fall for each species and placed into resealable plastic bags. Leaves were collected from multiple trees and






6
watersheds, wherever possible. Immediately upon return to the laboratory, leaves were air-dried for 48 hours and samples from separate watersheds were homogenized. After at least 100 g (dry mass) of leaf material for each species were obtained and milled for 10 minutes in a large Wiley Mill (Standard no. 3) through 3 mm mesh. Samples were then ground in a small Wiley Mill (A75-A) through 1 mm mesh. Ten g of ground samples were then placed into 6 oz. Whirl-pak bags. Enough material was collected to create a minimum of 4 samples per species.

Leaves of each species were analyzed for phosphorus, nitrogen, and in-vitro

organic matter digestibility (IVOMD). For nitrogen and phosphorus analyses, samples were digested using a modification of the aluminum block digestion procedure (Gallaher et al. 1975). Sample mass was 0.25 g, the catalyst used was 1.5 g of 9:1K2SO4:CuSO4, and digestion was conducted for at least 4 hours at 3750 C using 6 mL of H2SO4 and 2 mL H202. Nitrogen and phosphorus in the digestate were determined by semiautomated colorimetry (Hambleton 1977) (Appendix 1). The IVOMD was performed by a modification of the two-stage technique (Moore and Mott 1974) and measured those leaf constituents that are easily digestible by ruminant gut microflora (Appendix 2). The remaining recalcitrant material is of limited use to microflora and presumably forest and aquatic microbes.

Cuticle development was categorized qualitatively on a scale of 1-5, with 1 characterizing leaves with the most well developed cuticles, and 5 representing those with the least-developed c.ticles. Categorization was based on texture, pliability. and knife scrapings. At least twice weekly from September 2000 through June 2001, surveys of each of the four watersheds were made to determine which riparian tree species were







7
currently abscising. The first incidence of pronounced leaf fall for a particular species was noted as the beginning of abscission for that species in north-central Florida. Termination of sustained leaf fall for a species anywhere in north-central Florida was designated as the end of abscission for that species. The mean date of leaf fall was also determined. Dates were transformed using a modified Julian date system whereby the beginning of the year was designated as September 1.

Data analysis was performed with a hierarchical cluster analysis using SPSS v.

9.0 (SPSS 1998). Cluster analysis was selected to segregate taxa into discrete groupings based on phosphorus, nitrogen, and recalcitrant compound content, as well as cuticle development and timing of leaf fall. These groupings should segregate based on the commonalities of when taxa are likely to become available to be processed. All data were standardized using Z-scores and calculated using the squared Euclidean distance method. Linkages were determined with between-groups linkages.

Results

Leaves of 31 common riparian trees were collected from October 2000- May 2001 (Table 2-1). All taxa with the exception of Cephalanthus occidentalis and Chionanthus virginicus were canopy species. Almost half the taxa were near the southern-most limit of their distributions, including Platanus occidentalis, which has been introduced as an ornamental from its native range under 200 km north and west of north-central Florida. No taxa were endemic to Florida, with all ranging north into the temperate zone. Besides P. occidentalis, a non-invasive introduction, the exotic Sapium sebiferum was also found in north-central Florida. All other taxa were native.







8


Table 2-1. Scientific and common names of the 31 common riparian species from north
central Florida used in this study and the degree of cuticle development.
Cuticle development is on a scale of 1-5, with 1 representing the most
heavily-developed cuticle and 5 the least developed


Scientific name Acer negundo Acer rubrum Betula nigra Carpinus caroliniana Carya aquatica Carya glabra Celtis laevigata Cephalanthus occidentalis Chionanthus virginicus Fraxinus caroliniana Fraxinus pennsylvanica Liquidambar styraciflua Liriodendron tulipifera Magnolia grandiflora Nyssa ogeche Nyssa sylvatica var. biflora Ostrya virginica Pinus elliottii Planera aquatica Platanus occidentalis Populus deltoides Quercus laurifolia Quercus lyrata Quercus michauxii Quercus nigra Salix caroliniana Sapium sebiferum Tilia americana Taxodium distichum Ulmus alata Ulmus americana


Common name box elder red maple river birch ironwood water hickory pignut hickory hackberry buttonbush fringetree Carolina ash green ash sweet gum tulip poplar magnolia Ogechee lime swamp tupelo American hophombeam slash pine water elm sycamore cottonwood laurel oak overcup oak swamp chestnut oak water oak Carolina willow Chinese tallowwood basswood bald cypress winged elm American elm


Cuticule development
5
4 4
4
3 3 5
3 3 3 3 3 3 1
2 2 5

2 2 2 2 2 2 2 4 4
5
3 3 3






9
Four discrete groupings of trees and two outlier taxa were delineated with cluster analysis (Figure 2-1). Group 1 consists of 10 species that abscise early, have light to moderate cuticle development, and generally are nutrient-rich and low in recalcitrant compounds. Group 2 is 11 species that, in general, had greater cuticle development, abscission that ranged from early to middle of the period of total leaf fall, and were variable in both nutrients and recalcitrant compounds. Group 3 was composed of five species that were generally low in nutrients and high in recalcitrant compounds, had welldeveloped cuticles, and abscised over a wide time range. Group 4 consisted of three species that did not appear to fit any pattern, as degree of cuticle development, nutrients, and recalcitrant compounds were variable.

Two species were relatively unrelated to the other species (Tilia americana and Magnolia grandifora) based on the variables included in the analysis (Figure 2-1). Tilia americana was consistently high in nutrients and low in recalcitrant compounds and cuticle development. Magnolia grandiflora abscised at a time considerably different from that of any other species, was low in nutrients and recalcitrant compounds, and had a very well developed cuticle.

Leaf fall occurred in two phases, from early October through late February and one species accounted for leaf fall from mid-March through early June (Figure 1-2). All taxa except M grandiflora, Quercus nigra, and Q. laurifolia abscised between early October and the end of December. Quercus nigra dropped its leaves from early January through early February and partially overlapped Q. laurifolia, which began abscission in mid-January and ended in late February. Magnolia grandiflora abscised from mid-March through early June. Ulmus alata and Carya aquatica had the shortest periods of leaf fall,










0


5


Carpinus caroliniana
Salix caroliniana Planera aquatica
Fraxinus caroliniana
Carya glabra
Ostrya virginica
Acer negundo
Acer rubrum
Sapium sebiferum
Celtis laevigata
Populus deltoides Ulmus americana
Chionanthus virginicus Nyssa sylvatica var. biflora
Liriodendron tulipifera Liquidambar styraciflua
Ulmus alata
Fraxinus pennsylvanica
Quercus lyrata
Quercus michauxii
Taxodium distichum
Quercus laurifolia
Quercus nigra Nyssa ogeche
Platanus occidentalis
Pinus elliottii


10
Rescaled Distance Cluster Combine
10 15 20 25
1 1 1


1















2











3





4


Betula nigra
Carya aquatica Cephalanthus occidentalis
Tilia americana
Magnolia grandiflora

Figure 2-1. Dendrograrn showing associations of 31 common riparian tree species
from north-central Florida based on phosphorus, nitrogen, recalcitrant
and inhibitory compounds, cuticle development, and timing of leaf fall.
Numbers on left indicate four discrete groupings.



-
-
--- ----


--


---



-


---
-


-



-





--







S1I


Platanus occidentalis
Cephalanthus occidentalis
Chionanthus virginicus
Fraxinus caroliniana
Fraxinus pennsylvanica
Nyssa ogeche

Nyssa sylvatica var. biflora
Tilia americana
Celtis laevigata
Betula nigra
Liriodendron tulipifera
Populus deltoides
Acer negundo
Pinus elliottii
Planera aquatica
Liquidambar styraciflua
Sapium sebiferum
Salix caroliniana
Ulmus alata
Carya glabra
Ulmus americana
Carya aquatica
Taxodium distichum
Acer rubrum
Carpinus caroliniana
Ostrya virginica
Quercus lyrata
Quercus michauxii
Quercus nigra
Quercus laurifolia
Magnolia grandiflora
I I I I
S O N D J F M A M J J A

Month

Figure 2-2. Period of leaf fall of 31 common riparian tree species from northcentral Florida from September 2000-August 2001.






12

lasting less than a month. Conversely, P. occidentalis (early October through early December) and M grandiflora abscised over the longest time period.

Mean phosphorus content of leaves was 0.16% of leaf dry mass and ranged from

0.02% (Pinus elliottii) to 0.50% (T. americana) (Figure 2-3). Nitrogen content was higher than P for all species. Mean N was 0.94% of leaf dry mass and ranged from

0.44% (P. elliottii) to 1.70% (Carya aquatica) (Figure 2-4). The IVOMD was below 50% for all species except T. americana, M grandiflora, and S. sebiferum (Figure 2-5). Mean IVOMD was 31.9% and ranged from 11.4% (P. occidentalis) to 66.0% (T. americana). Thus, with recalcitrant compounds being 100% - IVOMD, P. occidentalis had the highest levels of recalcitrant compounds and T. americana had the lowest levels. Most leaves had intermediate cuticle development, with M grandiflora, P. elliottii, and Nyssa ogeche having the best-developed cuticles and Acer negundo, Celtis laevigata, Ostrya virginica, and T. americana having the least-developed cuticles (Table 2-1).

Discussion

All species in this study are ubiquitous flora of riparian areas across north-central Florida. Not every species is found in every habitat, as some (e.g., Taxodium distichum, N. ogeche, and Betula nigra) are obligate wetland plants, while others (e.g., C. virginicus, 0. virginica, and M grandiflora) are upland trees that frequently grow on bluffs or other raised portions of riparian areas (Tobe et al. 1998).

Unlike the Northern Hemisphere temperate zone where leaf fall is limited to a

brief autumnal period (Fisher and Likens 1973), in north-central Florida it occurs over an extended period. Both tropical and Southern Hemisphere temperate systems experience leaf fall throughout the year, but as opposed to north-central Florida and the Northern








13


Tilia americana Liquidam bar styraciflua Ulm us alata
Quercus lyrata
Fraxinus pennsylvanica
Betula nigra
Quercus michauxii
Taxodium distichum
Carya aquatica Nyssa sylvatica var. biflora
Chionanthus virginicus
Ulm us americana
Ce/tis laevigata
Populus deltoides
Carya glabra
Sapium sebiferum
Ostrya virginica
Carpinus caroliniana Platanus occidentalis
Acer rubrum
Acer negundo
MAgnolia grandiflora
Nyssa ogeche Cephalanthus occidentalis
Quercus nigra
Salix caroliniana
Liriodendron tulipifera
Planera aquatica Quercus laurifolia
Fraxinus caroliniana
Pinus elliottil


-Eu..
El. EKE El. El' El' IEE-~
'El




I.




F


0.1 0.2


0.3 0.4 0.5 0.6


Percent of dry weight


Figure 2-3. Percentage of dry mass ( 1 standard error) consisting of phosphorus
for 31 common riparian tree species from north-central Florida.


0








14


Carya aquatca Cephalanthus occidentalis
Betula nigra
Tilia americana
Planera aquatca Salix caroliniana
Carya glabra Quercus lyrata
Carpinus caroliniana Platanus occidentalis
Quercus michauxii
Taxodium distichum
Celts laevigata Ostrya virginica
Nyssa ogeche
Populus deltoides
Fraxinus caroliniana
Ulm us aleta
Acernegundo
Quercus laurifolia Ulmus americana
Magnolia grandiflora
Liriodendron tulipifera
Fraxinus pennsylvanica
Quercus nigra
Sapium sebiferum
Acer rubrum Nyssa sylvatca var. bifiora Chionanthus virginicus Liquidamb ar styraciflua
Pinus elia/ti


I
- iini EEl.. ~IE I.
im I.' IKI I.
__________________U
U
















El'


-


0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

Percent of dry weight


Figure 2-4. Percentage of dry mass ( 1 standard error) consisting of nitrogen for 31
common riparian tree species from north-central Florida.








15


Platanus occidentalis
Pinus elliotti
Ostrya virginica
Betula nigra
Quercus lyrata Quercus nigra Carya glabra
Taxodium distichum
Quercus laurifolia
Ulmus alata
Quercus michauxii
Acer negundo
Fraxinus caroliniana
Carya aquatica Nyssa ogeche
Liquidamb ar styraciflua
Planera aquatica Salix caroliniana Nyssa sylvatica var. biflora
Carpinus caroliniana
Ulmus americana
Fraxinus pennsylvanica Cephalanthus occidentalis
Populus deltoides
Acerrubrum
Chionanthus virginicus
Celtis laevigata
Liriodendron tulipifera
Sapium sebiferum
Agnolia grandiflora
Tilia americana

0 10 20 30 40 50 60 70 80 90 100

Percent of dry weight


Figure 2-5. In vitro organic matter digestibility (IVOMD) ( 1 standard error) as a
percentage of dry mass for 31 common riparian tree species from northcentral Florida.






16
Hemisphere temperate zone, peak leaf fall occurs during the warmest months of the year, which also corresponds to the dry season (Lake and Barmuta 1986, Spain 1984, Stewart and Davies 1990). As such, the Northern Hemisphere subtropics appear to occupy a niche intermediate between those of Northern temperate, tropical, and Southern Hemisphere temperate zone streams.

Such differences are based on variance in photoperiod and on vegetation.

Although the temperate zone is home to various evergreen conifers, north-central Florida hosts both evergreen conifers such as Pinus and evergreen hardwoods. In fact, the three riparian species (Q. nigra, Q. laurfolia, and Magnolia) that have the latest leaf fall in north-central Florida are evergreen hardwoods. Without these species, it is likely that leaf fall would more closely resemble that of the Northern Hemisphere temperate zone. Not all riparian systems in north-central Florida support these three taxa, particularly Magnolia, which is not often found in more frequently inundated floodplains. Thus, the length of time leaves are available, and consequently the spatial and temporal availability of energy sources for microbes and invertebrates, varies considerably based on the vegetative community.

The processing rates of leaves depend on their chemical and structural

components in concert with various environmental factors (Kaushik and Hynes 1971). Nutrients and levels of inhibitory and structural compounds are important regulators of microbial and invertebrate communities associated with leaf litter in both aquatic and terrestrial systems (Cummins et al. 1989, Merritt et al. 1984). Those leaves with high concentrations of materials that processing organisms require and low concentrations of






17
those that inhibit colonization are processed preferentially and thus more rapidly (Petersen and Cummins 1974).

Leaf nutrient levels can influence decomposition rates in both aquatic and

terrestrial systems. Nitrogen concentration is an important determinant of leaf quality in both aquatic (Day 1982, Suberkropp et al. 1976) and terrestrial systems (Melillo et al. 1982), but P appears to be of marginal importance compared to N and inhibitory and recalcitrant compounds such as lignins and tannins. In both the Great Dismal Swamp, Virginia, USA (Day 1982) and in terrestrial shrubland in Spain (Gallardo and Merino 1999), however, P concentration are positively correlated with processing rates. Addition of aqueous P also accelerates decomposition (Elwood et al. 1981).

Neither N nor P appear to be sole predictors of processing timing. Pinus elliottii had the lowest concentration of both N and P in its needles and T. americana had the highest P concentration and very high N concentration. Although the low nutrient content of Pinus is typically reflected in very slow processing rates, and high nutrients in Tilia is reflected in very rapid processing (Webster and Benfield 1986), the relation between nutrients and processing rate is less clear for the remainder of the taxa. It would appear that nutrient content itself is not as important as the ratio between nutrients and recalcitrant compounds, a relation that has been noted elsewhere (Canhoto and Graga 1995).

Relatively recalcitrant and inhibitory compounds in leaves impede colonization and utilization of leaf litter (Chew and Rodman 1979). TLignins and tannins are among the most important of these because they are relatively recalcitrant (Boulton and Boon 1991), bind with proteins (Haslam 1989), have antibiotic properties (Haslam 1989), and






18
inhibit processing of structural components such as cellulose and hemicellulose (Kirk et al. 1977). Such properties slow the rate of processing and decrease the efficiency of energy uptake of both microbes and invertebrates. Recalcitrant compounds in leaves in the present study were also related to processing rates. Pinus elliottii and P. occidentalis had the lowest levels of recalcitrant compounds and T americana the least and this is reflected in the fact that Pinus and Platanus are slow-processed and Tilia is fastprocessed (Webster and Benfield 1986). In fact, processing rates of tree families in Webster and Benfield (1986) are very similar to the plot of trees and their recalcitrant compounds measured in the present study.

Cuticle development is also important in influencing processing rates. In the present study, Magnolia was one of the species with the best-developed cuticle. In Chapter 4, shows that although nutrient content of M grandiflora is high and levels of recalcitrant and inhibitory compounds are low, their leaves were the slowest-processed of the five tree species examined. Furthermore, Quercus michauxii was the second slowest to be processed (Chapter 4) and also has a well-developed cuticle. Bunn (1988) and Canhoto and Graga (1995) found that heavy cuticles retard leaf processing.

The clustering of these taxa into four groups represents a continuum over which

leaves are available to processing organisms. The low cuticle development, high

nutrients, and low recalcitrant compound levels of T. americana would make it available

very quickly. Group 1 is made up of taxa that are likely to be available for processing

soon after abscission. Taxa in Group 2 are likely to become available later, as they have higher cuticle development. Those in Group 3 would be available even later. In the case of both Q. laurifolia and Q. nigra, abscission occurs well after that of most taxa, and thus






19
would not enter the system until January-February. The other taxa in Group 3 have very well developed cuticles and high recalcitrant compound levels, and would not become available until later for that reason. Magnolia grandiflora, predominantly because of its late abscission, would be the last taxon to be utilized.

Although not frequently found associated with streams in undisturbed habitats, Pinus elliottii was included in this study because it is planted along streams in vast monoculture plantations throughout much of northern Florida. Sapium sebiferum is an exotic native to Asia and has escaped cultivation and is forming monocultures in riparian areas of the southern U.S. (Bruce et al. 1995) and is increasingly encountered in floodplains of northern Florida. Monocultures of exotic Eucalyptus in Portugal and Spain (Abelho and Graga 1996, Pozo et al. 1998) and Salix in New Zealand (Lester et al. 1994.) have altered aquatic macroinvertebrate communities as a result of differences in leaf fall quality and timing. By altering timing and quality of leaf inputs, and consequently macroinvertebrate communities, litter processing and nutrient recycling could be altered, as could entire food webs.

Both P. elliottii and S. sebiferum are unlikely to cause similar problems in mixed forests in Florida based on their close statistical associations to other species, but in monospecific stands, they could elicit effects similar to those of Eucalyptus and Salix. P. elliottii is low in nutrients and high in recalcitrant compounds, making their needles of limited use to both terrestrial and aquatic decomposers. Monocultures of S. sebiferum would exclude late-abscising trees as well as those taxa with higher levels of recalcitrant compounds and better-developed cuticles. These trees have leaves that would be available for processing later than more easily processed leaves (Gessner and Schwoerbel






20

1989, Suberkropp et al. 1976), such as those of S. sebiferum. Thus, an energy source would be available only during a brief period of time.

Some exotic understory trees such as Ligustrum spp. are increasingly being found in north-central Florida riparian zones (Nelson 1996). Others that are unable to tolerate periodic freezing conditions of north-central Florida, such as Casuarina equisetifolia (Nelson 1994), Cupaniopsis anacardioides (Lockhart et al. 1997), Melaleuca quinquenervia (Woodall 1982), and Schinus terebinthifolius (Woodall 1982), have invaded many areas of southern Florida and could expand their ranges northward with continued global warming and monocultures of S. sebiferum and P. elliottii in similar ways. With this method, future species introductions may be analyzed for chemical constituents and leaf fall timing, and relations to other trees can be measured. As such, it may be ascertained what effects a highly competitive exotic may have on biota that depend on seasonal leaf inputs.














CHAPTER 3
STORAGE AND TRANSPORT

Introduction

Small streams in forested watersheds typically rely on materials derived from outside the stream channel as their primary energy base (e.g., Cummins et al. 1973, Fisher and Likens 1973, Hynes 1975). Such allochthonous materials entering a stream are primarily in the form of coarse particulate organic matter (CPOM), which is predominantly composed of leaves and wood, but also reproductive structures such as blossoms, fruits, and nuts, and dead fauna. Once in the stream, most CPOM is trapped by in stream structure such as fallen tree limbs, roots, and rocks. Here, it is processed by a suite of organisms, ranging from bacteria and fungi to macroinvertebrates, and broken down into smaller particles (fine particulate organic matter (FPOM) and dissolved organic matter (DOM)), that are exported downstream and subsequently utilized. This process was synthesized into the "River Continuum Concept" (Vannote et al. 1980), which describes the structure and function of streams from their headwaters to terminus.

Although a great deal of research has been performed in the temperate zone on organic matter budgets and litter processing, but relatively little has been done in the subtropics (Cuffney and Wallace 1987b, Dudgeon 1982, Dudgeon 1989, Wallace and Benke 1984), and almost nothing has been done in Florida (Elder and Cairns 1982). The subtropics are unlike temperate regions in that leaf fall is not limited to a brief autumnal period. Instead, it can be extended over several months, and in the case of north-central


21






22
Florida, litter fall begins in early October and continues through February (Elder and Cairns 1982, Lugo and Zucca 1983), with species like magnolia (Magnolia grandifolia) continuing to drop leaves though late May (Chapter 2).

As a result, the dynamics of CPOM fluxes, storage, and processing should be considerably different from in temperate areas, where leaf litter is introduced to the system during a temporally brief period. The current study addresses the quantity and periodicity of particulate organic matter inputs, storage, and transport in a lightly developed subtropical stream in north-central Florida. Stream communities are greatly dependent on organic matter, and quantification of inputs, storage, and transport will provide us with an idea of when particulate organic matter is available to stream biota. Whether inputs are reflected in in stream storage and transport will also be investigated. Finally, an organic matter budget will be created to account for many of the important sources and fates of organic matter in Possum Creek.

Methods

Study Site

Possum Creek is a small stream (approximately 4.8 km long and 1.5-2 m channel width) within the larger Hogtown Creek watershed in Alachua County, Florida (Figure 21). Water depth varied depending on discharge and location within the stream, but at base flow, depths in runs were around 0.15 m and pool depths ranged from 0.25-1 m. The stream runs through predominantly undisturbed floodplain and low density residential areas. The study site was a nature preserve approximately 1 km upstream of









23


2


0


NU
U



U

IL
U
















El I-tter traps i Lateral do 0 Saworacc \ Begirnend
o Swerac


4 .,r&


T
W.U.I















0


position traps ass
in-stream storage transect


Figure 3-1. Map of Possum Creek, Florida, study site, and placement of sampling

equipment in the floodplain of the stream.






24
the confluence with Hogtown Creek. At this point, it is a second-order stream. Like most Florida creeks, the streambed is predominantly sand. The channel is meandering and incised, with the streambed about one meter below the adjacent floodplain. Dominant floodplain tree species include box elder (Acer negundo), red maple (A. rubrum), sweet gum (Liquidambar styraciflua), American elm (Ulmus americana), ironwood (Carpinus caroliniana), water oak (Quercus nigra), swamp chestnut oak (Q. michauxii), laurel oak (Q. laurifolia), Carolina ash (Fraxinus caroliniana), and magnolia (Magnolia grandiflora).

Experimental Approach

Inputs, storage, and transport of particulate organic matter were collected

biweekly from Possum Creek from November 2000-October 2001. Direct deposition (litterfall) and lateral deposition were measured inputs. In stream benthic, leaf packs, debris dams, and floodplain were also measured. Downstream transport was the only transport mechanism quantified. Litterfall and lateral deposition traps were placed in the field two weeks before the first collection. Samples were placed in resealable plastic bags on collection and returned to the laboratory for processing.

Direct deposition was measured by placing 10 litter traps (open side facing up) randomly along a 400-m channel length and within 20 m of the channel (Figure 2-1). Traps were modified plastic storage boxes (60.5 x 45 x 45 cm) that had 1-mm diameter holes drilled through the bottom to facilitate water drainage. It was assumed that litter quantity and species composition were representative of material directly deposited into the stream channel.






25
Lateral deposition (the CPOM wind-transported from the floodplain to the stream) was measured by placing five traps at the boundary of the floodplain and stream channel (Figure 3-1). These traps were constructed of plastic boxes (45 x 25 x 13 cm) with 1-mm diameter holes drilled to facilitate water drainage, and were placed on their edges (on long side) to capture debris entering the channel from the floodplain.

Leaf packs are those aggregations of leaves and other organic materials occurring on the stream bottom in quiescent areas. Debris dams are found where woody debris traps organic matter, generally near the stream surface. Leaf pack and debris dam volumes were enumerated from a 200-m stretch of stream that was measured throughout the study. To avoid modification of leaf packs and debris dams, thereby possibly biasing the next temporal sampling, ten samples of CPOM were obtained from each of five leaf packs and debris dams from randomly selected locations downstream of the 200 m stretch. The CPOM was placed into a 12.5 x 7.5 x 7.5-cm box, to provide approximately equivalent measurements, then each of the ten samples were transferred to separate resealable plastic bags for processing in the laboratory. These samples were collected to provide relative proportions of CPOM constituents in the stream at any one time.

Benthic organic matter was collected using a stainless steel cylindrical-corer 7.1 cm in diameter and 26.5 cm long with a 64 cm attached handle. Ten cores were taken midstream to a depth of 10 cm every 2 m along the stream length from a randomly selected starting point in the channel within the aforementioned 200 m stretch.

Floodplain CPOM lying on the soil surface was collected from five 0.5 m

quadrats positioned randomly within 10 m of the stream channel. Samples were collected






26
only in areas of native vegetation and avoided areas with coverage by exotic ground cover such Tradescantiafluminensis.

The CPOM and FPOM export was measured by anchoring a 60 x 45 cm drift net to the stream channel and collecting organic trapped in them over the course of six hours. Stream flow was measured in the center of the water column where the drift net was deployed at the beginning and end of the 6 hours with a Marsh-McBirney Model 2000 Flo-Mate flow meter. Five separate 10 second-averaged measurements were taken both prior to and immediately after the 6-hour period. It was assumed that the mean of these measurements represented the mean flow of the stream between the two measurements. Using these data, the amount of organic matter transported downstream per day could be quantified. The FPOM and CPOM were collected using a 35 pim mesh drift net. At the end of 6 hours, the contents of each net were washed into a bottle and returned to the laboratory for analysis.

For laboratory processing of CPOM in litter traps and in stream accumulations, all materials collected in the field were first dried at 1100 C for 24 hours in a drying oven. Five-gram (dry mass) subsamples of each constituent were then ashed in a muffle furnace at 550 'C for 1 hour and were reweighed to obtain ash-free dry mass (AFDM). For leaf pack and debris dam samples, no subsamples were taken and the entire sample was ashed. Benthic organic matter (BOM) from cores was first separated from the mineral substrate by thoroughly washing it through two sieves. Material retained by a 1 mm sieve constituted CPOM and that retained by a 63 ptrm sieve was designated FPOM.

Stream gage height data were obtained from the Hydrologic Data Services

Division of the St. John's Water Management District. These data were measured at the






27
NW 16th Avenue gage station on Possum Creek approximately 0.5 km upstream of the study site. All gage height levels were measured daily at midnight. Data Analysis

Spearman rank correlations were used to test the for a relationship between initial leaf cuticle development and nutrient and recalcitrant compound content were unrelated to leaf processing rate. Simple linear regressions were used to relate both CPOM and FROM export to stream gage height and litterfall.

Results and Discussion

Latitudinal Comparisons

The extended period of leaf fall in the Possum Creek floodplain differs

considerably from the brief autumnal leaf fall typical of Northern Hemisphere temperate systems (Figure 3-2). The duration of peak leaf fall is far more similar to other subtropical and tropical ecosystems than to that of the temperate zone. The timing of leaf fall in the Possum Creek floodplain most closely resembles that of other Florida and Georgia systems (Lugo and Zucca 1983, Watt and Golladay 1999), but also is very similar to several Southern Hemisphere locations (e.g., Daniel and Adams 1984, King et al. 1987, Neiff and Neiff 1990, Enright 1999). Latitude appears to have some influence on both the timing and duration of leaf fall. Amount of litterfall is also in the range of many Southern Hemisphere forests and greater than those of the Northern Hemisphere temperate zone (Table 3-1). Litterfall increases in eastern North America with increasing precipitation and decreasing latitude in response to lengthened growing seasons (Benfield 1997). Florida fits this trend well (Table 3-1), and it is likely that this trend is evident worldwide as well.







28


Norway (62* N) British Columbia Canada (49' 16')b
Ontario Canada (430 47' N)'
Spain (43* 19' N)d
Portugal (390 30'N)e
Kentucky USA (370 5' N)f
South Carolina USA (33* N)'
Georgia USA (310 N)h
India (300 41' N)'
Florida USA (29' 14' N) Florida USA (290 14' N)k
Hong Kong (21' 27' N)'
Mexico (190 30' N)"
Mexico (180 34' N)"
India (100 35' N)*
Cameroon (40 30' N)l
Bolivia (160 13' S)'
Australia (170 20' S)r Argentina (270 30' S)
Australia (280 S)'
South Africa (330 50' S)u New Zealand (360 46' S)V
Australia (370 30' S)w
New Zealand (410 20' S)X
Tasmania (43* 25' S)Y


I I I M I
J F M A M J


I I I I I
J A S 0 N D


Approximate timing of leaf abscission from selected forests of the world. Solid line indicates timing of leaf fall. a Haapala and Muotka 1998; b Richardson 1992; ' Hill and Brooks 1996; d Pozo et al. 1997;e Abelho and Graga 1998; f Muller and Martin 1983; g Shure and Gottschalk 1985; hWatt and Golladay 1999; Garkoti and Singh 1995; 1 present study; k Lugo and Zucca 1983; ' Lam and Dudgeon 1985; ' Williams-Linera and Tolome 1996; " Alvarez-Sanchez and Guevara Sada 1993; 0 Kumar and Deepu 1992; P Songwe et al. 1988; q Justiniano and Frederickson 2000; Spain 1984; S Neiff and Neiff 1990; ' Greenway 1994; u King et al. 1987; Enright 1999; ' Campbell et al. 1992; X Daniel and Adams 1984; Y Turnbull and Madden 1983.


Figure 3-2.


'







29


Table 3-1. Litterfall from representative forests along a latitudinal gradient


Location
Portugal Kentucky, USA South Carolina, USA Georgia, USA India
Florida, USA Florida, USA Florida, USA Hong Kong Mexico Australia India
Cameroon Argentina Australia South Africa New Zealand Australia New Zealand Tasmania


Latitude
390 N 370 N 330 N 31"N 31*N 300 N 290 N 290 N 21ON 190 N 170 S I11N
40N 270 S 280 S 340 S 370 S 380 S
410S 430 S


Litterfall (T ha~' y')
7.2 2.9
4.2-5.4
4.1-5.8
3.5-6.3
8.0 7.7
2.8-3.3
12.2 8.5
7.3-10.5
12.2-14.4 12.9-14.1
8.2
6.8-8.1
4.3-5.0
6.8-8.8
6.0-7.0
6.2
4.1-5.6


Literature source
Abeiho and Graqa 1998 Muller and Martin 1983 Shure and Gottschalk 1985 Watt and Golladay 1999 Garkoti and Singh 1995 Elder and Cairns 1982 Present study Lugo and Zucca 1983 Lam and Dudgeon 1985 Williams-Linera and Tolome 1996 Spain 1984 Kumar and Deepu 1992 Songwe et al. 1988 Neiff and Neiff 1990 Greenway 1994 King et al. 1987 Enright 1999 Campbell et al. 1992 Daniel and Adams Turnbull and Madden 1983






30
In the Northern Hemisphere temperate zone, leaf fall occurs during a brief

autumnal period (Anderson and Sedell 1979, Fisher and Likens 1973), but the subtropics and much of the Southern Hemisphere do not follow this pattern. In much of the Southern Hemisphere, vegetation is evergreen (Campbell and Fuchshuber 1994), and leaf fall occurs throughout the year, although much of this litter fall occurs during springsummer and at a smaller magnitude than that of autumnal inputs in Northern Hemisphere temperate systems (King et al. 1987, Lake and Barmuta 1986, Stewart and Davies 1990). In this respect, Possum Creek is similar to these Southern Hemisphere systems with litter inputs occurring over a longer period of time than in Northern Hemisphere temperate systems. In Chapter 4 it is suggested that subtropical Possum Creek lies intermediate to Northern Hemisphere temperate and Southern Hemisphere streams, not only in terms of timing of leaf fall, but also in terms of macroinvertebrate life histories.

This similarity between Possum Creek and the Southern Hemisphere and difference from the Northern Hemisphere temperate leaf fall reflects primarily the presence of evergreen vegetation in the Possum Creek floodplain. At Possum Creek, deciduous trees experienced leaf fall from October through early December, but evergreens such as Quercus nigra, Q. laurifolia, and M grandiflora abscised from January through May. In addition, the chemistry and cuticle development of these early and late-abscising species differ considerably, as in Chapter 4 it is shown that earlyabscising deciduous leaves were processed over twice as quickly as those of lateabscising evergreens. These late-abscising trees had leaves with greater cuticle development, which can impede microbial processing (Bunn 1988).






31
As is the case with litterfall (Benfield 1997), Jones (1997) found that organic

matter storage in eastern North American streams increases with increasing precipitation and decreasing latitude. Possum Creek is near the high end of values in eastern North America, but the relation between organic matter, precipitation, and latitude is not fully supported by literature values (Table 3-2). Varying methods and means of quantification are likely affecting these relations, and standardization of methods may allow for a more fully developed model in the future.

Organic Matter Budget

The organic matter budget shows inputs, storage, and export of particulate organic matter into Possum Creek and its floodplain (Figure 3-3). Dissolved organic matter, biotic storage, respiration, and flood export from the floodplain were not quantified in this study, but are nevertheless components of the budget. The majority of inputs occurred via direct deposition in the form of litterfall. However, DOM can also be a very important source of organic matter to streams (Benke and Meyer 1988) and should not be discounted as a source of energy for Possum Creek. Similar amounts of organic matter were stored both in stream and on the floodplain, and the majority of in stream organic matter was stored in debris dams, followed by leaf packs. Similar to import, exports were dominated by CPOM, but DOM export is often far greater (Benke and Meyer 1988) and is thus likely of considerable importance in Possum Creek.

Streams with dense canopies obtain most of their energy from autochthonous

sources (e.g., Cummins et al. 1973; Fisher and Likens 1973; Hynes 1975). The majority of these inputs are from litterfall, although lateral deposition and upstream transport are also important. In addition, DOM in groundwater can generally equal or surpass these






32


Table 3-2. Organic matter storage in representative first and second-order streams
along a latitudinal gradient


Location
Quebec, Canada (50* N) Quebec, Canada (50* N) Oregon, USA (450 N) Oregon, USA (450 N) New Hampshire, USA (440 N) Michigan, USA (42' N) Virginia, USA (370 N) Tennessee, USA (360 N) North Carolina, USA (350 N) North Carolina, USA (350 N) Florida, USA (290 N)


Storage (g m-2)
968 317
4607 5117
610 126
1730
175
740 506 850


Citation
Naiman 1982 Naiman 1982 Triska et a. 1982 Triska et al. 1982 Fisher and Likens 1973 Cummins et al. 1981 Smock 1990 Mulholland et al. 1985 Lugthart and Wallace 1992 Webster et al. 1990 Present study






33


Biota
Groundwater --..--DOM:? ?


521.7fl Benthic storage
521.7CPOM: 3.4 FPOM: 0.6

Leaf pack storage Lateral deposition , 317.3
37.1
Debris dam storage 528.9






Upstream transport Floodplain storage
I? 433


Respiration


Export CPOM: 325 FPOM: 28.7 DOM: ?


Figure 3-3. Organic matter budget for the Possum Creek system, Florida, from
October 2000-November 2001. Solid arrows indicate measured
movements, and dashed arrows indicate hypothesized movements of
organic matter between input, storage, and export compartments. Inputs
(left column of boxes) and exports (right column) are expressed as g
AFDM m- yrf. Standing stock in storage (middle column) is expressed as g2
g AFDM m


I


i I ,






34
inputs (Minshall 1996). Upstream transport was not measured in the present study, but should be approximately that of export, as has been found in other studies (Fisher and Likens 1973, Fisher et al. 1982). However, although upstream import and downstream export are generally in approximate balance, the total amount in each can vary considerably annually (Cummins et al. 1983). Thus, these measurements, as well as others, in Possum Creek represent a "snapshot" view of the stream. It has been strongly advocated that organic matter budgets not be constructed using data collected over a short temporal frame, but rather multi-year data sets (Cummins et al. 1983).

All dominant storage compartments were measured in the present study, and thus the storage portion of the organic matter budget is likely fairly complete. Most organic matter budgets do not distinguish between buried organic matter and that stored in leaf packs or debris dams. As such, it is not possible to compare compartments among studies.

In general, little organic matter is stored permanently in streams (Fisher and

Likens 1973), but rather processed in situ and exported as smaller particles. Although not quantified in the present study, the majority of this organic matter is exported as DOM (Benke and Meyer 1988). Of the remaining exported materials, CPOM generally is responsible for the larger fraction than is FPOM (Fisher and Likens 1973). This was the case in Possum Creek as well, with an order of magnitude greater CPOM export than FPOM.

Inputs

Litter fall totaled 10.4 kg AFDM m-2 yr-' in the floodplain for the period

November 2000-October 2001. Peak litter fall occurred from early October through late






35
December, and a smaller peak was evident during February (Figure 3-4). The majority of the litter was leaves (73%), although wood (15%) and reproductive structures (11%) were important temporally (Figure 3-5). Leaf fall occurred in the Possum Creek floodplain from mid-October through late February and again from early March through late May. Of the dominant canopy trees, leaves of A. negundo, A. rubrum, Liquidambar, Ulmus, Carpinus, and Fraxinus abscised from mid-October through mid-December. During late November, Q. michauxii leaves began to fall and continued through early January. This was followed by Q. nigra and Q. laurifolia, both of which dropped leaves from early January through late February. Finally, M grandiflora abscised from mid-March through late May.

Those trees whose leaves abscised from mid-October through mid-December

contributed the vast majority of the litter fall during the study, with peak litter fall of 12.2 g AFDM m- day', but the smaller peak in litter fall during February resulted from both Q. nigra and Q. laurifolia leaf fall and the addition of reproductive structures, primarily catkins (Figure 3-5).

Nuts and seeds were the major constituents of the reproductive structures and

were only briefly part of the litter fall (Figure 3-5). Nuts were predominantly acorns of the three Quercus species and, to a lesser extent, pignut hickory (Carya glabra). Peak nut fall occurred during early September through late October. Seeds were dominated in both number and mass by Liquidambar with Fraxinus, Acer, and Tilia seeds as secondary contributors. Seed fall occurred at various times throughout the year, but was orders of magnitude less than that of leaf fall. Epiphytes, primarily Spanish moss (Tillandsia







36


16
~14
E 12 UR 10
8


E 4 'E2

S 0
0 N D J F M A M J J A S 0
2000 2001


Figure 3-4. Organic matter inputs via litterfall (circles) and lateral deposition
(triangles) from October 2000 through November 2001 into Possum
Creek, Florida. Error bars represent one standard error







37


100



80

S60


40


A 20



o N D J F M A M J J A S
2000 2001

ULeaves U Epiphytes
[]Wood U Reproductive structures

Figure 3-5. Percent by dry mass of constituents of organic matter in litterfall from
October 2000 through November 2001 at Possum Creek, Florida.
.Reproductive structures include nuts, seeds, blossoms, and other
structures associated with floral reproduction.






38
usneoides), were present in litter traps throughout the year, but were a very minor component of total litter fall.

CPOM can originate from both autochthonous and allochthonous sources, but in heavily forested first or second order streams, the canopy precludes much autochthonous production from algae and macrophytes (Vannote et al. 1980). However, algae was observed in Possum Creek from December through February when leaf abscission of many canopy trees allowed greater light penetration. Although algae were not specifically measured for this study, it would appear that they could represent a sizable energy source during winter, although allochthonous production, especially leaf litter, was likely the primary energy source. Wood also serves as a CPOM source, but its major role in small streams is often in retaining leaf litter in debris dams (Bilby 1981, Bilby and Likens 1980, Smock et al. 1989). Leaves enter a stream primarily during abscission, and this is generally the most important input of CPOM annually in allochthonous systems (Hynes 1975). Reproductive structures enter when riparian vegetation flowers, and again when seeds or nuts are mature. Wood inputs are unpredictable and are related to storm and wind events.

Little lateral deposition of litter into the stream channel occurred (Figure 3-4) and consisted predominantly of leaves (Figure 3-6). Peak lateral deposition occurred during mid-December 2000 (0.64 g AFDM m- day') and was positively related in general with litter fall (r2 = 0.56, p < 0.001), and particularly leaf fall (r2= 0.94, p < 0.001).

Lateral deposition can account for 10-50% of total annual input of CPOM to a

stream (Conners and Naiman 1984, Fisher and Likens 1973, Gurtz et al. 1988, Webster et al. 1990, Weigelhofer and Waringer 1994), but contributed far less in Possum Creek.







39


"Imp
. . . . . . . . .


N D J F M A M J J A S
2001


Epiphytes Reproductive structures


0 Leaves : Wood


Percent by dry mass of constituents of organic matter in lateral deposition from October 2000 through November 2001 at Possum Creek, Florida. Gap in data represents a flooding event that disabled lateral deposition traps. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction.


100 80 60


40 20


0 -M
0
2000


Figure 3-6.






40
Possum Creek lies at the lower end of the range of values reported in the literature (Table 3-3). Organic matter such as leaves and blossoms is more readily transported to the stream via this route due to its lower mass, although some wood and nuts were collected from lateral deposition traps during the course of the study. Possum Creek lies in a heavily forested area with pronounced ground cover. These factors, in concert with the very low floodplain slope, appeared to decrease transport of materials into the stream channel by acting as windbreaks. Even during periods of high wind, little was experienced near the stream.

Storage

Floodplain organic matter was dominated by leaves throughout much of the year (Figure 3-7). However, a flood in late June 2001 redistributed much of the floodplain litter, and another flood in late September 2001 did the same, to a lesser extent. Much of the litter was not washed into the stream channel but became aggregated in wrack piles, thus variability between samples increased considerably after the flood (Figure 3-8). Wood became proportionally more important in the floodplain litter, as much of the lighter leaf litter was washed into the stream during the flooding. Overall, floodplain organic matter increased to 604 g AFDM m2 in late February 2001 and decreased slightly temporally preceding June flooding (Figure 3-8).

CPOM not directly deposited into the stream during storms or abscission is stored on the floodplain. Stream channels are dynamic systems, with CPOM constantly being imported, exported, and transformed (Smock and Roeding 1986). In contrast, many floodplains are less dynamic, with these processes occurring to a lesser degree, especially where spates infrequently inundate portions of the floodplain (Smock 1990), which is






41

Table 3-3. Lateral deposition from representative forests along a latitudinal gradient

Location Latitude Lateral deposition Citation
g m2 yr-1
Alaska, USA 650 N 19 Irons and Oswood 1997
Quebec, Canada 500 N 344 Naiman 1982
Quebec, Canada 500 N 56 Naiman 1982
Oregon, USA 450 N 667 Triska et al. 1982
Oregon, USA 450 N 1111 Triska et al. 1982
Tennessee, USA 360 N 106 Mulholland et al. 1985
North Carolina, USA 350 N 71 Webster et al. 1990
North Carolina, USA 350 N 137 Wallace et al. 1995
Florida, USA 290 N 37 Present study






42


100





S60


. . 40


20


0
0 N 0 J F M A M J J A S
2000 2001
U Leaves UEpiphytes
:: Wood U Reproductive structures

Figure 3-7. Percent by dry mass of constituents of organic matter stored on the
Possum Creek floodplain, Florida, from October 2000 through November
2001. Reproductive structures include nuts, seeds, blossoms, and other
structures associated with floral reproduction.






43


800 E 700
600 .
U
'~500 S400 E 300 2U 00
E



0 N D J F M A M J J A S
2000 2001


Figure 3-8. Organic matter storage on the Possum Creek floodplain, Florida, from October 2000 through November 2001. Error bars represent one
standard error.






44
often the case in low-order streams such as Possum Creek. These factors, combined with the broader area of storage of floodplains as compared to the stream channels, result in greater floodplain storage of CPOM. Some processing occurs on the floodplain by terrestrial invertebrates, bacteria, and fungi (Merritt and Lawson 1992), but perhaps the greatest importance of floodplains for stream biota is as a source of CPOM (Cuffney 1988, Smock 1990).

CPOM is brought into streams by spates and lateral wind deposition. As a result, CPOM can be present year round, although its addition to the stream channel is mediated by storm events (Hill and Brooks 1996, Ractliffe et al. 1995). Floodplain CPOM is present year round for Possum Creek, although quantities decreased during the summer wet-season as a result of both decreased deposition and increased processing due to higher temperatures and wetter conditions (Peterson and Rolfe 1982, Singh 1984) and transport to the stream channel during floods.

Most in stream organic matter was stored in debris dams (Figure 3-9). This storage component was positively related to leaf fall (r2 = 0.46, p < 0.001), with peak storage (0.16 m3 -100 m stream length-) occurring during late December 2000. Leaf packs were present only during and immediately after peak leaf fall, with peak storage (0.081 m3 100 m stream length') occurring during early December 2000 (Figure 3-9). Debris dam organic matter was dominated by leaves from early October 2000 through early January 2001, but by small diameter woody debris for much of the remainder of the year until leaf fall again increased during late September 2001 (Figure 3-10). Leaf packs gradually became dominated by small-diameter woody debris after mid-November, but







45


E
0
E L.
E


Figure 3-9.


0.25


0.2 0.15 0.1


0.05


0
0 N D J F
2000 2001


In-stream organic matter storage in leaf packs (white) and debris dams (black) per 100 m of stream length from October 2000 through November 2001 in Possum Creek, Florida.


M A M J J A S 0






46


100


80


60





40



S100


80


60


40 20


0
0 N D J F M A M J J A S
2000 2001
* Leaves U Reproductive structures
El Wood U Epiphytes


Figure 3-10. Percent by dry mass of constituents of in leaf packs (upper) and debris
dams (lower) from October 2000 through November 2001 in Possum
Creek, Florida. Gaps in data represent periods where no leaf packs were
present. Reproductive structures include nuts, seeds, blossoms, and
other structures associated with floral reproduction.






47
by May almost completely consisted of woody debris, and disappeared altogether after that.

Debris dam storage is an important reservoir of CPOM in streams, especially in high velocity streams where storage on the streambed is small (Angradi 1996, Bilby 1981, Bilby and Likens 1980, Bretschko and Leichtfried 1988, Gonzalez and Pozo 1996, Smock 1990, Smock et al. 1989). CPOM storage in debris dams is greatest during senescence, when most leaves are entering the stream and decreases as they are processed and fragmented. CPOM in temperate stream debris dams can be available all year, as those leaves in the center of the dam are protected against processing because of low dissolved oxygen (Reice 1974). As outer leaves are processed, the oxygenated environment proceeds towards the inner leaves allowing them to be processed (Reice 1974).

However, despite an extended period of leaf fall, CPOM was not present in Possum Creek throughout the year, and very little organic matter was present during summer. Elevated annual water temperatures in subtropical streams such as Possum Creek, as compared to temperate systems, facilitate more rapid processing of organic matter (Benstead 1996) and, subsequently, the presence of little in stream organic matter during summer. Even during winter, mean stream temperatures rarely dropped below 100 C, allowing fungal processing of organic matter to occur at higher rates than would occur in cooler temperate areas during the same time period (Suberkropp and Chauvet 1995).

Leaf packs can be either an important or unimportant store of CPOM and this is predicated by discharge (Angradi 1996, Smock 1990, Smock et al. 1989). In streams with low discharge, leaf packs can form throughout the streambed and remain intact






48
unless disturbed by a spate. Conversely, higher discharge streams have fewer areas (generally pools and backwaters) where leaf packs may form. As with debris dams, leaf packs can sustain stream communities by providing CPOM throughout the year. High discharge in Possum Creek limited leaf packs to only the most quiescent areas (i.e. deep pools and backwaters).

Benthic CPOM peaked at 7.4 g AFDM m-3 in the surface sediments at the time of peak leaf fall during October 2000 and approximated 3 g AFDM m-3 for the remainder of the year (Figure 3-11). FPOM was consistently an order of magnitude lower than CPOM, about 0.5 g AFDM m-3 throughout the year.

Spates can cause the substrate to be either a source or sink of CPOM (Metzler and Smock 1990). During high leaf fall, spates will generally bury more leaf litter relative to that released. Conversely, spates generally release CPOM to the stream channel. Low processing rates in the oxygen-deprived BOM shifts its utilization to when it is brought to the sediment surface by spates (Strommer and Smock 1989). Along with green leaf fall associated with storm events, this appears to be the dominant manner in which CPOM in Possum Creek is made available to the stream community, even though relatively little organic matter was stored in the sediment. Transport from the floodplain is also of some importance, but the low flood frequency limited its value as a CPOM source to the two flooding events during the study.

Buried CPOM or benthic organic matter (BOM) is often the most important

reservoir for CPOM in streams (Cummins et al. 1983, Metzler and Smock 1990, Rounick and Winterbourn 1983, Smith and Lake 1993, Smock 1990). Most streams have an ample supply of CPOM stored in their sediments for processing throughout the year, and







49


1.2

El

<0.8

i0.6

E 0.4 .2
C
0.2

0
o N D J F M A M J J A S 0
2000 2001

Figure 3-11. In-stream benthic organic matter storage as CPOM (circles) and FPOM
(triangles) from October 2000 through November 2001 in Possum Creek,
Florida. Error bars represent one standard error.







50


25

E 20 I 15

10




E,0
0 N D J F M A M M J J A S 0
2000 2001

Figure 3-12. In-stream organic matter transport as FPOM (white) and CPOM (black)
from October 2000 through November 2001 in Possum Creek, Florida.










*


2 1.8 1.6

1.4 .0i 1.2



~0.8 AR


*


M A M J J A S


Figure 3-13.


Gage height at Possum Creek, Florida, from September 2000 to October 2001. Asterisks (*) represent instances when water height .exceeded bankfull stage.


51


KAR


0.4


0.2

0
S 0 N D
2000


F


J
2001


(6






52

Table 3-4. Organic matter export from representative first and second-order streams
along a latitudinal gradient


Location
Alaska, USA (650 N) Alaska, USA (65* N) Oregon, USA (450 N) Oregon, USA (450 N) New Hampshire, USA (440 N) Virginia, USA (370 N) North Carolina, USA (35* N) Florida, USA (290 N)) Puerto Rico (180 N)


Export (kg yr-)
1071 1133
37
245
1700
13,751
4,326 2,506
211


Citation
Irons and Oswood 1997 Irons and Oswood 1997 Triska et al. 1982 Triska et al. 1982 Fisher and Likens 1973 Jones and Smock 1991 Webster et al. 1990 Present study McDowell and Asbury 1994






53
this is a more stable source of CPOM than either leaf packs or debris dams, which are rather dependent on seasonal leaf abscission (Smock 1990). However, the sand sediment of Possum Creek contained little organic matter as compared to debris dams, leaf packs, and floodplain storage. Smock (1990) found that organic matter in a sand substrate was very high, although no storage occurred in clay substrates. Possum Creek appears to be intermediate in that although the substrate is predominantly sand, it overlies a clay layer 0-60 cm below the sand. This clay may have limited the organic matter storage potential of the sand substrate as the relatively homogenous particle size of the sand substrate may have acted as an impediment for trapping of organic matter. Transport

CPOM and FPOM transport were related to both leaf fall and stream discharge. Two peaks in transport were evident, one from mid-November through mid-December (peak 22 kg AFDM m-3 day-') resulting from leaf litter inputs during that period, and another from early June through late July (peak of 16 kg AFDM m3 day-) (Figure 3-12). The summer peak was related to high discharge during the Florida summer wet-season (Figure 3-13). Transport values are highly variable across North America (Table 3-4), and Florida is intermediate. Organic matter is positively correlated to discharge (Webster and Meyer 1997), and comparing transport rates is difficult among streams that vary considerably in gradient, order, and precipitation.

Downstream transport of both CPOM and FPOM is of great importance in the energy dynamics of stream ecosystems (Vannote et al. 1980). Although studies have shown that discharge and CPOM are poorly correlated, they are sensitive to the effects of both spates and maximum leaf fall (Bilby and Likens 1979). Coarse particulate organic






54
matter export was not related to discharge over the entire study (r2= 0.051, p = 0.278), but leaf fall partially obscured this. Looking only at CPOM export during March-August 2001, there was a very strong correlation (r2 = 0.816, p < 0.001). From SeptemberFebruary there was a strong correlation between CPOM transport and litter fall (r2 =

0.557, p < 0.02). Peak CPOM transport occurred during the period of maximum leaf fall and again during the summer wet-season. Storms occurred almost daily from late May through late July and kept the discharge of Possum Creek higher than base flow (Figure 3-13).

FPOM transport is more accurately predicted from discharge (Wallace et al.

1991). This is the case in Possum Creek, as FPOM export was related to discharge (r2

0.196, p = 0.03). This was particularly the case from March-August 2001 when the effect of leaf fall could be factored out (r2 = 0.60, p < 0.001). Higher FPOM export during summer spates may also be influenced by lower retention in depleted debris dams (Golladay et al. 1989). In many temperate streams, macroinvertebrate shredders have been linked to FPOM generation and, consequently, transport (Cuffney et al. 1990, Wallace et al. 1991). However, the relative lack of shredders in Possum Creek (Chapter 4) suggests that they are of minor importance in FPOM generation and transport, and this may also be the case in other subtropical streams with low shredder densities (Cuffhey and Wallace 1987a, Dudgeon 1994, McArthur et al. 1994). This is partially evident as the lack of correlation between litter fall and FPOM (r2 = 0.01, p = 0.636).

Organic matter inputs and storage in leaf packs and debris dams were greatly influenced by the extended period of leaf fall as compared to temperate systems. This relation was less pronounced for lateral deposition, benthic storage, and floodplain






55

storage. Spatial floodplain organic matter storage, however, appeared to be related to moisture and flooding events associated with the Florida summer wet-season. Although rainfall data were not available from this study site, floodplain leaf litter was noticeably more moist during the summer than during the winter dry-season. Downstream organic matter transport was also related to increased discharge during the wet-season, as well as to leaf fall. This information is critical to understanding the sources and periodicity of organic matter inputs into streams, and consequently, the food resources for in stream fauna. Alteration in either the timing of leaf fall or ability to store organic matter effectively could have implications for the stream biotic community by reducing inputs during periods to which macroinvertebrates may have synchronized their life histories.













CHAPTER 4
LITTER PROCESSING

Introduction

Many macroinvertebrates in low-order streams are dependent on seasonal inputs of leaf litter as an energy source. Although studied in great detail in the temperate zone, a paucity of information exists for the subtropics on litter processing (Dudgeon 1982; Elder and Cairns 1982; Hauer et al. 1986; Cuffhey and Wallace 1987; McArthur et al. 1994). Many temperate areas are characterized by leaf fall limited to a relatively brief autumnal period corresponding to the end of the growing season, and macroinvertebrate life cycles are adapted to this seasonality of organic matter inputs (Cummins 1974). However, in many Southern Hemisphere streams, macroinvertebrate life cycles are not synchronized to leaf fall that is spread over much of the year (Campbell and Fuchshuber 1994).

The subtropics differ from temperate areas in terms of both leaf fall seasonality and length of growing season. North-central Florida is located near the northern boundary of the subtropics in North America and is characterized by a continuous growing season and leaf fall that occurs from October through May (Chapter 2). The input of leaves during this period occurs over a considerably longer period than further north, and may have considerable influence on stream macroinvertebrate life cycles.

Although timing of leaf fall can structure macroinvertebrate communities, the nutritional content and physical structure of each leaf species will help dictate how


56






57
rapidly each is processed and, as a result, how long a species may remain available to stream macroinvertebrates (Petersen and Cummins 1974). Processing is the biological and physical processes by which coarse particulate organic matter (CPOM) is utilized by stream biota to yield fine particulate organic matter (FPOM) and dissolved organic matter (DOM) (Petersen and Cummins 1974). Both physical and biotic means are involved in leaf processing. CPOM is physically fragmented by abrasive forces governed by current. Biological processing initially involves fungi and bacteria (Benfield et al. 1977), and later in many systems, invertebrate shredders (Cummins et al. 1973). Leaves with low nutritional value may not become available for macroinvertebrate processing until they are first conditioned by fungi and bacteria, and it is this microbial community that can provide much of the energy requirements for macroinvertebrates (Anderson and Sedell 1979; Cummins and Klug 1979).

This study will determine the rates at which leaves of five common riparian trees are processed and examine the temporal patterns of abundance for stream macroinvertebrates in relation to sequential leaf organic matter additions. It is expected that the processing rates of the species used in this study and macroinvertebrate colonization rates reflect their nutrient content and cuticle development. Finally, the trophic structure of subtropical stream macroinvertebrate communities will be examined to determine the abundance of functional feeding groups.

Methods

Study Site

Possum Creek is a small, second order stream (approximately 4.8 km in length) in the larger Hogtown Creek watershed, Alachua County, Florida. The study area was in a






58
nature preserve approximately 1 km upstream from the confluence with Hogtown Creek. The floodplain has few obvious anthropogenic alterations with the exception of a sewer line that at places parallels and bisects the streambed. Despite few outward appearances of anthropogenic activity, at least 15 exotic plants inhabit the floodplain. Like many Florida creeks, the streambed is predominantly sand, although outcroppings of clay are present where the stream cuts through the Hawthorne layer. The channel is incised, and the streambed is about a meter below the surrounding floodplain. Dominant canopy tree species include box elder (Acer negundo), red maple (A. rubrum), sweet gum (Liquidambar styraciflua), American elm (Ulmus americana), ironwood (Carpinus caroliniana), water oak (Quercus nigra), swamp chestnut oak (Q. michauxii), laurel oak (Q. laurifolia), Carolina ash (Fraxinus caroliniana), and magnolia (Magnolia grandiflora).

Experimental Approach

Leaves of individual tree species were collected for a mesh-bag study once each species became abundant in litter traps. Ten litter traps were placed randomly throughout the floodplain within 20 m of the stream channel, and contents of each trap were collected biweekly beginning in October 2000. Five species were selected based on ease of collection and their relative timing of leaf fall. Acer rubrum, Liquidambar syraciflua, and Ulmus americana were selected to represent the early abscising taxa, Q. michauxii was the mid-season-abscising taxa, and Magnolia grandiflora represented the only lateabscising taxon. One hundred g (dry mass) of freshly abscised leaves from each species

were collected from the floodplain during peak leaf fall and placed into plastic bags.

Leaves were air-dried for 48 hours to standardize for water content.






59
Initial nutrient content for leaves of each species was analyzed for nitrogen, phosphorus, and recalcitrant compounds (including phenolics). For nitrogen and phosphorus analyses, samples were digested using a modification of the aluminum block digestion procedure (Gallaher et al. 1975). Sample mass was 0.25 g, the catalyst used was 1.5 g of 9:1K2SO4:CuSO4, and digestion was conducted for at least 4h at 3750C using 6 ml of H2SO4 and 2 ml H202. Nitrogen and phosphorus in the digestate were determined by semiautomated colorimetry (Hambleton 1977) (Appendix 1). In-vitro organic matter digestion (IVOMD) was performed by a modification of the two-stage technique (Moore and Mott 1974) (Appendix 2). The IVOMD measures those leaf constituents that are easily digestible by ruminant gut microflora. Material that is not digested, such as tannins and lignins, is recalcitrant and of limited use to processing microflora.

Each 1 cm-mesh bag contained 5 g of leaf material of a single species. -These bags measured 45 cm2, leaves placed in the center, corners folded to the center, and cinched with rigid plastic ties. Seventy-five bags were constructed for each species and placed in the stream, tethered to steel chains stretched across the channel in five randomly selected areas displaying approximately the same velocity (0.25-0.30 m s-1 at base flow). Bags were placed in the stream corresponding to the period of leaf fall for a particular species: A. rubrum, L. styraciflua, and U americna in mid December, Quercus michauxii in early January, and M. grandiflora in early April. Bags were collected every 100, 200, 300, 500, 700, 900 degree-days (sum of mean daily air temperatures) to standardize for temperature effects. One or two bags of each species were collected randomly from each steel chain. Remaining bags were removed from the stream after






60
900 degree days: A. rubrum, L. styraciflua, and U. americana in late February, Q. michauxii in mid March, and M grandiflora in late May. For the first three sampling periods, five bags each were collected, and for the final three periods, ten bags each were collected. Bags were collected more frequently initially because of anticipated rapid processing of the material, and more samples were collected later in the study to reduce variability among bags. After collection, bags were immediately brought to the laboratory for processing.

In the laboratory, samples were gently cleaned to remove macroinvertebrates and the contents washed into a 125 prm sieve. The collected material was then transferred to bottles containing 70% ethyl alcohol and stained with the vital stain Rose Bengal to aid in sorting. All macroinvertebrates were removed manually, transferred to vials containing 70% ethyl alcohol, and identified to lowest practical taxonomic level using Pennak (1989), Epler (1995), and Merritt and Cummins (1996). Macroinvertebrate identification primarily took place under a 4.5x (with IOx oculars) Meiji stereoscope, though chironomid dipterans were placed in CMC- 10 mounting medium on a clean glass slide and viewed under a Fisher Scientific Micromaster CK compound microscope (4x to 1 Ox with lOx oculars). Those samples with large numbers of macroinvertebrates were subsampled on a gridded plate under a stereoscope, and at least 200 individuals were removed for identification.

Leaf material separated from the macroinvertebrates was placed into drying ovens for 24 hours. Initially, unprocessed 5 g leaf samples were placed into the muffle furnace at 550 'C for 1 hour to determine initial ash-free dry mass (AFDM). AFDM was also determined for leaf material remaining at each collection date.






61


Data Analysis

The rate coefficient for leaf pack AFDM (k) for each leaf species was estimated by regressing mesh bag AFDM against time using the exponential decay model (Petersen and Cummins 1974):

Wt= Wie-ki

Where W is AFDM after time t and W is the initial AFDM. To control for temperature effects, t was expressed in degree-days. One-way ANOVA's with Scheffe's post-hoc tests were performed to test the null hypothesis that leaf processing does not differ among species (a = 0.05), and that macroinvertebrate abundance does not differ among leaf species. Spearman rank correlations were used to test the null hypothesis that initial invertebrate abundance and leaf processing rate were unrelated. Simple linear regressions were used to relate macroinvertebrate abundance to organic matter storage (Chapter 4). Organic matter quantity was interpolated assuming a linear change in value between sampling dates.

Results

Mean air temperatures were as follows: December (9.7 *C), January (9.7 *C),

February (16.5 'C), March (17.8 'C), April (18.9 *C), and May (21.4 *C). Acer rubrum and Ulmus americana leaf packs were processed most rapidly (-k = 0.0021 degree days-' and 0.0022 degree days- , respectively), followed closely those of by Liquidambar styraciflua (- k = 0.0019 degree days-'). Quercus michauxii and Magnolia grandiflora leaf packs wcrc processed much more slowly (-k = 0.0009 degree days' and -k = 0.0008 degree days- , respectively). As a result of the increasing stream temperatures from December-May, M grandiflora was processed over a shorter time period than A. rubrum,






62
U. americana, L. styraciflua, and Q. michauxii. Nevertheless, the standardization to degree-days controls for temperature statistically.

Significant interspecific differences (p < 0.01) were noted in leaf content for each of phosphorus, nitrogen, and recalcitrant compounds, as all species were significantly different from one another for each variable. Liquidambar styraciflua had highest percentage as phosphorus (0.32%), and M grandiflora had the least (0.07%) (Figure 41). Nitrogen was highest in U americana (0.97%) and lowest in A. rubrum (0.44%) (Figure 4-1). Q. michauxii had the highest percentage of recalcitrant compounds (74%), and M grandiflora had the lowest (46%) (Figure 4-1). Neither P, N, nor recalcitrant compounds were correlated to processing rates, although cuticle development was marginally correlated (r = 0.32, p = 0.052).

Over the course of the study, 34 macroinvertebrate taxa were collected from leaf packs in Possum Creek. Macroinvertebrate abundance decreased through time for A. rubrum, L. styraciflua, and U americana leaf packs, but peaked midway through processing for Q. michauxii and M grandiflora (Figure 4-2).

Few shredders were collected from leaf packs, and were represented primarily by Tipula sp. and secondarily Hyalella azteca and some chironomids (Table 4-1). Tipula sp. were collected frequently on A. rubrum, L. styraciflua, and U americana leaf packs, but were less frequent on Q. michauxii and especially M grandiflora (Figure 4-3). No trend could be seen over the course of processing for any leaf packs.


Collector-gatherers were dominated by Stenonema sp. and Microcylloepus sp., and some chironomids (Table 4-1). Stenonema sp. abundance increased somewhat over the course of processing for A. rubrum, L. styraciflua, and U americana, but remained







63
Phosphorus Liquidambar styraciflua Quercus michauxii Ulmus americana

Acer rubrum

Magnolia grandiflora

0 0.1 0.2 0.3 0.4

Nitrogen

Ulmus americana Quercus michauxii Liquidambar styraciflua Magnolia grandiflora

Acer nub rum

0 0.2 0.4 0.6 0.8 1

Recalcitrant compounds Querus michauxii Liquidambar styraciflua Ulmus americana

Acer rub rum

Magnolia grandiflora

0 10 20 30 40 50 60 70 80 Percent of dry weight

Figure 4-1. Percent of dry weight of phosphorus, nitrogen, and refractory compounds
in the senescent leaves of five tree species collected from October 2000May 2001. Error bars represent one standard error.












Table 4-1. Macroinvertebrates collected from Possum Creek, Florida, on leaf packs from December 2000 to May 2001. Functional
feeding group designations from Pennak (1989) and Merritt and Cummins (1996)

Taxon Functional feeding group Taxon Functional feeding group


Turbellaria
Planariidae
Dugesia sp.
Annelida
Oligochaeta.
Lumbricidae Mollusca Gastropoda Ancylidae
Ferrissia sp.
Physidae
Physella sp.
Pelecypoda Corbiculidae
Corbiculafluminea
Amphipoda Hyalellidae Hyallela azteca
Decapoda Cambaridae Procambarus sp.
Insecta Ephemeroptera Baetidae
Acentrella sp.
Heptageniidae
Stenonema sp.
Odonata Aeshnidae
Boyeria sp.
Calopterygidae
Calopteryx sp.
Coenagrionidae
Argia sp.
Corduliidae
Macromia s.


predator collector-gatherer scraper scraper collector-filterer shredder, collector-gatherer collector-gatherer collector-gatherer scraper, collector-gatherer predator predator predator


predator


Gomphidae Progomphus sp.
Megaoloptera Corydalidae Corydalus sp.
Trichoptera Hydropsychidae Cheumatopsyche sp.
Coleoptera Dryopidae Pelenomus sp.
Elmidae
Dubiraphia sp.
Microcylloepus sp.
Hydrophilidae Sperchopsis sp.
Diptera Ceratopogonidae Bezzia sp.
Chironomidae Cricotopus/Orthocladius sp.
Corynoneura sp.
Lubrundinia sp
Nilotanypusfimbriatus
Pentaneura inconspicua
Phaenopsectra obediens gp.
Polypedilumfallax
Polypedilum illinoense gp.
Rheocricotopus robacki
Rheotanytarsus sp.
Tanytarsus sp.
Simuliidae Simulium sp.
Tipulidae Tipula sp.


predator


predator


collector-filterer


scraper, collector-gatherer

scraper, collector-gatherer scraper, collector-gatherer

predator


predator

shredder, collector-gatherer collector-gatherer predator
predator
predator, collector-gatherer scraper, collector-gatherer shredder, collector-gatherer shredder, collector-gatherer shredder, collector-gatherer collector-filterer collector-gatherer, filterer

collector-filterer

shredder


Macromia~ s eao
U. r








65


1100 Acer rubrum . 2.5
1000
900 2
800
700 1.5
600 T


300
200 0.5
100
0 0
12/1212/19 1/7 1/16 2/3 2/16 2/27
0 100 200 300 500 700 900


800 700 600 500
400 300
200 100
0


iquidambar styiracifla -


2

1.5

1

0.5 .0


12/12 12/19 17 1/16 2/3 2/16 2/27 0 100 200 300 500 700 900


700 Quercus michauxil . 2.5
600 2
500
400 1.5
300 1
2000.
0.5
100
0 0
1/6 1/16 1/25 2/3 2/16 2/27 3/11 0 100 200 300 500 700 900

160 Magnolia grandiflora 2.5
S140
2
120
100 1.5
80
1
60
40 0.5
20
0 -


4/8
0


4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900


Ulmus americana







i-


2.5

2

1.5

1

0.5 .0


12/1212/19 1/7 1/16 2/3 2/16 2/27
0 100 200 300 500 700 900


Macroinvertebrate abundance and leaf carbon content over time collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack~1 and white bars represent macroinvertebrates-g C~1-leaf pack-'. Both sampling date and cumulative degree days are shown. Error bars represent one standard error.


Cu 0.
I
Cu


Cu

.0


1800 1600
1400 1200 1000 800 600
400 200
0


Figure 4-2.


L uidambar stvraciflua









lftfini


2c.5






66
relatively stable for Q. michauxii and M grandiflora leaf packs (Figure 4-3). Microcylloepus sp. was rarely collected from leaf packs during December-January, but made up a considerable portion of non-chironomid abundance from February-May (Figure 4-3). Because A. rubrum, L. styraciflua, and U. americana leaf packs were placed into Possum Creek in December and removed in late February, few Microcylloepus sp. were noted until 700-900 degree days. Conversely, Microcylloepus sp. were abundant on M grandiflora leaf packs and became more so through the course of processing. There was a marginal relationship between Microcylloepus sp. Abundance and organic matter storage (r2 = 0.21, p = 0.06), and Stenonema sp. was strongly related to organic matter storage (r2 = 0.67, p < 0.001).

Cheumatopsyche sp. and Simulium sp. were the most frequent collector-filterers associated with leaf packs (Table 4-1). Cheumatopsyche sp. abundance increased over the course of processing in A. rubrum, L. styraciflua, and U americana, but remained relatively stable in Q. michauxii and M. grandiflora, although lower than peak levels in the former three species (Figure 4-4). Simulium sp. were frequently collected from leaf packs from December through early May, but virtually disappeared thereafter As a result, Simulium sp. was rarely found on M grandiflora leaf packs after 200 degree days (Figure 4-4). The abundance of both taxa was positively related to organic matter storage (Cheumatopsyche sp., r2 = 0.34, p = 0.03; Simulium sp., r2 = 0.49, p = 0.006)

Predators such as Corydalus sp. and Argia sp. were not frequently collected from Possum Creek leaf packs until February (Figure 4-5), thus lagging behind prey abundance. In general, not only did predator abundance increase from December-May, but also over the coarse of processing for each leaf species (Figure 4-5). The abundance








67


40 Acer rubrum

30

20

10


12/19 1/7 1/16 2)3 2/16 227 100 200 300 500 700 900 20 -Liquidambar styraciflua

15

10

5 T

0
12/19 1/7 1/16 2/3 2/16 2127 100 200 3C 500 700 900 70 Ulmus americana
60
50
40
30 .



0
12/19 1/7 1/16 2 2/16 227 100 200 300 500 700 900 12 T Quercus michauxii


10
8
6
4.
2
0


20 15 10

5

0



10

8

6

4

2

0


4A


1/16 1/25 2,3 2/16 2/27 3/11 100 200 300 500 700 900


Acer rubrum










12/19 117 1/16 2/3 2/16 227 100 200 300 500 700 900 Liquidambar styraciflua










12/19 117 1/16 2/3 2/16 2/27 100 200 300 500 700 900
UImus anmericana


3

2


1
OLJ"
12/19 1/7 1/16 2,3 2/16 2/27 100 200 300 500 700 900
6 Quercus michauxii
5



2

0 11

1/16 1/25 2,3 2/16 2/27 3/11 100 200 300 500 700 900


4 Acer rubrum

3 -

2



0
12/19 117 1/16 2/3 2/16 2/27 100 200 300 500 700 900
3 Liquidambar styraciflua
2.5
2
1.5
1
0.5
0
12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900
7 Ulmus americana
6
5
4
3
2



12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900
1 -Quercus michauxii


0.8 0.6

0.4 0.2

0


1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900


lo Magnolia grandflora 25 Magnolia grandiflora 0.3 Magnolia grandiflora

8 20 - 025

6 15 02
0.15-4 10 .
2 ot_ .05


4/13 4/17 4/23 5/4 5/14 5/22 4/13 4/17 4/23 5/4 5/14 5/22 4/13 4/17 4/23 5/4 5/14 5/22
100 200 300 500 700 900 100 200 300 500 700 900 100 200 300 500 700 900
Stenonema (Ephemeroptera) Microcylloepus (Coleoptera) Tipula (Diptera)


Figure 4-3.


Dominant shredders and collector-gatherers collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack-' and white bars represent macroinvertebrates g C~1'leaf pack~. Both sampling date and cumulative degree days are shown. Error bars represent one standard error.


U




U
C

C
C









68


25

204 15 10


Acer rubrum


0


12/19 1/7 1/16 2/3 2/16 2,27 100 200 300 500 700 900
16- Liquidambar styraciflua
14
12 10 8 6
4
2 0
12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900
14 Ulmus americana
12
- 10
S 8
''6
4

C0
- 12/19 1/7 1/16 2/3 2/16 227
100 200 300 500 700 900


2.


1.


0.


350 Acer rubrum
300 250
200 150 100

50

12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 35 Liquidambar styraciflua
30 25
20 15 10



12/19 1/7 1/16 2/3 2/16 227 100 200 300 500 700 900 600- Ulmus americana


500
400


200 100
0


0- ar
12/19 1/7 1/16 20 2/16 227 100 200 300 500 700 900


3 Quercus michauxii 30 Quercus michauxii
5 25
2 20
5 15



0 - r
1/16 1/25 2/3 2/16 2/27 3/11 1/16 1/25 2/3 2/16 2/27 3/11
100 200 300 500 700 900 100 200 300 500 700 900
5 Magnolia grandiflora 7 Magnolia grandiflora

4 6
5
3 4

2 30 -N IN ,


4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900
Cheumatopsyche (Trichoptera)


4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900
Simulium (Diptera)


Figure 4-4. Dominant filter-feeders collected from Possum Creek from December 2000
to May 2001. Gray bars represent macroinvertebrates-leaf pack-' and white
bars represent macroinvertebrates-g C-1-leaf pack-'. Both sampling date
and cumulative degree days are shown. Error bars represent one standard

error.


----|---








69


10

8

6

4-

2

0


5

4

3

2


0



6 5 4 3
2


0


Acer rubrum








- ii *


12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Liquidambar styracyiua










12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Ulmus americana


12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900
2 T Quercas michauxii


1.5 +


1 +


0.5


0


10 8t


4

2

0


60 50
40 30 20 10
0



14 12 10
8
6
4
2
0



5-


4



~IT~i~i~3


0


Acer rubrum


12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Liquidambar styraciflua


12/19 1/7 1/16 213 2/16 227 100 200 300 500 700 900 Ulmus americana





M--


12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900
Quercus michauxii


1/16 1,25 2/3 2/16 2/27 3/11 1/16 1/25 2/3 2/16 2/27 3/11
100 200 300 500 700 900 100 200 300 500 700 900
3 Magnolia grandiflora 5 Magnolia grandiflora
2-5 4


2


4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Corydalus (Megaloptera)


4/13 4/17 4/23 5/4 5/14 522 100 200 300 500 700 900
Argia (Odonata)


Dominant predators collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack-' and white bars represent macroinvertebrates-g C-1-leaf pack-'. Both sampling date and cumulative degree days are shown. Error bars represent one standard error.


Figure 4-5.


. . . . .- e






70
of both Corydalus sp. (r2 = 0.01, p = 0.7) nor Argia sp. (r2= 0.02, p = 0.82) was not related to organic matter storage.

Chironomids were the most frequently collected macroinvertebrates from all leaf packs. They were far more abundant during December-February than later and their abundance was strongly related to organic matter storage (r2 = 0.66, p < 0.001). Peak abundance per gram carbon of A. rubrum, L. styraciflua, and U. americana was on degree-day 900, although absolute abundance was highest earlier and decreased though the course of processing (Figure 4-6). However, absolute abundance in Q. michauxii and M grandiflora leaf packs peaked around degree-day 300. The abundance of chironomids was strongly related to organic matter storage (r2 = 0.66, p < 0.001).

Eleven chironomid taxa were collected from leaf packs from December-May (Table 4-1). Cricotopus/Orthocladius sp. was not collected from leaf packs until early February, and only infrequently after early May (Figure 4-7). Pentaneura inconspicua, the dominant predaceous chironomid, Rheocricotopus robacki, and Tanytarsus sp. were dominant throughout the study for all leaf species, with numbers remaining relatively stable (Figure 4-7).

Corynoneura sp. was a dominant chironomid through late January, but became a minor part of the chironomid community, then decreased and remained stable through May (Figure 4-7). Rheocricotopus robacki and Corynoneura sp. displayed similar patterns, but unlike Corynoneura, Rheotanytarsus was collected from M grandiflora leaf packs (Figure 4-7).

Polypedilum illinoense gp. was fairly common during December-early January in A. rubrum, L. styraciflua, and U americana, but were virtually absent for the rest of the












Acer rubrum






- ~ ~ ~


12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 8c- Liquidambar styraciflua
700
600
500
400



2100 -Eno i


12/19 1/7 1/16 2/3 2/16 227 100 200 300 500 700 900 Ulmus americana










12/19 1/7 1/16 2/3 2/16 227 100 200 300 500 700 900


700 Que
600 500
400 3W0


100


rcus michauxii


1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900
180 Magnolia grandiflora
160
140 120 100 8060
40 20

4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900


Figure 4-6. Chironomidae (Diptera) collected from Possum Creek from December
2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack-' and
white bars represent macroinvertebrates-g C~'-leaf pack-. Both sampling
date and cumulative degree days are shown. Error bars represent one
standard error.


71


1200 1000


600 400200
0-


C.'

I


C.'

=


14001200 1000800600
400 200
0







Liquidambar styracflua


100 80 60

40 20


'7')


100


0 0
12/19 1/7 1/16 2/3 2/16 2/27 12/19 1/7 1/16 2/3 2/16 2/27
100 200 300 500 700 900 100 200 300 500 700 900

Ulmus americana Quercus michauxii


100


100 0 80

60

40 20


0 0
12/19 1/7 1/16 2/3 2116 2/27 1/16 1/25 2/3 2116 2127 3/11
100 200 300 500 700 900 100 200 300 500 700 900


Magnolia grandiflora


100

80 60

40


4/13 4/17 4/23


5/4 5/14 5/


100 200 300 500 700


900


Others

Cricotopus/Orthocladius Corynoneura sp. Pentaneura inconspicua E Polypedilum illinoense gp.

Rheocricotopus robacki Rheotanytarsus sp. 22 Tanytarsus sp.


Figure 4-7. Chironomidae (Diptera) relative abundance over time collected from
Possum Creek from December 2000 to May 2001. Both sampling date
and cumulative degree days are shown.


. . . . . . . . . . . . . "I


Acer r~brum


OLO


20

0






73
processing period for these three species (Figure 4-7). No P. illinoense gp. were collected from Q. michauxii leaf packs, but they were fairly abundant throughout in M grandiflora leaf packs.

Discussion

Leaf Chemistry

Leaf litter varies in the degree to which it can be processed. This depends

oncharacteristics of both the organic matter and stream including: the tree species from which it originates (Petersen and Cummins 1974; Webster and Benfield 1986), whether it has been conditioned on the floodplain prior to entrainment in the stream (Merritt et al. 1984; Merritt and Lawson 1992), water velocity (Reice 1974), water temperature (Petersen and Cummins 1974; Webster and Benfield 1986), and chemical constituents of the stream water, primarily nitrogen (Kaushik and Hynes 1971; Howarth and Fisher 1976), phosphorus (Howarth and Fisher 1976), and pH (Griffith and Perry 1993). The present study controlled for these factors except tree species by placing all leaf packs in the same stream under similar conditions and using degree days to normalize temporal temperature differences.

After initial leaching of soluble compounds from leaf litter, colonization by

microorganisms increases both the amount of nitrogen and protein and degradation of structural components such as cellulose, thus making them more palatable for shredders (Kaushik and Hynes 1971; Suberkropp et al. 1976; Anderson and Sedell 1979). This process, termed conditioning, occurs to all leaves, but at varying rates, depending on leaf chemistry and structure. As a result, shredders generally utilize leaves only after they have been conditioned (Petersen and Cummins 1974; Suberkropp et al. 1983).






74
Initial leaf quality is often the factor determining processing rates of leaves (Kaushik and Hynes 1971; Merritt et al. 1984). Individual species contain different concentrations of nutrients as well as both chemical and physical deterrents, and these influence processing rates. Chemical defenses are of greatest importance in many species, but thick, waxy cuticles may also retard processing in others (Bunn 1988; Canhoto and Graga 1995).

The fact that L. styraciflua, A. rubrum, and U. americana had the highest initial macroinvertebrate abundance, especially chironomids, may very much reflect the fact that these three species had the highest processing rates of those tested. Conversely, low initial invertebrate abundances on Q. michauxii and M grandiflora may be indicative of the well-developed cuticles of these leaves. There was a very strong correlation between initial macroinvertebrate abundance and processing rate. The fact that recalcitrant compounds differed considerably between the two species, as did nutrient concentrations, suggests that none of these variables were the primary determinant of their processing rates.

Declining chironomids and Tipula and increasing Microcylloepus sp. and

Corydalus from December through May may be related more to biotic or abiotic factors than strictly nutritional quality or cuticle development of Q. michauxii and M grandiflora. Few taxa encountered in Possum Creek were shredders, and as such, neither quantity nor quality of leaf litter entering the stream should be the primary factor controlling abundance of individual taxa. However, slow-processed leaves result in slower generation of smaller CPOM fractions, FPOM, and DOM, and this could affect collector-gatherer and collector-filterer communities that depend on such energy sources.






75
Leaf packs and debris dams in Possum Creek appear to function more as habitat than an energy source, although trapping of FPOM in these structures likely attracts those organisms that utilize this energy source.

The extent of insect herbivory and/or tree health may alter the chemical

constituents of leaves from expected levels (Bryant et al. 1983). As a result of a severe drought, some trees on the Possum Creek floodplain dropped their leaves during summer 2000, and it is likely that those that retained their leaves may have had altered leaf chemistry compared to non-drought years. Water availability can be a factor influencing not only individual trees, but also entire stands where soil water deficits occur. The effect of the drought on energy sources in Possum Creek and its floodplain is unknown, but both quality and quantity of leaf litter may have been affected. Minoletti (1994) found higher P and N levels in Viburnum acerifolium during a drought year, although Pilon (1996) found that drought did not alter nutrient concentrations in sugar maple (Acer saccharum) leaves. The quantity of leaf litter may also be affected by drought. Melaleuca quinquenervia on an Australian floodplain contributed 10% less leaf litter in a drought than a wet year (Greenway 1994).

Nitrogen content of leaves is among the most important chemical constituents determining leaf quality (Bunn 1988; Sinsabaugh and Linkins 1990). It provides the highest nutritive value of leaves for detritivores, thus initial concentration is positively correlated with processing rate (Cummins and Klug 1979). Growth rate and survivorship are heavily influenced by N intake, and those leaves high in N should be consumed preferentially (Anderson and Sedell 1979). Despite this, of the three fast-processed leaves, although both L. styraciflua and U americana were high in N, A. rubrum had the






76
least N of the five species used in this study. Thus, N does not appear to be of much predictive value for processing rates of taxa used in the study.

Phenolic chemicals, as measured by the IVOMD procedure used in the present study, are relatively recalcitrant compounds that also dictate processing rates (Chew and Rodman 1979; Bryant et al. 1983; Bunn 1988). Many phenols are water-soluble and can either have little influence on processing rates (Suberkropp et al. 1976), or may indeed slow it (Edwards and Heath 1975; Barlocher and Oertli 1978; Bunn 1988). Phenols interfere with the breakdown of proteins (Walker 1975) and carbohydrates, including cellulose (Kirk et al. 1977; Zucker 1983), and may have antibiotic capability (Haslam 1989).

Although not specifically measured in this study, tannins and lignins are the most commonly encountered recalcitrant compounds, and lignins are especially important in limiting leaf processing and may be more important than N or P in dictating processing rates (Melillo et al. 1982). These compounds are likely the most important of the recalcitrant compounds measured by IVOMD employed in this study. However, although Q. michauxii and M grandiflora were the slowest processed leaves in this study, Q. michauxii had the highest levels of recalcitrant compounds and M grandiflora the least. As such, levels of recalcitrant compounds do not appear to serve as predictors for processing rates for taxa in this study.

Processing rates are a reflection of many variables. The very well-developed cuticle of M grandiflora may be dictate its slow processing, but in the similarly slowprocessed Q. michauxii, high levels of recalcitrant compounds and a well-developed cuticle may be of greater importance. The faster-processed A. rubrum, L. styraciflua, and






77
U. americana display variability in N, P, and recalcitrant compound levels, but all have less-developed cuticles than either Q. michauxii or M grandiflora. As such, no single chemical constituent appears to govern the processing of these leaf species, but cuticle development may have a considerable role. Timing of litter fall

Along with chemical content of leaves, timing of leaf fall is an important factor governing leaf processing. In the temperate zone, leaf fall is limited to a relatively brief autumnal period when bulk of allochthonous energy inputs occur (Cummins 1974). In the subtropics and tropics, leaf fall generally extends over a longer time period and often occurs at times other than during autumn. In north-central Florida, leaf fall begins in early October and can continue through early June (Chapter 2). Much of the total leaf fall in the Possum Creek floodplain falls from October through December, but additional leaf inputs from oaks (Quercus spp.) in January-February and M grandiflora from midMarch through early June also add considerable carbon (Chapter 3). Differences in timing of leaf fall results in predictable sources of fresh litter in Possum Creek for over half the year.

As a result of temporally staggered leaf fall, macroinvertebrates in Possum Creek have a somewhat constant energy supply entering the stream from October through May, that supplements the highly variable inputs from lateral deposition and flooding of the floodplain. As a result of lower temperatures and subsequent slower processing (Chergui and Pattee 1990), organic matter in temperate streams remains available for macroinvertebrates for a period of time far surpassing the period of leaf fall. Possum Creek displays higher temperatures than temperate streams, and if not for the extended






78
period of leaf fall, it is likely that organic matter would be less available temporally for macroinvertebrate communities.

Macroinvertebrates

The paucity of shredders in Possum Creek is a departure from most temperate systems. Most shredders are amphipods of insect larvae in the orders of Plecoptera and Trichoptera and the dipteran family Tipulidae. No Plecoptera were collected from Possum Creek, and the only trichopteran was Cheumatopsyche sp., a collector-filterer. Tipulids and amphipods were collected and represent the only shredders in Possum Creek, although several chironomid taxa may be partially fit into this classification. A relative lack of shredders has been noted in other subtropical streams (Cuffney and Wallace 1987; Dudgeon 1994; McArthur et al. 1994), although this is not a universal phenomenon (Hauer et al. 1986). It is likely that shredding organisms are not important leaf processors in many subtropical streams. Rather, bacterial and fungal processing and physical abrasion may be of greatest importance.

Both of the dominant predators, Corydalus sp. and Argia sp., increased

numerically over the course of the study, corresponding to decreased abundance of many macroinvertebrates during this same period. The fact that predators are important in controlling prey population levels is clear (Caswell 1978). Fish are generally the most important predators controlling prey populations in streams (Allan 1982), but macroinvertebrates may also play a large role in structuring macroinvertebrate prey communities (Oberndorfer et al. 1984; Wallace and Webster 1995).

Despite the fact that both taxa are collector-gatherers, Microcylleopus sp., as opposed to Stenonema sp. was not regularly collected from leaf packs until late in the






79
study. By the time Microcylleopus sp. abundance peaked in early May, in stream organic matter was low compared to earlier in the study. As a result, it does not appear that increased Microcylleopus sp. abundance was associated with leaf litter inputs. Instead, it is likely more closely related to life history patterns and/or abiotic factors. The decreased abundance of both Stenonema sp. and the collector-filterer Cheumatopsyche sp. may be related to a reduction in leaf pack and debris dam habitat during this period (Chapter 3) as may have predation, life history, and abiotic factors.

Collector-filterers feed primarily on FPOM and do not appear to be heavily influenced by leaf fall. The drop in Simulium sp. abundance from December through May likely occurred independently of changes in either quality or quantity of leaf fall. The increase in temperature during this period may have indirectly contributed to their decrease in abundance by increasing the metabolism of predators such as mosquitofish (Gambusia holbrooki), Corydalus sp., and Argia sp.. Simulium sp. appears relatively prone to predation, as they attach to the surface of in stream structure and are thus readily available to predators. Metamorphosis into adults during this period of time may also have contributed to their in stream decline, but no adults were seen.

Changes in chironomid dominance from December-May were fairly consistent among leaf types. This suggests that leaf chemistry had less to do with chironomid community composition than other factors. This is particularly evident with chironomids, as one might conclude, based on changes in chironomid abundance for A. rubrum, L. stvraciflua, and U. americana, that certain taxa such as Cricotopus/Orthocladius and Rheotanytarsus became dominant as processing continued. However, chironomid dominance on M grandiflora packs showed that although this






80
conclusion may be valid for Cricotopus/Orthocladius, it would appear to be incorrect for Rheotanytarsus, which maintains population levels throughout processing. The same can be said of Polypedilum illinoense gp. and Corynoneura, both of which decreased in abundance during A. rubrum, L. styraciflua, and U. americana processing, but were present on M grandiflora packs throughout its processing. Although leaf chemistry can play a considerable role in determining macroinvertebrate species composition of leaf packs (Anderson and Sedell 1979; Cummins and Klug 1979), at least in Possum Creek, it did not appear to be as important in dictating macroinvertebrate, specifically chironomid, dominance. This is particularly the case with those taxa that only depend indirectly on leaf litter as an energy source (e.g., collectors and predators).

Life histories of some macroinvertebrates in Possum Creek may be related to peak seasonal inputs of leaf litter. However, separating the effects of seasonal litter accession, warming water temperatures, and predation is difficult within the design of this study. Furthermore, the fact that shredders are generally those macroinvertebrates whose life histories are most synchronized with leaf litter inputs (Cummins 1974), and yet are so uncommon in Possum Creek and other Northern Hemisphere subtropical streams, makes it difficult to assess the extent to which extended period of leaf fall affects their life histories.

The Northern Hemisphere subtropics likely occupy a niche different from those of both northern temperate and both tropical and temperate of the Southern Hemisphere streams. Northern temperate streams are limited to a brief period of autumnal leaf fall that occurs just prior to the coldest period of the year (Fisher and Likens 1973). Conversely, in Southern Hemisphere streams, leaf fall occurs over an extended period of






81
time, but during the warmest time of the year (Lake and Barmuta 1986; Stewart and Davies 1990). Possum Creek is intermediate in that leaf fall begins just prior to the coldest time of year, yet the period of leaf fall extends far beyond autumn. The fact that macroinvertebrate life histories in north temperate streams are synchronized to seasonal leaf inputs (Cummins 1974), Southern Hemisphere streams are not (Campbell 1994), and the possibility that Possum Creek macroinvertebrate life histories are only related to the habitat afforded by leaf aggregations further reinforces the uniqueness of Northern Hemisphere subtropical streams.













CHAPTER 5
CONCLUSIONS

Organic matter inputs in subtropical north-central Florida are available from September through early June. These inputs are reflected in in stream storage, predominantly in debris dams and leaf packs. Sequential additions of organic matter resulting from trees with different abscission periods not only result in an extended period of leaf fall as compared to the Northern Hemisphere temperate zone, but in fresh organic matter being made available for stream organisms. Although in Northern Hemisphere temperate streams, organic matter is processed by both shredding macroinvertebrates and microbes, shredders in Possum Creek were uncommon, suggesting that they played a lesser role in processing as compared to those in Northern Hemisphere temperate streams.

This research may have implications for management and restoration of riparian areas in north-central Florida. Both natural and anthropogenic disturbances have altered riparian vegetation in many areas. Severe fires can alter the quantity of inputs to streams (Minshall et al. 1989), and vegetation succession after such an event may vary the quality and timing of leaf inputs, consequently affecting invertebrates (Minshall et al. 1995). Logging, particularly clear-cutting, is similar to fires in that an initial decrease of leaf quantity is observed (Fisher and Likens 1973). Subsequent succession can have implications for leaf litter quality and accession (Griffith and Perry 1991), which in turn can affect invertebrate communities (Benfield et al. 1991). Combining knowledge of


82






83
what species are likely to colonize a riparian system after disturbance, from pioneer species to climax communities, with the information in the present studies, will enable land managers and scientists to determine temporal availability of organic matter based on tree species present.

Restoration of degraded riparian habitats is increasing (Kondolf and Micheli

1995). In many cases, these habitats have little to no woody vegetation, or the vegetation may be undesirable. Knowledge of tree associations can increase the probability for long-term success by planting trees from each of the four groupings classified in this study. By maximizing nutrient content, levels of recalcitrant and inhibitory compounds, and timing of leaf fall, organic matter can be available for extended periods of the year and more closely resemble communities in undisturbed areas.














REFERENCES


Abelho, M., and M. A. S. Graqa. 1996. Effects of eucalyptus afforestation on leaf litter
dynamics and macroinvertebrate community structure of streams in central
Portugal. Hydrobiologia 324: 195-204.

Allan, J. D. 1982. The effects of reduction in trout density on the invertebrate community
of a mountain stream. Ecology 63:1444-1455.

Anderson, N. H., and J. R. Sedell. 1979. Detritus processing by macroinvertebrates in
stream ecosystems. Ann. Rev. Entom. 24: 351-377.

Angradi, T. R. 1996. Inter-habitat variation in benthic community structure, function, and
organic matter storage in three Appalachian headwater streams. J. N. Am. Benthol.
Soc. 15: 42-63.

Barlocher, F., and J. J. Oertli. 1978. Inhibitors of aquatic hyphomycetes in dead conifer
needles. Mycologia 70:964-974.

Benfield, E. F. 1997. Comparison of litterfall input to streams. J. N. Am. Benthol. Soc.
16: 104-108.

Benfield, E. F., D. S. Jones, and M. F. Patterson. 1977. Leaf pack processing in a
pastureland stream. Oikos 29:99-103.

Benfield, E. F., J. R. Webster, S. W. Golladay, G. T. Peters, and B. M. Stout. 1991.
Effects of forest disturbance on leaf breakdown in southern Appalachian streams.
Verh. Internat. Verein. Limnol. 24: 1687-1690.

Benke, A. C. and J. L. Meyer. 1988. Structure and function of a blackwater river in the
Southeastern U.S.A. Verh. Internat. Verein. Limnol. 23: 1209-1218.

Benstead, J. P. 1996. Macroinvertebrates and the processing of leaf litter in a tropical
stream. Biotropica 28: 367-375.

Bilby, R. E., and G. E. Likens. 1979. Effect of hydrologic fluctuations on the transport of
fine particulate organic carbon in a small stream. Limnol. Oceanogr. 24: 69-75.

Bilby, R. E. 1981. Role of organic debris dams in regulating the export of dissolved and
particulate matter from a forested watershed. Ecology 62: 1234-1243.


84






85


Bilby, R. E. 1980. Importance of organic debris dams in the structure and function of
stream ecosystems. Ecology 61: 1107-1113.

Boulton, A. J., and P. I. Boon. 1991. A review of methodology used to measure leaf litter
decomposition in lotic environments: Time to turn over an old leaf? Aus. J. Mar.
Freshwat. Res. 42: 1-43.

Bretschko, G., and M. Leichtfried. 1988. Distribution of organic matter and fauna in a
second order, alpine gravel stream (Ritrodat-Lunz study area, Austria). Verh.
Internat. Verein. Limnol. 23: 1333-1339.

Brinson, M. M., A. E. Lugo, and S. Brown. 1981. Primary productivity, decomposition
and consumer activity in freshwater wetlands. Ann. Rev. Ecol. Syst. 12: 123-161.

Bruce, K. A., G. N. Cameron, and P. A. Hacombe. 1995. Initiation of a new woodland
type on the Texas Coastal Prairie by the Chinese tallow tree (Sapium sebiferum
(L.) Roxb.). Bull. Torrey Bot. Club 122: 215-225.

Bryant, J. P., F. S. Chapin, III, and D. R. Klein. 1983. Carbon/nutrient balance of boreal
plants in relation to vertebrate herbivory. Oikos 40:357-368.

Bunn, S. E. 1988. Processing of leaf litter in two northern jarrah forest streams, Western
Australia: 2. The role of macroinvertebrates and the influence of soluble
polyphenols and inorganic sediment. Hydrobiologia 162:211-223.

Campbell, I. C. 1994. Ecology of Australian detritus-based streams - A note in response
to Graga. Limnologica 24:291-293.

Campbell, I. C., and L. Fuchshuber. 1994. Amount, composition and seasonality of
terrestrial litter accession to an Australian cool temperate rainforest stream. Arch.
Hydrobiol. 130: 499-512.

Canhoto, C., and M. A. S. Graga. 1995. Food value of introduced eucalypt leaves for a
Mediterranean stream detritivore: Tipula lateralis. Freshwat. Biol. 34:209-214.

Caswell, H. 1978. Predator-mediated coexistence: A non-equilibrium model. Amer. Nat.
112:127-154.

Chergui, H., and E. Pattee. 1990. The influence of season on the breakdown of
submerged leaves. Arch. Hydrobiol. 120:1-12.

Chew, F. S., and J. E. Rodman. 1979. Plant resources for chemical defense, Pages 271307 In G. A. Rosenthal and D. H. Janzen (eds.) Herbivores: their interaction with
secondary plant metabolites. New York, Academic Press.






86
Conners, M. E., and R. J. Naiman. 1984. Particulate allochthonous inputs: Relationships
with stream size in an undisturbed watershed. Can. J. Fish. Aquat. Sci. 41: 14731484.

Cuffhey, T. F. 1988. Input, movement and exchange of organic matter within a
subtropical coastal blackwater river-floodplain system. Freshwat. Biol. 19: 305320.

Cuffney, T. F., and B. J. Wallace. 1987. Leaf litter processing in coastal plain streams
and floodplains of southeastern Georgia, U.S.A. Arch. Hydrobiol., Suppl. 76: 124.

Cuffiey, T. F., J. B. Wallace, and G. J. Lugthart. 1990. Experimental evidence
quantifying the role of benthic invertebrates in organic matter dynamics of
headwater streams. Freshwat. Biol. 23: 281-299.

Cummins, K. W. 1974. Structure and function of stream ecosystems. Bioscience 24:631
641.

Cummins, K. W., and M. J. Klug. 1979. Feeding ecology of stream invertebrates. Annu.
Rev. Ecol. Syst. 10:147-172.

Cummins, K.W., M.J. Klug, G.M. Ward, G.L. Spingler, R.W. Speaker, R.W. Ovink, D.C.
Mahan, and R.C. Petersen. 1981. Trends in particulate organic matter fluxes,
community processes and macroinvertebrate functional froups along a Great
Lakes drainage basin river continuum. Verh. Internal. Verein. Limnol. 21: 841849.

Cummins, K. W., R. C. Petersen, F. 0. Howard, J. C. Wuycheck, and V. I. Holt. 1973.
The utilization of leaf litter by stream detritivores. Ecology 54: 336-345.

Cummins, K. W., J. R. Sedell, F. J. Swanson, G. W. Minshall, S. G. Fisher, C. E.
Cushing, R. C. Petersen, and R. L. Vannote. 1983. Organic matter budgets for
stream ecosystems: problems in their evaluation. Pages 299-354, In J. R. Barnes and G. W. Minshall (eds). Stream Ecology: Application and Testing of General
Ecological Theory. Plenum, New York.

Cummins, K. W., M. A. Wilzbach, D. M. Gates, J. B. Perry, and W. B. Taliaferro. 1989.
Shredders and riparian vegetation. Bioscience 39: 24-30.

Day, F. P., Jr. 1982. Litter decomposition rates in the seasonally flooded Great Dismal
Swamp. Ecology 63: 670-678.

Dudgeon, D. 1982. An investigation of physical and biological processing of two species
of leaf litter in Tai Po Kau Forest Stream, New Territories, Hong Kong. Arch.
Hydrobiol. 96:1-32.






87


Dudgeon, D. 1989. The influence of riparian vegetation on the functional organization of
four Hong Kong stream communities. Hydrobiologia 179: 183-194.

Dudgeon, D. 1994. The influence of riparian vegetation on macroinvertebrate community
structure and functional organization in six New Guinea streams. Hydrobiologia
294:65-85.

Edwards, C. A., and G. W. Heath. 1975. Studies in leaf litter breakdown. III. The
influence of leaf age. Pedobiologia 15:348-354.

Elder, J. F., and D. J. Cairns. 1982. Production and decomposition offorest litter fall on
the Apalachicola River flood plain, Florida. Alexandria, VA, U.S. Geological
Survey.

Elwood, J. W., J. D. Newbold, A. F. Trimble, and R. W. Stark. 1981. The limiting role of
phosphorus in a woodland stream ecosystem: Effects of P enrichment on leaf
decomposition and primary producers. Ecology 62: 146-158.

Epler, J.H. 1995. Identification Manual for the Larval Chironomidae (Diptera) of
Florida. Florida Department of Environmental Protection, Tallahassee, FL.

Fisher, S. G., and G. E. Likens. 1973. Energy flow in Bear Brook, New Hampshire: an
integrative approach to stream ecosystem metabolism. Ecol. Monogr. 43:421-439.

Fisher, S. G., L. J. Gray, N. B. Grimm, and D. E. Busch. 1982. Temporal succession in a
desert stream ecosystem following flash flooding. Ecological Monographs 52: 93110.

Gallaher, R. N., C. 0. Weldon, and J. G. Futral. 1975. An aluminum block digester for
plant and soil analysis. Soil Sci. Soc. Amer. Proc. 39:803-806.

Gallardo, A., and J. Merino. 1999. Control of leaf litter decomposition rate in a
Mediterranean shrubland as indicated by N, P and lignin concentrations.
Pedobiologia 43: 64-72.

Gessner, M. W., and J. Schwoerbel. 1989. Leaching kinetics of fresh leaf-litter with
implications for the current concept of leaf-processing in streams. Arch.
Hydrobiol. 115: 81-90.

Golladay, S. W., J. R. Webster, and E. F. Benfield. 1989. Changes in stream benthic
organic matter following watershed disturbance. Holarc. Ecol. 12: 96-105.

Gonzalez, E., and J. Pozo. 1996. Longitudinal and temporal patterns of benthic coarse
particulate organic matter in the Aguera stream (northern Spain). Aquat. Sci. 58:
355-366.






88


Greenway, M. 1994. Litter accession and accumulation in a Melaleuca quinquenervia
(Cav.) S.T. Blake wetland in south-eastern Queensland. Aust J. Mar Freshwat.
Res. 45:1509-1519.

Griffith, M. B., and S. A. Perry. 1991. Leaf pack processing in two Appalachian
mountain streams draining catchments with different management histories.
Hydrobiologia 220: 247-254.

Griffith, M. B., and S. A. Perry. 1993. Colonization and processing of leaf litter by
macroinvertebrate shredders in streams of contrasting pH. Freshwat. Biol. 30:93103.

Gurtz, M. E., G. R. Marzolf, K. T. Killingbeck, D. L. Smith, and J. V. McArthur. 1988.
Hydrologic and riparian influences on the import and storage of coarse particulate
organic matter in a prairie stream. Can. J. Fish. Aquat. Sci. 45: 655-665.

Haefner, J. D., and J. B. Wallace. 1981. Shifts in aquatic insect populations in a firstorder southern Appalachian stream following a decade of old field succession.
Can. J. Fish. Aquat. Sci. 38: 353-359.

Hambleton, L. G. 1977. Semiautomated method for simultaneous determination of
phosphorus, calcium and crude protein in animal feeds. J. Assoc. Off Anal.
Chem. 60:845-852.

Haslam, E. 1989. Plant Polyphenols. Cambridge, Cambridge University Press.

Hauer, F. R., N. L. Poff, and P. L. Firth. 1986. Leaf litter decomposition across broad
thermal gradients in southeastern coastal plain streams and swamps. J Freshwat.
Ecol. 3:545-552.

Hill, A. R., and A. M. Brooks. 1996. Coarse particulate organic matter inputs to a headwater swamp stream. Arch. Hydrobiol. 137: 25-38.

Howarth, R. W., and S. G. Fisher. 1976. Carbon, nitrogen, and phosphorus dynamics
during leaf decay in nutrient enriched stream microecosystems. Freshwat. Biol.
6:221-228.

Hynes, H. B. N. 1975. The stream and its valley. Verh. Internat. Verein. Limnol. 19: 1-15.

Irons, J. G., and M. W. Oswood. 1997. Organic matter dynamics in 3 subarctic streams
of interior Alaska, USA. J. N. Am. Benthol. Soc. 16: 23-28.

Jones, J. B. 1997. Benthic organic matter storage in streams: influene of detrital import
and export, retention mechanisms, and climate. J. N. Am. Benthol. Soc. 16: 109119.






89


Jones, J. B., and L. A. Smock. 1991. Transport and retention of particulate organic
matter in two low-gradient headwater streams. J. N. Am. Benthol. Soc. 47: 588594.

Kaushik, N. K., and H. B. N. Hynes. 1971. The fate of dead leaves that fall into streams.
Arch. Hydrobiol. 68:465-515.

King, J. M., J. A. Day, B. R. Davies, and M. P. Henshall-Howard. 1987. Particulate
organic matter in a mountain stream in the south-western cape, South Africa.
Hydrobiologia 154: 165-187.

Kirk, T. K., W. J. Connors, and J. G. Zeikus. 1977. Advances in understanding the
microbiological degradation of lignin In F. A. Loewus, and V. C. Runeckles, eds.
Recent Advances in Phytochemistry. Vol. 11. The Structure, Biosysnthesis and
Degradation of Wood. New York, Plenum.

Kondolf, G. M., and E. R. Micheli. 1995. Evaluating stream restoration projects. Environ.
Manage. 19: 1-15.

Lake, P. S., and L. A. Barmuta. 1986. Stream benthic communities: Persistent
presumptions and current speculations. Monogr. Biol. 61: 263-276.

Lester, P. J.,, S. F. Mitchell, and D. Scott. 1994. Effects of riparian willow trees (Salix
fragilis) on macroinvertebrate densities in two small central Otago, New Zealand,
streams. N. Z J. Mar. Freshwat. Res. 28: 267-276.

Liu, S., H. Riekerk, and H. L. Gholz. 1996. Leaf litterfall, leaf area index, and radiation
transmittance in cypress wetlands and slash pine plantations in north-central
Florida. Wed. Ecol. Manage. 4: 257-271.

Lockhart, C. S., D. F. Austin, W. E. Jones, and L. M. Downey. 1997. The invasion of
carrotwood (Cupaniopsis anacardioides) in natural areas. Final study report.
Pages 24. Habitat Specialists, INC. for Florida Department of Environmental
Protection, Tallahassee, FL.

Lugo, A. E., and C. P. Zucca. 1983. Comparison of litter fall and turnover in two Florida
ecosystems. Fla. Sci. 46: 101-111.

Lugthart, G. J., and J. B. Wallace. 1992. Effects of disturbance on benthic functional
structure and production in mountain streams. J. N. Am. Benthol. Soc. 11: 138164.

Lyons, J., S. W. Trimble, and L. K. Paine. 2000. Grass versus trees: Managing riparian
areas to benefit streams of central North America. J. Am. Water Res. Assoc. 36:
919-930.







90


Maity, S. K., and V. C. Joy. 1999. Impact of antinutritional chemical compounds of leaf
litter on detritivore soil arthropod fauna. J. Ecobiol. 11: 193-202.

McArthur, J. V., J. M. Aho, R. B. Rader, and G. L. Mills. 1994. Interspecific leaf
interactions during decomposition in aquatic and floodplain ecosystems. J. N. Am.
Benthol. Soc. 13:57-67.

McClaugherty, C. A., J. Pastor, J. D. Aber, and J. M. Melillo. 1985. Forest litter '
decomposition in relation to soil nitrogen dynamics and litter quality. Ecology 66:
266-275.

McDowell, W.H., and C.E. Asbury. 1994. Export of carbon, nitrogen, and major ions
from three tropical montane watersheds. Limnol. Oceanogr. 39: 111-125.

Melillo, J. M., J. D. Aber, and J. F. Muratore. 1982. Nitrogen and lignin control of
hardwood leaf litter decomposition dynamics. Ecology 63:621-626.

Merritt, R. W., and D. L. Lawson. 1992. The role of leaf litter macroinvertebrates in
stream-floodplain dynamics. Hydrobiologia 248: 65-77.

Merritt, R. W., W. Wuerthele, and D. L. Lawson. 1984. The effect of leaf conditioning on
the timing of litter processing on a Michigan woodland floodplain. Can. J. Zool.
62:179-182.

Metzler, G. M., and L. A. Smock. 1990. Storage and dynamics of subsurface detritus in a
sand-bottomed stream. Can. J. Fish. Aquat. Sci. 47: 588-594.

Minoletti, M. L., and R. E. J. Boerner. 1994. Drought and site fertility effects on foliar
nitrogen and phosphorus dynamics and nutrient resorption by the forest understory shrub Viburnum acerifolium L. Am. Mid. Nat. 131:109-119.

Minshall, G. W. 1996. Organic matter budgets. Pages 591-605 in F. R. Hauer and R. H.
Lamberti, eds. Methods in Stream Ecology. Academic Press, San Diego, CA.

Minshall, G. W., R. C. Petersen, K. W. Cummins, T. L. Bott, J. R. Sedell, C. E. Cushing,
and R.L. Vannote. 1983. Interbiome comparison of stream ecosystem dynamics.
Ecol. Monogr.53: 1-25.

Minshall, G. W., K. W. Cummins, R. C. Petersen, C. E. Cushing, D. A. Bruns, J. R.
Sedell, and R. L. Vannote. 1985. Developments in stream ecosystem theory. Can.
J Fish. Aquat. Sci. 42: 1045-1055.

Minshall, G. W., J. T. Brock, and J. D. Varley. 1989. Wildfires and Yellowstone's stream
ecosystems. Bioscience 39: 707-715.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EI7LH76WU_RD3581 INGEST_TIME 2015-04-13T20:37:20Z PACKAGE AA00029975_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

RIPARIAN TREE ASSOCIATIONS AND STORAGE, TRANSPORT, AND PROCESSING OF PARTICULATE ORGANIC MATTER IN A SUBTROPICAL STREAM ' By ; ; CHRISTOPHER RICHARD ROBERTS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE L'NIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIIUiMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2002

PAGE 2

ACKNOWLEDGMENTS I greatly appreciate all the efforts of my major professor, Dr. Thomas Crisman. His advice and guidance, both dealing with this project and in more personal areas, through this arduous process was considerable. I am also grateful to my other committee members, Drs. Lauren Chapman, Frank Nordlie, and William Wise for their counsel. My wife was inspirational through this entire process, never letting me get down on my setbacks and always pushing me to work my hardest and best. She has worked equally hard in her studies and although our time together has been limited, her constant inspiration has led me to the point where I am now. Special thanks go to Todd Osborne for helping with processing of leaf materials in the laboratory. All chemical analyses were performed by the Forage Evaluation Support Laboratory at the University of Florida, with specific thanks to Richard Fethiere. Without all their help, the completion of this project and thesis would never have been accomplished. Thanks also go to all the student workers who slaved away in anonymity to allow others to reap the benefits. I would also like to thank the Center for Wetlands staff, specifically Sherl Brinkley, for their behind the scene efforts in making this research a reality. ii

PAGE 3

TABLE OF CONTENTS page ACKNOWLEDGMENTS ii LIST OF TABLES v LIST OF FIGURES vi ABSTRACT ix CHAPTERS 1 INTRODUCTION 1 2 LEAF ASSOCIATIONS 4 Introduction •. 4 Methods 5 Results 7 Discussion 12 3 STORAGE AND TRANSPORT 21 Introduction 21 Methods 22 Study Site 22 Experimental Approach 23 Data Analysis 27 Results and Discussion 27 Latitudinal Comparisons 27 Organic Matter Budget 31 Inputs 34 Storage 41 Transport 50 4 LITTER PROCESSING 56 Introduction 56 Methods 57 iii

PAGE 4

Study Site •'^ Data Analysis Results ^1 Discussion Leaf Chemistry ^3 Timing of Litter Fall 77 Macroinvertebrates 5 CONCLUSIONS 82 REFERENCES 84 APPENDICES 94 BIOGRAPHICAL SKETCH 97 iv

PAGE 5

I LIST OF FIGURES Figure Page 21 Dendrogram showing associations of 3 1 common riparian tree species from north-central Florida based on phosphorus, nitrogen, recalcitrant and inhibitory compounds, cuticle development, and timingof leaf fall. Numbers on left indicate four discrete groupings 10 2-2 Period of leaf fall of 3 1 common riparian tree species from north-central Florida from September 2000 to August 200 1 11 2-3 Percentage of dry weight (± 1 standard error) consisting of phosphorus for 3 1 common riparian tree species from north-central Florida 13 2-4 Percentage of dry weight (± 1 standard error) consisting of nitrogen for 3 1 common riparian tree species from north-central Florida 14 25 In vitro organic matter digestibility (IVOMD) (± 1 standard error) as a percentage of dry weight for 3 1 common riparian tree species from north-central Florida 15 31 Map of Possum Creek, Florida, study site and placement of sampling equipment in floodplain 24 3-2 Approximate timing of leaf abscission from selected forests of the world. Solid line indicates timing of leaf fall 28 3-3 Organic matter budget for the Possum Creek system, Florida, from October 2000-November 2001 . Solid arrows indicate measured movements, and dashed arrows indicate hypothesized movements of organic matter between input, storage, and export compartments. Inputs (left column of boxes) and exports (right column) are expressed as annual summation of each transport vehicle. Storage (middle column) is expressed as the annual mean. All data are expressed as g AFDM m-^yr' 33 3-4 Organic matter inputs via litterfall (circles) and lateral deposition (triangles) from October 2000 through November 2001 into Possum Creek, Florida. Error bars represent ± one standard error 35 V

PAGE 6

3-5 Percent by weight of constituents of organic matter in litterfall from October 2000 through November 2001 at Possum Creek, Florida. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction 37 3-6 Percent by weight of constituents of organic matter in lateral deposition from October 2000 through November 2001 at Possum Creek, Florida. Gap in data represents a flooding event that disabled lateral deposition traps. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction 39 3-7 Percent by weight of constituents of organic matter stored on the Possum Creek floodplain, Florida, from October 2000 through November 2001. Reproductive structiires include nuts, seeds, blossoms, and other structures associated with floral reproduction 42 3-8 Organic matter storage on the Possum Creek floodplain, Florida, from October 200 through November 2001 . Error bars represent ± one standard error 43 3-9 In-stream organic matter storage in leaf packs (white) and debris dams (black) per 100 m of stream length from October 2000 through November 2001 in Possum Creek, Florida 45 3-10 Percent by weight of constituents of in leaf packs (upper) and debris dams (lower) from October 2000 through November 2001 in Possum Creek, Florida. Gaps in data represent periods where no leaf packs were present. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction 46 3-11 In-stream benthic organic matter storage as CPOM (circles) and FPOM (triangles) from October 2000 through November 2001 in Possum Creek, Florida. Error bars represent ± one standard error 49 3-12 In-stream organic matter transport as FPOM (white) and CPOM (black) from October 2000 through November 2001 in Possum Creek, Florida 51 3-13 Gage height at Possum Creek, Florida, from September 2000 to October 2001 . Asterisks (*) represent instances when water height exceeded bankfull stage 52 4-1 Percent of dry weight of phosphorus, nitrogen, and refractory compounds in the senescent leaves of five tree species collected from October 2000-May 2001 . Error bars represent ± one vi

PAGE 7

Macroinvertebrate abundance and leaf carbon content over time collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack'' and white bars represent macroinvertebrates-g C'-leaf pack''. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error Dominant shredders and collector-gatherers collected from Possum Creek from December 2000 to May 2001 . Gray bars represent macroinvertebrates-leaf pack'' and white bars represent macroinvertebrates-g C'-leaf pack''. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error Dominant filter-feeders collected from Possum Creek from December 2000 to May 2001 . Gray bars represent macroinvertebrates-leaf pack'' and white bars represent macroinvertebrates-g C -leaf pack''. Both sampling date and cumulative degree days are shown Error bars represent ± one standard error Dominant predators collected from Possum Creek from December 2000 to May 2001 . Gray bars represent macroinvertebrates-leaf pack'' and white bars represent macroinvertebrates-g C'-leaf pack''. Both sampling date and cumulative degree days are shown Error bars represent ± one standard error Chironomidae (Diptera) collected from Possum Creek from December 2000 to May 2001 . Gray bars are macroinvertebratesleaf pack'' and white bars are macroinvertebrates-g C' -leaf pack''. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error Chironomidae (Diptera) relative abundance over time collected from Possum Creek from December 2000 to May 2001 . Both sampling date and cumulative degree days are shown vii

PAGE 8

LIST OF TABLES Table Page 21 Scientific and common names of the 3 1 common riparian species from north central Florida used in this study and the degree of cuticle development. Cuticle development is on a scale of 1-5, with 1 representing the most heavily developed cuticle and 5 the least developed 8 31 Litterfall from representative forests along a latitudinal gradient 29 3-2 Lateral deposition from representative forests along a latitudinal Gradient 32 3-3 Organic matter storage in representative first and second-order streams along a latitudinal gradient 40 34 Organic matter export from representative first and second-order sfreams along a latitudinal gradient 53 41 Macroinvertebrates collected from Possum Creek, Florida, on leaf packs from December 2000 to May 2001 65 viii

PAGE 9

Abstract of Dissertation Presented to the Graduate School in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RIPARIAN TREE ASSOCIATIONS AND STORAGE, TRANSPORT, AND PROCESSING OF PARTICULATE ORGANIC MATTER IN A SUBTROPICAL STREAM By Christopher Richard Roberts May 2002 Chairperson: Thomas L. Crisman Major Department: Environmental Engineering Sciences Riparian tree associations, leaf-litter storage, transport, and in-stream processing were investigated in subtropical north-central Florida. Chemical constituents, cuticle development, and timing of leaf fall were quantified for 31 species and analyzed with hierarchical cluster analysis. Four groups and two outlying species were classified. Pinus elliottii, uncommon in natural riparian systems, but widely planted regionally in plantations, and Sapium sebiferum, an exotic, were statistically associated with other species, but monocultures could deleteriously affect invertebrates that process leaf litter. These associations can be used to better understand vegetative community management and restoration of riparian areas. Storage and transport of leaf litter were analyzed biweekly over a year in both a stream and floodplain system. Peak leaf fall occurred from September-December with a smaller peak during January-F ebruary and were reflected in peaks of both lateral and longitudinal organic matter transport. Longitudinal transport was also correlated to stream discharge. Floodplain storage was highest during winter 2000-2001 during and after the period of peak leaf fall. Variability in floodplain organic matter increased ix

PAGE 10

markedly after flooding during July and September 2001 . These results suggest that leaf fall is temporally extended and that fresh litter is available for much of the year. Leaves of five common riparian tree species were compared for in-stream processing rates. Early-abscising deciduous trees were processed twice as rapidly as late-abscising evergreen trees. In contrast to most temperate streams, shredding macroinvertebrates were rarely collected and appear to be of limited importance in leaf processing. This suggests that fungal and bacterial processing is of greater importance in subtropical streams. Macroinvertebrate abundance on leaf packs decreased from December through May, rather than over the course of processing, suggesting that factors other than leaf availability influence macroinvertebrate abundance. The fact that few shredders were present suggests that invertebrates were using leaf packs as habitat and, in the case of scrapers and collector-gatherers, for feeding fungal commimities and zone of fine particulate organic matter accumulation, respectively. This research fills a void in imderstanding stream ecosystem function along a latitudinal gradient. Subtropical streams, in terms of both timing of leaf fall and a macroinvertebrate community only loosely synchronized to seasonal leaf inputs, behave in a way intermediate between conditions in temperate streams and those of the Southern Hemisphere.

PAGE 11

CHAPTER 1 INTRODUCTION Small streams in forested watersheds typically rely on materials derived from outside the stream channel as their primary energy base (e.g., Cummins et al. 1973; Fisher and Likens 1973; Hynes 1975). AUochthonous materials that enter a stream are primarily in the form of coarse particulate organic matter (CPOM), which is predominantly leaves and wood, but also reproductive structures such as blossoms, fruits, and nuts, and dead animals. Once in the stream, the most CPOM is trapped by in stream structure such as fallen tree limbs, roots, and rocks. Here, it is processed by a suite of organisms, ranging from bacteria and fungi to macroinvertebrates and broken down into smaller particles (fine particulate organic matter (FPOM) and dissolved organic matter (DOM)) that are exported downstream and utilized by other organisms. This process has been synthesized into the "River Continuum Concept" (Vannote et al. 1980), which describes the structure and function of streams from their headwaters to where they enter a larger stream or lentic body of water. These processes are well known in Northen Hemisphere temperate systems (e.g., Cummins et al. 1973; Fisher and Likens 1973; Hynes 1975), but poorly studied in the subtropics. The temperate zone and subtropics differ in many frmdamental respects, including climate, pliutopci iod, aiid liming of litter accession (Lugo and Zacca 1983; Watt and Golladay 1999). Applying concepts based on temperate systems to those in other regions is potentially inappropriate. For example, lowland tropical systems 1

PAGE 12

2 Operate differently than what is outlined in the "River Continuum Concept" (V annote et al. 1980), but rather is more properly expressed by Junk (1984) in the "Flood-Pulse Concept". Furthermore, Southern Hemisphere temperate systems operate differently than those of the Northern Hemisphere (King et al. 1987, Lake and Barmuta 1986), with leaf fall more closely associated with precipitation (dry season) than with temperature and photoperiod as is the case in the Northern Hemisphere. Because systems outside the Northern Hemisphere temperate zone operate differently than what has been accepted by scientists in those regions, it stands to reason that Northern Hemisphere subtropics may differ from temperate systems as well. Subtropical systems, such as those in north-central Florida, experience an extended period of leaf fall as compared to the brief autumnal period in the Northern Hemisphere temperate zone (Lugo and Zacca 1983; Watt and Golladay 1999). Not only will availability of leaves be governed by the extended subtropical leaf fall as compared to the Northern Hemisphere temperate zone, but also leaf quality will differ among trees. Initial leaf quality is often the most important factor determining processing rates of leaves (Kaushik and Hynes 1971; Merritt et al. 1984). Different tree species have leaves that contain different concentrations of chemical and physical deterrents as well as nutrients, which influence their processing rates in streams. Chemical defenses are of greatest importance in many species, but thick, waxy cuticles may also retard processing in others (Bunn 1988; Canhoto and Gra9a 1995). This research addresses these issues in three parts. First, riparian trees taxa from north-central Florida were segregated into discrete groupings based on the commonalities of when taxa are likely to become available to be processed, using

PAGE 13

3 phosphorus, nitrogen, and recalcitrant compound content of leaves, as well as cuticle development and timing of leaf fall. Second, temporal patterns of organic matter inputs, storage, and transport were quantified in a subtropical stream. Possum Creek. Finally, rates at which leaves of five common riparian trees on the Possum Creek floodplain are processed was determined, and the temporal patterns of abundance for stream macroinvertebrates was examined in relation to sequential leaf organic matter additions. Addressing these issues allows us to better understand the importance of organic matter in the subtropics and how processes differ from the Northern Hemisphere temperate zone.

PAGE 14

CHAPTER 2 LEAF ASSOCIATIONS Introduction Leaf fall is often the dominant carbon source for forested streams (Hynes 1975), forested wetlands, (Brinson et al. 1981) and upland-forest floors (Maity and Joy 1999). Utilization of this energy is governed by a number of factors, including moisture (Singh 1984); initial leaf nutrient content (Petersen and Cummins 1974); presence of inhibitory and recalcitrant compounds (Haslam 1989); biota adapted to processing leaves (Merritt and Lawson 1992); and in aquatic systems, water chemistry (Griffith and Perry 1993, Howarth and Fisher 1976). When leaves are introduced to the system over a short period of time, as is the case in the Northern Hemisphere temperate zone, the above factors can typically be of greatest importance in leaf processing. However, in the subtropics (Elder and Cairns 1982, Liu et al. 1996, Lugo and Zucca 1983), the tropics (Songwe et al. 1988, Spain 1984), and Southern Hemisphere temperate zone (King et al. 1987, Lake and Barmuca 1986), leaf fall extends over a longer period of time, and leaves of different species are added to the system sequentially. Differences in nutritional quality and inhibitory and recalcitrant compounds, along with timing of leaf fall, are plant-specific factors that influence temporal utilization of these energy sources. By imderstanding leaf quality and timing of abscission, one can better predict the relative importance of particular species to those 4

PAGE 15

5 organisms directly feeding on leaf litter and those grazing on microbes involved in leaf processing. Environmental changes and disturbance can affect the plant community of an ecosystem and, consequently, those organisms depending on energy derived from leaves. For example, afforestation by exotic species such as Eucalyptus can alter in stream macroinvertebrate communities (Abelho and Graga 1996) because of much higher levels of inhibitory and recalcitrant compounds and differences in timing of leaf fall. The influence of nutritional content of leaves of tree species along a temporal successional gradient can also influence stream macroinvertebrates (Haefner and Wallace 1981, Lyons et al. 2000) and terrestrial communities (McClaugherty et al. 1985). This study addresses these issues by measuring the amount of phosphorus, nitrogen, and recalcitrant compounds in leaves, as well as cuticle development, and timing of leaf fall for common tree species in north-central Florida riparian forests. Based on these data, relations between these trees will be established that can be used to identify when taxa are likely to become available for processing. The relations between two invasive species and native riparian trees will also be addressed by discerning how they cluster. Methods Leaves were collected from 3 1 common tree species growing in riparian areas in multiple locations in north central Florida, including the Hogtown Creek, Blues Creek, Santa Fe River, and Oklawaha River watersheds (approximately 29° 50' N, 82° 20' W). Freshly abscised leaves were collected during the period of leaf fall for each species and placed into resealable plastic bags. Leaves were collected from multiple trees and

PAGE 16

6 watersheds, wherever possible. Immediately upon return to the laboratory, leaves were air-dried for 48 hours and samples from separate watersheds were homogenized. After at least 100 g (dry mass) of leaf material for each species were obtained and milled for 10 minutes in a large Wiley Mill (Standard no. 3) through 3 mm mesh. Samples were then ground in a small Wiley Mill (A75-A) through 1 mm mesh. Ten g of ground samples were then placed into 6 oz. Whirl-pak bags. Enough material was collected to create a minimum of 4 samples per species. Leaves of each species were analyzed for phosphorus, nitrogen, and in-vitro organic matter digestibility (IVOMD). For nitrogen and phosphorus analyses, samples were digested using a modification of the aluminum block digestion procedure (Gallaher et al. 1975). Sample mass was 0.25 g, the catalyst used was 1.5 g of 9:lK2S04:CuS04, and digestion was conducted for at least 4 hours at 375° C using 6 mL of H2SO4 and 2 mL H2O2. Nitrogen and phosphorus in the digestate were determined by semiautomated colorimetry (Hambleton 1977) (Appendix 1). The IVOMD was performed by a modification of the two-stage technique (Moore and Mott 1974) and measured those leaf constituents that are easily digestible by ruminant gut microflora (Appendix 2). The remaining recalcitrant material is of limited use to microflora and presumably forest and aquatic microbes. Cuticle development was categorized qualitatively on a scale of 1-5, with 1 characterizing leaves with the most well developed cuticles, and 5 representing those with the least-developed ciitirles. Categorization was based on texttire, pliability, and knife scrapings. At least twice weekly from September 2000 through June 2001, surveys of each of the four watersheds were made to determine which riparian tree species were

PAGE 17

7 currently abscising. The first incidence of pronounced leaf fall for a particular species was noted as the beginning of abscission for that species in north-central Flonda. Termination of sustained leaf fall for a species anywhere in north-central Florida was designated as the end of abscission for that species. The mean date of leaf fall was also determined. Dates were transformed using a modified Julian date system whereby the beginning of the year was designated as September 1 . Data analysis was performed with a hierarchical cluster analysis using SPSS v. 9.0 (SPSS 1998). Cluster analysis was selected to segregate taxa into discrete groupings based on phosphorus, nitrogen, and recalcitrant compound content, as well as cuticle development and timing of leaf fall. These groupings should segregate based on the commonalities of when taxa are likely to become available to be processed. All data were standardized using Z-scores and calculated using the squared Euclidean distance method. Linkages were determined with between-groups linkages. Results Leaves of 31 common riparian trees were collected from October 2000May 2001 (Table 2-1). All taxa with the exception of Cephalanthus occidentalis and Chionanthus virginicus were canopy species. Almost half the taxa were near the southern-most limit of their distributions, including Platanus occidentalis, which has been introduced as an ornamental from its native range under 200 km north and west of north-central Florida. No taxa were endemic to Florida, with all ranging north into the temperate zone. Besides P. occidentalis, a non-invasive introduction, the exotic Sapium sebiferum was also found in north-central Florida. All other taxa were native.

PAGE 18

8 Table 2-1. Scientific and common names of the 31 common riparian species from north central Florida used in this study and the degree of cuticle development. Cuticle development is on a scale of 1-5, with 1 representing the most heavily-developed cuticle and 5 the least developed ^riPTitifir name Common name Cuticule development Acer negundo DUX eiucr 5 jiLur ruurufii lllCiL./lV' 4 Ho ft t //Tf W I €Tffl river hirch 4 K^urpifius curuiiniunu 4 watpr hipknrv 3 niomit hirWnrv 3 HnpVHprrv 5 hiittnnhiisVi 3 fVi n fTPffPf* 3 rruxinui curuiinturiu r^arnlina fi^h X^CUVJiiilCl ClOAA 3 r ruXlnu^ psnnsyivunicu ^ICCil aoil 3 LiquiuumOur SiyruClJlUu i>WCCl ^Ulli 3 i^iriucifCriCirOfi luiijJijcru tiilin nnnlar 3 A A/it^vi r\ #1/1 rrw/~i n/V J Tl/~iW*j~l iviu^rioiiu ^runuij luru llla^iiuiici 1 lyyssu ogtcne W^CLIICC lllilC 2 lyyssci syivuiiLU vor. oijiuru cw/Qtirn tiinplr* oVVcUllU lupcivj 2 Ostrya virginica /Miicncan nopnomuediu I inus eiiioiiii alaMl piIlC • 1 Planera uquatica wdicr Clin 2 riaicxyms occiueniciiis syLoiTiorc 2 Populus deltoidcs couonwoou 2 1 ntiyiTrili/i L-Uj lidtif IJUllL* lanrel oflW 2 Quercus lyrata overcup oak 2 Quercus michauxii swamp chestnut oak 2 Quercus nigra water oak 2 Salix caroliniana Carolina willow 4 Sapium sehiferum Chinese tallowwood 4 Tilia americana basswood 5 Taxodium distichum bald cypress 3 Ulmus alata winged elm 3 Ulmus americana American elm 3

PAGE 19

9 Four discrete groupings of trees and two outlier taxa were delineated with cluster analysis (Figure 2-1). Group 1 consists of 10 species that abscise early, have light to moderate cuticle development, and generally are nutrient-rich and low in recalcitrant compounds. Group 2 is 1 1 species that, in general, had greater cuticle development, abscission that ranged from early to middle of the period of total leaf fall, and were variable in both nutrients and recalcitrant compounds. Group 3 was composed of five species that were generally low in nutrients and high in recalcitrant compounds, had welldeveloped cuticles, and abscised over a wide time range. Group 4 consisted of three species that did not appear to fit any pattern, as degree of cuticle development, nutrients, and recalcitrant compounds were variable. Two species were relatively unrelated to the other species (Tilia americana and Magnolia grandijlora) based on the variables included in the analysis (Figure 2-1). Tilia americana was consistently high in nutrients and low in recalcitrant compounds and cuticle development. Magnolia grandijlora abscised at a time considerably different from that of any other species, was low in nutrients and recalcitrant compounds, and had a very well developed cuticle. Leaf fall occurred in two phases, from early October through late February and one species accounted for leaf fall from mid-March through early June (Figure 1-2). All taxa except M. grandijlora, Quercus nigra, and Q. laurijolia abscised between early October and the end of December. Quercus nigra dropped its leaves from early January through early Februar>' and partially overiapped Q. laurijolia, which began abscission in mid-January and ended in late February. Magnolia grandijlora abscised from mid-March through early June. Ulmus alata and Carya aquatica had the shortest periods of leaf fall,

PAGE 20

10 ^ uRescaled Distance Cluster Combine 5 10 15 20 25 Carpinus caroliniana — Salix caroliniana Planera aquatica — ' Fraxinus caroliniana — Carya glabra — Ostrya virginica Acernegundo — ' Acer rubrum Sapium sebiferum Celtis laevigata Populus deltoides lllmus americana Chionanthus virginicus Nyssa sylvatica var. biflora Liriodendron tulipifera Liquidambar styraciflua — Ulmus alata J Fraxinus pennsylvanica — ' Quercus lyrata Quercus michauxii Taxodium distichum — ' Quercus laurifolia — r Quercus nigra — ' Nyssa ogeche Platanus occidentalis Pinus elliottii Betula nigra — r Carya aquatica — ' Cephalanthus occidentalis — Tilia americana — Magnolia grandijlora Figvire 2-1 . Dendrogram showing associations of 3 1 common riparian tree species from north-central Florida based on phosphorus, nitrogen, recalcitrant and inhibitory compounds, cuticle development, and timing of leaf fall. Numbers on left indicate four discrete groupings.

PAGE 21

11 Platanus occidentalis Cephalanthus occidentalis Chionanthiis virginicus Fraxinus caroliniana Fraxinus pennsylvanica Nyssa ogeche Nyssa sylvatica var. biflora Tilia amehcana Celtis laevigata Betula nigra Liriodendron tulipifera Populus deltoides Acer negundo Pinus elliottii Planera aquatica Liquidambar styraciflua Sapium sebiferum Salix caroliniana Ulmus alata Carya glabra Ulmus americana Carya aquatica Taxodium distichum Acer rubrum Carpinus caroliniana Ostrya virginica Quercus lyrata Quercus michauxii Quercus nigra Quercus laurifolia Magnolia grandiflora O N D F M Month M J Figure 2-2. Period of leaf fall of 31 common riparian tree species from northcentral Florida from September 2000-August 2001.

PAGE 22

12 lasting less than a month. Conversely, P. occidentalis (early October through early December) and M grandiflora abscised over the longest time period. Mean phosphorus content of leaves was 0.16% of leaf dry mass and ranged from 0.02% {Pinus elliottii) to 0.50% {T. americand) (Figure 2-3). Nitrogen content was higher than P for all species. Mean N was 0.94% of leaf dry mass and ranged from 0.44% {P. elliottii) to 1.70% {Carya aquatica) (Figure 2-4). The IVOMD was below 50% for all species except T. americana, M. grandiflora, and S. sebiferum (Figure 2-5). Mean IVOMD was 31.9% and ranged from 1 1.4% {P. occidentalis) to 66.0% (T. americana). Thus, with recalcitrant compounds being 100% IVOMD, P. occidentalis had the highest levels of recalcitrant compounds and T. americana had the lowest levels. Most leaves had intermediate cuticle development, with M. grandiflora, P. elliottii, and Nyssa ogeche having the best-developed cuticles and Acer negundo, Celtis laevigata, Ostrya virginica, and T. americana having the least-developed cuticles (Table 2-1). Discussion All species in this study are ubiquitous flora of riparian areas across north-central Florida. Not every species is found in every habitat, as some (e.g., Taxodium distichum, N. ogeche, and Betula nigra) are obligate wetland plants, while others (e.g., C. virginicus, O. virginica, and M. grandiflora) are upland trees that frequently grow on bluffs or other raised portions of riparian areas (Tobe et al. 1998). Unlike the Northern Hemisphere temperate zone where leaf fall is limited to a brief autumnal period (Fisher and Likens 1973), in north-central Florida it occurs over an extended period. Both tropical and Southern Hemisphere temperate systems experience leaf fall throughout the year, but as opposed to north-central Florida and the Northern

PAGE 23

13 Tilia americana Liquidambar styracillua Ulmus alata Quercus lyrata Fraxinus pennsylvanica Betula nigra Quercus michauxii Taxodium distich um Car/a aquatica Nyssa sylvatica var. biflora ChionanOius virginicus Ulmus americana Celtis laevigata Populus deltoides Carya glabra Sapium sebiferum Ostrya virginica Carpinus caroliniana Piatanus occidentalis Acer rub rum Acer negundo Magnolia grandiflora Nyssa ogeche Cephalanthus occidentalis Quercus nigra Salix caroliniana Liriodendron tuiipifera Pianera aquatica Quercus laurifolia Fraxinus caroliniana Pinus elliottii 0.1 0.2 0.3 0.4 Percent of dry weight 0.5 Figure 2-3. Percentage of dry mass (± 1 standard error) consisting of phosphorus for 3 1 common riparian tree species from north-central Florida.

PAGE 24

14 Carya aquaUca Cephalanthus occidentalis Betula nigra Tilia americana Planera aquatica Salix caroliniana Carya glabra Quercus lyrata Carpinus caroliniana Platanus occidentalis Quercus michauxii Taxodium disdchum CeWs laevigata Ostrya virginica Nyssa ogeche Populus deltoides Fraxinus caroliniana Ulmus alata Acer negundo Quercus laurifolia Ulmus americana Magnolia grandillora Liriodendron tulipifera Fraxinus pennsylvanica Quercus nigra Sapium sebiferum Acer rub rum Nyssa sylvatica var biHora Chionanthus virginicus Liquidambar styraciflua Pinus elliottii 0.6 0.8 1 1.2 Percent of dry weight 1.4 1.6 Figure 2-4. Percentage of dry mass (± 1 standard error) consisting of nitrogen for common riparian tree species from north-central Florida.

PAGE 25

15 Platanus occidentalis Pinus ellioWi Ostrya virginica Betula nigra Quercus lyrata Quercus nigra Carya glabra Taxodium distichum Quercus laurifdiia Ulmus alata Quercus michauxii Acer negundo Fraxinus caroliniana Carya aquatica Nyssa ogeche Liquidam bar styraciflua Planera aquatica Salix caroliniana Nyssa sylvatica var. biHora Carpinus caroliniana Ulmus americana Fraxinus pennsylvanica Cephalanthus occidentalis Populus deltoides Acer rub rum Chlonanthus virginicus CeWs laevigata Liriodendron tullpifera Sapium seblferum Magnolia grandlflora Tilla americana Percent of dry weight Figure 2-5. In vitro organic matter digestibility (IVOMD) (± 1 standard error) as a percentage of dry mass for 31 common riparian tree species from northcentral Florida.

PAGE 26

16 Hemisphere temperate zone, peak leaf fall occvirs during the warmest months of the year, which also corresponds to the dry season (Lake and Barmuta 1986, Spain 1984, Stewart and Davies 1990). As such, the Northern Hemisphere subtropics appear to occupy a niche intermediate between those of Northern temperate, tropical, and Southern Hemisphere temperate zone streams. Such differences are based on variance in photoperiod and on vegetation. Although the temperate zone is home to various evergreen conifers, north-central Florida hosts both evergreen conifers such as Pinus and evergreen hardwoods. In fact, the three riparian species (Q. nigra, Q. laurifolia, and Magnolia) that have the latest leaf fall in north-central Florida are evergreen hardwoods. Without these species, it is likely that leaf fall would more closely resemble that of the Northern Hemisphere temperate zone. Not all riparian systems in north-central Florida support these three taxa, particularly Magnolia, which is not often found in more frequently inundated floodplains. Thus, the length of time leaves are available, and consequently the spatial and temporal availability of energy sources for microbes and invertebrates, varies considerably based on the vegetative community. The processing rates of leaves depend on their chemical and structural components in concert with various environmental factors (Kaushik and Hynes 1971). Nutrients and levels of inhibitory and structural compounds are important regulators of microbial and invertebrate communities associated with leaf litter in both aquatic and terrestrial systems (Cummins et al. 1989, Merritt et al. 1984). Those leaves with high concentrations of materials that processing organisms require and low concentrations of

PAGE 27

17 those that inhibit colonization are processed preferentially and thus more rapidly (Petersen and Cummins 1974). Leaf nutrient levels can influence decomposition rates in both aquatic and terrestrial systems. Nitrogen concentration is an important determinant of leaf quality in both aquatic (Day 1982, Suberkropp et al. 1976) and terrestrial systems (Melillo et al. 1982), but P appears to be of marginal importance compared to N and inhibitory and recalcitrant compounds such as lignins and tannins. In both the Great Dismal Swamp, Virginia, USA (Day 1982) and in terrestrial shrubland in Spain (Gallardo and Merino 1999), however, P concentration are positively correlated with processing rates. Addition of aqueous P also accelerates decomposition (Elwood et al. 1981). Neither N nor P appear to be sole predictors of processing timing. Pinus elliottii had the lowest concentration of both N and P in its needles and T. americana had the highest P concentration and very high N concentration. Although the low nutrient content of Pinus is typically reflected in ver>' slow processing rates, and high nutrients in Tilia is reflected in very rapid processing (Webster and Benfield 1986), the relation between nutrients and processing rate is less clear for the remainder of the taxa. It would appear that nutrient content itself is not as important as the ratio between nutrients and recalcitrant compounds, a relation that has been noted elsewhere (Canhoto and Gra9a 1995). Relatively recalcitrant and inhibitory compounds in leaves impede colonization and utilization of leaf litter (Chew and Rodman 1979). T,ignins and tannins are among the most important of these because they are relatively recalcitrant (Boulton and Boon 1991), bind with proteins (Haslam 1989), have antibiotic properties (Haslam 1989), and

PAGE 28

18 inhibit processing of structural components such as cellulose and hemicellulose (Kirk et al. 1977). Such properties slow the rate of processing and decrease the efficiency of energy uptake of both microbes and invertebrates. Recalcitrant compounds in leaves in the present study were also related to processing rates. Pinus elliottii and P. occidentalis had the lowest levels of recalcitrant compounds and T. americana the least and this is reflected in the fact that Pinus and Platanus are slow-processed and Tilia is fastprocessed (Webster and Benfield 1986). In fact, processing rates of tree families in Webster and Benfield (1986) are very similar to the plot of trees and their recalcitrant compounds measured in the present study. Cuticle development is also important in influencing processing rates. In the present study, Magnolia was one of the species with the best-developed cuticle. In Chapter 4, shows that although nutrient content of M grandiflora is high and levels of recalcitrant and inhibitory compounds are low, their leaves were the slowest-processed of the five tree species examined. Furthermore, Quercus michauxii was the second slowest to be processed (Chapter 4) and also has a well-developed cuticle. Bunn (1988) and Canhoto and Gra9a (1995) found that heavy cuticles retard leaf processing. The clustering of these taxa into four groups represents a continuum over which leaves are available to processing organisms. The low cuticle development, high nutrients, and low recalcitrant compound levels of T. americana would make it available very quickly. Group 1 is made up of taxa that are likely to be available for processing soon after abscission. Taxa in Group 2 are likely to become available later, as they have higher cuticle development. Those in Group 3 would be available even later. In the case of both Q. laurifolia and Q. nigra, abscission occurs well after that of most taxa, and thus

PAGE 29

19 would not enter the system until January-February. The other taxa in Group 3 have very well developed cuticles and high recalcitrant compound levels, and would not become available until later for that reason. Magnolia grandiflora, predominantly because of its late abscission, would be the last taxon to be utilized. Although not frequently found associated with streams in undisturbed habitats. Pirns elliottii was included in this study because it is planted along streams in vast monoculture plantations throughout much of northern Florida. Sapium sebiferum is an exotic native to Asia and has escaped cultivation and is forming monocultures in riparian areas of the southern U.S. (Bruce et al. 1995) and is increasingly encountered in floodplains of northern Florida. Monocultures of exotic Eucalyptus in Portugal and Spain (Abelho and Gra9a 1996, Pozo et al. 1998) and Salix in New Zealand (Lester et al. 1994.) have altered aquatic macroinvertebrate communities as a result of differences in leaf fall quality and timing. By altering timing and quality of leaf inputs, and consequently macroinvertebrate communities, litter processing and nutrient recycling could be altered, as could entire food webs. Both P. elliottii and S. sebiferum are unlikely to cause similar problems in mixed forests in Florida based on their close statistical associations to other species, but in monospecific stands, they could elicit effects similar to those of Eucalyptus and Salix. P. elliottii is low in nutrients and high in recalcitrant compounds, making their needles of limited use to both terrestrial and aquatic decomposers. Monocultures of S. sebiferum would exclude late-abscising trees as well as those taxa wdth higher levels of recalcitrant compounds and better-developed cuticles. These trees have leaves that would be available for processing later than more easily processed leaves (Gessner and Schwoerbel

PAGE 30

20 1989, Suberkropp et al. 1976), such as those of 5. sebiferum. Thus, an energy source would be available only during a brief period of time. Some exotic understory trees such as Ligustrum spp. are mcreasingly being foun( in north-central Florida riparian zones (Nelson 1996). Others that are unable to tolerate periodic freezing conditions of north-central Florida, such as Casuarina equisetifolia (Nelson 1994), Cupaniopsis anacardioides (Lockhart et al. 1997), Melaleuca quinquenervia (Woodall 1982), and Schinus terebinthifolius (Woodall 1982), have invaded many areas of southern Florida and could expand their ranges northward with continued global warming and monocultures of S. sebiferum and P. elliottii in similar ways. With this method, future species introductions may be analyzed for chemical constituents and leaf fall timing, and relations to other trees can be measured. As such, may be ascertained what effects a highly competitive exotic may have on biota that depend on seasonal leaf inputs.

PAGE 31

CHAPTER 3 STORAGE AND TRANSPORT Introduction Small streams in forested watersheds typically rely on materials derived from outside the stream channel as their primary energy base (e.g., Cummins et al. 1973, Fisher and Likens 1973, Hynes 1975). Such allochthonous materials entering a stream are primarily in the form of coarse particulate organic matter (CPOM), which is predominantly composed of leaves and wood, but also reproductive structures such as blossoms, ftoiits, and nuts, and dead fauna. Once in the stream, most CPOM is trapped by in stream structure such as fallen tree limbs, roots, and rocks. Here, it is processed by a suite of organisms, ranging from bacteria and fungi to macroinvertebrates, and broken down into smaller particles (fine particulate organic matter (FPOM) and dissolved organic matter (DOM)), that are exported downstream and subsequently utilized. This process was synthesized into the "River Continuum Concept" (Vannote et al. 1980), which describes the structure and function of streams from their headwaters to terminus. Although a great deal of research has been performed in the temperate zone on organic matter budgets and litter processing, but relatively little has been done in the subtropics (Cuffney and Wallace 1987b, Dudgeon 1982, Dudgeon 1989, Wallace and Benke 1984), and almost nothing has been done in Florida (Elder and Cairns 1982). The subtropics are unlike temperate regions in that leaf fall is not limited to a brief autumnal period. Instead, it can be extended over several months, and in the case of north-central 21

PAGE 32

22 Florida, litter fall begins in early October and continues through February (Elder and Cairns 1982, Lugo and Zucca 1983), with species like magnolia (Magnolia grandifolia) continuing to drop leaves though late May (Chapter 2). As a result, the dynamics of CPOM fluxes, storage, and processing should be considerably different from in temperate areas, where leaf litter is introduced to the system during a temporally brief period. The current study addresses the quantity and periodicity of particulate organic matter inputs, storage, and transport in a lightly developed subtropical stream in north-central Florida. Stteam communities are greatly dependent on organic matter, and quantification of inputs, storage, and transport will provide us with an idea of when particulate organic matter is available to stream biota. Whether inputs are reflected in in stream storage and transport wA\ also be investigated. Finally, an organic matter budget will be created to account for many of tlie important sources and fates of organic matter in Possum Creek. Methods Study Site Possum Creek is a small stream (approximately 4.8 km long and 1.5-2 m channel width) within the larger Hogtovm Creek watershed in Alachua County, Florida (Figure 21). Water depth varied depending on discharge and location within the stream, but at base flow, depths in runs were around 0.15 m and pool depths ranged from 0.25-1 m. The stream runs through predominantly undisturbed floodplain and low density residential areas. The study site was a nature preserve approximately 1 km upstream of

PAGE 33

;ure 3-1. Map of Possum Creek, Florida, study site, and placement of sampling equipment in the floodplain of the stream.

PAGE 34

24 the confluence with Hogtown Creek. At this point, it is a second-order stream. Like most Florida creeks, the streambed is predominantly sand. The channel is meandering and incised, with the streambed about one meter below the adjacent floodplain. Dominant floodplain tree species include box elder (Acer negundo), red maple (A. rubrum), sweet gum {Liquidambar styraciflud), American ehn (JJlmus americand), ironwood (Carpinus caroliniana), water oak {Quercus nigra), swamp chestnut oak (Q. michauxii), laurel oak {Q. laurifolid), Carolina ash {Fraxinus caroliniana), and magnolia {Magnolia grandiflora). Experimental Approach Inputs, storage, and transport of particulate organic matter were collected biweekly from Possum Creek from November 2000-October 2001 . Direct deposition (litterfall) and lateral deposition were measured inputs. In stream benthic, leaf packs, debris dams, and floodplain were also measured. Downsfream transport was the only transport mechanism quantified. Litterfall and lateral deposition traps were placed in the field two weeks before the first collection. Samples were placed in resealable plastic bags on collection and returned to the laboratory for processing. Direct deposition was measured by placing 10 litter traps (open side facing up) randomly along a 400-m channel length and within 20 m of the channel (Figure 2-1). Traps were modified plastic storage boxes (60.5 x 45 x 45 cm) that had 1-mm diameter holes drilled through the bottom to facilitate water drainage. It was assumed that litter quantity and species composition were representative of material directly deposited into the stream channel.

PAGE 35

25 Lateral deposition (the CPOM wind-transported from the floodplain to the stream) was measured by placing five traps at the boundary of the floodplain and stream channel (Figure 3-1). These traps were constructed of plastic boxes (45 x 25 x 13 cm) with 1-mm diameter holes drilled to facilitate water drainage, and were placed on their edges (on long side) to capture debris entering the channel from the floodplain. Leaf packs are those aggregations of leaves and other organic materials occurring on the stream bottom in quiescent areas. Debris dams are found where woody debris traps organic matter, generally near the stream surface. Leaf pack and debris dam volumes were enumerated from a 200-m stretch of stream that was measured throughout the study. To avoid modification of leaf packs and debris dams, thereby possibly biasing the next temporal sampling, ten samples of CPOM were obtained fi-om each of five leaf packs and debris dams fi-om randomly selected locations downstream of the 200 m stretch. The CPOM was placed into a 12.5 x 7.5 x 7.5-cm box, to provide approximately equivalent measurements, then each of the ten samples were transferred to separate resealable plastic bags for processing in the laboratory. These samples were collected to provide relative proportions of CPOM constituents in the stream at any one time. Benthic organic matter was collected using a stainless steel cylindrical-corer 7.1 cm in diameter and 26.5 cm long with a 64 cm attached handle. Ten cores were taken midstream to a depth of 10 cm every 2 m along the stream length from a randomly selected starting point in the channel within the aforementioned 200 m stretch. Floodplain CPOM lying on the soil surface was collected fi-om five 0.5 m quadrats positioned randomly within 10 m of the stream channel. Samples were collected

PAGE 36

26 only in areas of native vegetation and avoided areas with coverage by exotic ground cover such Tradescantia fluminensis. The CPOM and FPOM export was measured by anchoring a 60 x 45 cm drift net to the stream channel and collecting organic trapped in them over the course of six hours. Stream flow was measured in the center of the water column where the drift net was deployed at the beginning and end of the 6 hours with a Marsh-McBimey Model 2000 Flo-Mate flow meter. Five separate 10 second-averaged measurements were taken both prior to and immediately after the 6-hour period. It was assumed that the mean of these measurements represented the mean flow of the stream between the two measurements. Using these data, the amount of organic matter transported downstream per day could be quantified. The FPOM and CPOM were collected using a 35 ^im mesh drift net. At the end of 6 hours, the contents of each net were washed into a bottle and returned to the laboratory for analysis. For laboratory processing of CPOM in litter traps and in stream accumulations, all materials collected in the field were first dried at 1 10° C for 24 hours in a drying oven. Five-gram (dry mass) subsamples of each constituent were then ashed in a muffle ftimace at 550 °C for 1 hour and were reweighed to obtain ash-free dry mass (AFDM). For leaf pack and debris dam samples, no subsamples were taken and the entire sample was ashed. Benthic organic matter (BOM) from cores was first separated from the mineral substrate by thoroughly washing it through two sieves. Material retained by a 1 mm sieve constituted CPOM and that retained by a 63 jam sieve was designated FPOM. Stream gage height data were obtained from the Hydrologic Data Services Division of the St. John's Water Management District. These data were measured at the

PAGE 37

27 NW 16* Avenue gage station on Possum Creek approximately 0.5 km upstream of the study site. All gage height levels were measured daily at midnight. Data Analysis Spearman rank correlations were used to test the for a relationship between initial leaf cuticle development and nutrient and recalcitrant compound content were unrelated to leaf processing rate. Simple linear regressions were used to relate both CPOM and FROM export to stream gage height and litterfall. Results and Discussion Latitudinal ComparisoDS The extended period of leaf fall in the Possum Creek floodplain differs considerably from the brief autumnal leaf fall typical of Northern Hemisphere temperate systems (Figure 3-2). The duration of peak leaf fall is far more similar to other subtropical and tropical ecosystems than to that of the temperate zone. The timing of leaf fall in the Possum Creek floodplain most closely resembles that of other Florida and Georgia systems (Lugo and Zucca 1983, Watt and Golladay 1999), but also is very similar to several Southern Hemisphere locations (e.g., Daniel and Adams 1984, King et al. 1987, Neiff and Neiff 1990, Enright 1999). Latitude appears to have some influence on both the timing and duration of leaf fall. Amount of litterfall is also in the range of many Southern Hemisphere forests and greater than those of the Northern Hemisphere temperate zone (Table 3-1). Litterfall increases in eastern North America with increasing precipitation and decreasing latitude in response to lengthened growing seasons (Benfield 1997). Florida fits this trend well (Table 3-1), and it is likely that this trend is evident worldwide as well.

PAGE 38

28 Norway (62° N)" British Columbia Canada (49° 16' f Ontario Canada (43° 47' N)' Spain (43°19'N)'' Portugal (39° 30'N)' Kentucky USA (37° 5'N/ South Carolina USA (33° N/ Georgia USA (31°N)'' India (30°4rNy Florida USA (29° 14' N)* Florida USA (29° 14' N)"" Hong Kong (21°27'N)' Mexico (19° 30' N)"" Mexico (18° 34' N)" India (10° 35' N)° Cameroon (4° 30' N/ Bolivia (16° 13' S)' Australia (1 7° 20' S)r Argentina (27° 30' Sf Australia (28° S)' South Africa (33° 50' S)" New Zealand (36° 46' Sf Australia (37° 30' S)' New Zealand (41° 20' S) Tasmania (43° 25' Sf M 1 1 1 1 1 1 r1 M J J A S O N D Figure 3-2. Approximate timing of leaf abscission from selected forests of the world. Solid line indicates timing of leaf fall. * Haapala and Muotka 1998; Richardson 1992; Hill and Brooks 1996; Pozo et al. 1997;' Abelho and Gra(;a 1998; ^Muller and Martin 1983; ^ Shure and Gottschalk 1985; ''Watt and GoUaday 1999; ' Garkoti and Singh 1995; ^ present study; ^ Lugo and Zucca 1983; ' Lam and Dudgeon 1985; Williams-Linera and Tolome 1996; " Alvarez-Sanchez and Guevara Sada 1993; ° Kumar and Deepu 1 992; P Songwe et al. 1988; Justiniano and Frederickson 2000; ' Spain 1 984; ' Neiff and Neiff 1 990; ' Greenway 1 994; " King et al. 1 987; " Enright 1999; Campbell et al. 1992; " Daniel and Adams 1984; ^ TumbuU and Madden 1983.

PAGE 39

Table 3-1 . Litterfall from representative forests along a latitudinal gradient Location Latitude Litterfall (T ha' 7^1 Literature source Portugal 39° N 7.2 Abelhoand Gra^a 1998 Kentucky, USA 37° N 2.9 Muller and Martin 1983 South Carolina, USA 33° N 4.2-5.4 Shure and Gottschalk 1985 Georgia, USA 31°N 4.1-5.8 Watt and Golladay 1999 India 3rN 3.5-6.3 Garkoti and Singh 1995 Florida, USA 30° N 8.0 Elder and Cairns 1982 Florida, USA 29° N 7.7 Present study Florida, USA 29° N 2.8-3.3 Lugo and Zucca 1983 Hong Kong 21°N 12.2 Lam and Dudgeon 1985 Mexico 19° N 8.5 Williams-Linera and Tolome 1996 Australia 17° S 7.3-10.5 Spain 1984 India 11°N 12.2-14.4 Kumar and Deepu 1992 Cameroon 4°N 12.9-14.1 Songweetal. 1988 Argentina 27° S 8.2 Neiffand Neiff 1990 Australia 28° S 6.8-8.1 Green way 1994 South Africa 34° S 4.3-5.0 Kingetal. 1987 New Zealand 37° S 6.8-8.8 Enright 1999 Australia 38° S 6.0-7.0 Campbell et al. 1992 New Zealand 41° S 6.2 Daniel and Adams Tasmania 43° S 4.1-5.6 Tumbull and Madden 1983

PAGE 40

30 In the Northern Hemisphere temperate zone, leaf fall occurs during a brief autumnal period (Anderson and Sedell 1979, Fisher and Likens 1973), but the subtropics and much of the Southern Hemisphere do not follow this partem. In much of the Southern Hemisphere, vegetation is evergreen (Campbell and Fuchshuber 1994), and leaf fall occurs throughout the year, although much of this litter fall occurs during springsummer and at a smaller magnitude than that of autumnal inputs in Northern Hemisphere temperate systems (King et al. 1987, Lake and Barmuta 1986, Stewart and Davies 1990). In this respect. Possum Creek is similar to these Southern Hemisphere systems with litter inputs occurring over a longer period of time than in Northern Hemisphere temperate systems. In Chapter 4 it is suggested that subtropical Possum Creek lies intermediate to Northern Hemisphere temperate and Southern Hemisphere streams, not only in terms of timing of leaf fall, but also in terms of macroinvertebrate life histories. This similarity between Possum Creek and the Southern Hemisphere and difference from the Northem Hemisphere temperate leaf fall reflects primarily the presence of evergreen vegetation in the Possum Creek floodplain. At Possum Creek, deciduous trees experienced leaf fall from October through early December, but evergreens such as Quercus nigra, Q. laurifolia, and M. grandiflora abscised from January through May. In addition, the chemistry and cuticle development of these early and late-abscising species differ considerably, as in Chapter 4 it is shown that earlyabscising deciduous leaves were processed over twice as quickly as those of lateabscising e\'ergreens. These late-abscising trees had leaves with greater cuticle development, which can impede microbial processing (Bunn 1988).

PAGE 41

31 As is the case with litterfall (Benfield 1997), Jones (1997) found that organic matter storage in eastern North American streams increases with increasing precipitation and decreasing latitude. Possum Creek is near the high end of values in eastern North America, but the relation between organic matter, precipitation, and latitude is not fully supported by literature values (Table 3-2). Varying methods and means of quantification are likely affecting these relations, and standardization of methods may allow for a more fully developed model in the future. Organic Matter Budget The organic matter budget shows inputs, storage, and export of particulate organic matter into Possum Creek and its floodplain (Figure 3-3). Dissolved organic matter, biotic storage, respiration, and flood export from the floodplain were not quantified in this study, but are nevertheless components of the budget. The majority of inputs occurred via direct deposition in the form of litterfall. However, DOM can also be a very important source of organic matter to streams (Benke and Meyer 1988) and should not be discounted as a source of energy for Possum Creek. Similar amounts of organic matter were stored both in stream and on the floodplain, and the majority of in stream organic matter was stored in debris dams, followed by leaf packs. Similar to import, exports were dominated by CPOM, but DOM export is often far greater (Benke and Meyer 1988) and is thus likely of considerable importance in Possum Creek. Streams with dense canopies obtain most of their energy from autochthonous sources (e.g., Cummins et al. 1973; Fisher and Likens 1973; Hynes 1975). The majority of these inputs are from litterfall, although lateral deposition and upstream transport are also important. In addition, DOM in groundwater can generally equal or surpass these

PAGE 42

32 Table 3-2. Organic matter storage in representative first and second-order streams along a latitudinal gradient olordge ^^g m ; Quebec, Canada (50 N) yOo iNdimdll 170Z Quebec, Canada (50 N) XTa5»viar> 1 GO jNaiman lysz Oregon, USA (45° N) 401) / 1 nsKa ei ai. i yoz /~V T TO A /ACQ XT\ Oregon, USA (45 N) HIT J 1 1 / insKa ei ai. lyoz iNew riampsmre, uoa rs ^ 610 Pi^Vipr and T.ikpns 1973 Michigan, USA (42° N) 126 Cummins et al. 1981 Virginia, USA (37° N) 1730 Smock 1990 Tennessee, USA (36° N) 175 Mulholland et al. 1985 North Carolina, USA (35° N) 740 Lugthart and Wallace 1992 North Carolina, USA (35° N) 506 Webster etal. 1990 Florida, USA (29° N) 850 Present study

PAGE 43

33 Groundwater DOM:? Litterfall 521.7 Lateral deposition 37.1 Upstream transport 9 Biota 9 Benthic storage CPOM: 3.4 FPOM: 0.6 Leaf pack storage 317.3 Debris dam storage 528.9 Floodplain storage 433 Respiration 9 Export CPOM: 325 FPOM: 28.7 DOM:? Figure 3-3. Organic matter budget for the Possum Creek system, Florida, from October 2000-November 2001 . Solid arrows indicate measured movements, and dashed arrows indicate hypothesized movements of organic matter between input, storage, and export compartments. Inputs (left column of boxes) and exports (right column) are expressed as g AFDM m"^ yr"'. Standing stock in storage (middle column) is expressed as g AFDM m"l

PAGE 44

34 inputs (Minshall 1996). Upstream transport was not measured in the present study, but should be approximately that of export, as has been found in other studies (Fisher and Likens 1973, Fisher et al. 1982). However, although upstream import and downstream export are generally in approximate balance, the total amount in each can vary considerably annually (Cummins et al. 1983). Thus, these measurements, as well as others, in Possum Creek represent a "snapshot" view of the stream. It has been strongly advocated that organic matter budgets not be constructed using data collected over a short temporal frame, but rather multi-year data sets (Cummins et al. 1983). All dominant storage compartments were measured in the present study, and thus the storage portion of the organic matter budget is likely fairly complete. Most organic matter budgets do not distinguish between buried organic matter and that stored in leaf packs or debris dams. As such, it is not possible to compare compartments among studies. In general, little organic matter is stored permanently in streams (Fisher and Likens 1973), but rather processed in situ and exported as smaller particles. Although not quantified in the present study, the majority of this organic matter is exported as DOM (Benke and Meyer 1988). Of the remaining exported materials, CPOM generally is responsible for the larger fraction than is FPOM (Fisher and Likens 1973). This was the case in Possum Creek as well, with an order of magnitude greater CPOM export than FPOM. Inputs Litter fall totaled 10.4 kg AFDM m"^ yr"' in the floodplain for the period November 2000-October 2001. Peak litter fall occurred from early October through late

PAGE 45

35 December, and a smaller peak was evident during February (Figure 3-4). The majority of the litter was leaves (73%), although wood (15%) and reproductive structures (1 1%) were important temporally (Figure 3-5). Leaf fall occurred in the Possum Creek floodplain from mid-October through late February and again from early March through late May. Of the dominant canopy trees, leaves of A. negundo, A. rubrum, Liquidambar, Ulmus, Carpinus, and Fraxinus abscised from mid-October through mid-December. During late November, Q. michauxii leaves began to fall and continued through early January. This was followed by Q. nigra and Q. laurifolia, both of which dropped leaves from early January through late February. Finally, M. grandiflora abscised from mid-March through late May. Those trees whose leaves abscised from mid-October through mid-December contributed the vast majority of the litter fall during the study, with peak litter fall of 12.2 g AFDM m"^ day"', but the smaller peak in litter fall during February resulted from both Q. nigra and Q. laurifolia leaf fall and the addition of reproductive structures, primarily catkins (Figure 3-5). Nuts and seeds were the major constituents of the reproductive structures and were only briefly part of the litter fall (Figure 3-5). Nuts were predominantly acorns of the three Quercus species and, to a lesser extent, pignut hickory {Carya glabra). Peak nut fall occurred during early September through late October. Seeds were dominated in both number and mass by Liquidambar with Fraxinus, Acer, and Tilia seeds as secondary contributors. Seed fall occurred at various times throughout the year, but was orders of magnitude less than that of leaf fall. Epiphytes, primarily Spanish moss (Tillandsia

PAGE 46

36 ONDJFMAMJJASO 2000 2001 Figure 3-4. Organic matter inputs via litterfall (circles) and lateral deposition (triangles) from October 2000 through November 2001 into Possum Creek, Florida. Error bars represent ± one standard error

PAGE 47

37 gure 3-5. Percent by dry mass of constituents of organic matter in litterfall from October 2000 through November 2001 at Possum Creek, Florida. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction.

PAGE 48

38 usneoides), were present in litter traps throughout the year, but were a very minor component of total litter fall. CPOM can originate from both autochthonous and allochthonous sources, but in heavily forested first or second order streams, the canopy precludes much autochthonous production from algae and macrophytes (Vannote et al. 1980). However, algae was observed in Possum Creek from December through February when leaf abscission of many canopy trees allowed greater light penetration. Although algae were not specifically measured for this study, it would appear that they could represent a sizable energy source during winter, although allochthonous production, especially leaf litter, was likely the primary energy source. Wood also serves as a CPOM source, but its major role in small streams is often in retaining leaf litter in debris dams (Bilby 1981, Bilby and Likens 1980, Smock et al. 1989). Leaves enter a stream primarily during abscission, and this is generally the most important input of CPOM annually in allochthonous systems (Hynes 1975). Reproductive structures enter when riparian vegetation flowers, and again when seeds or nuts are mature. Wood inputs are unpredictable and are related to storm and wind events. Little lateral deposition of litter into the stream channel occurred (Figure 3-4) and consisted predominantly of leaves (Figure 3-6). Peak lateral deposition occurred during mid-December 2000 (0.64 g AFDM m'^ day"') and was positively related in general v^th litter fall (r^ = 0.56, p < 0.001), and particularly leaf fall (r^ = 0.94, p < 0.001). Lateral deposition can account for 10-50% of total annual input of CPOM to a stream (Conners and Naiman 1984, Fisher and Likens 1973, Gurtz et al. 1988, Webster et al. 1990, Weigelhofer and Waringer 1994), but contributed far less in Possum Creek.

PAGE 49

39 ONDJFMAMJ JA 2000 2001 Leaves I Epiphytes [x] Wood H Reproductive structures re 3-6. Percent by dry mass of constituents of organic matter in lateral deposition from October 2000 through November 2001 at Possum Creek, Florida. Gap in data represents a flooding event that disabled lateral deposition traps. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction.

PAGE 50

40 Possum Creek lies at the lower end of the range of values reported in the literature (Table 3-3). Organic matter such as leaves and blossoms is more readily transported to the stream via this route due to its lower mass, although some wood and nuts were collected from lateral deposition traps during the course of the study. Possum Creek lies in a heavily forested area with pronounced ground cover. These factors, in concert with the very low floodplain slope, appeared to decrease transport of materials into the stream channel by acting as v^ndbreaks. Even during periods of high wind, little was experienced near the stream. Storage Floodplain organic matter was dominated by leaves throughout much of the year (Figure 3-7). However, a flood in late June 2001 redistributed much of the floodplain litter, and another flood in late September 2001 did the same, to a lesser extent. Much of the litter was not washed into the stream channel but became aggregated in wrack piles, thus variability between samples increased considerably after the flood (Figure 3-8). Wood became proportionally more important in the floodplain litter, as much of the lighter leaf litter was washed into the stream during the flooding. Overall, floodplain organic matter increased to 604 g AFDM m'^ in late February 2001 and decreased slightly temporally preceding June flooding (Figure 3-8). CPOM not directly deposited into the stream during storms or abscission is stored on the floodplain. Stream channels are dynamic systems, with CPOM constantly being imported, exported, and transformed (Smock and P.oeding 1986). In contrast, many floodplains are less dynamic, with these processes occurring to a lesser degree, especially where spates infrequently inundate portions of the floodplain (Smock 1990), which is

PAGE 51

41 Table 3-3. Lateral deposition from representative forests along a latitudinal gradient Location Latitude Lateral deposition gm"^yr"' Citation Alaska, USA 65° N 19 Irons and Oswood 1997 Quebec, Canada 50° N 344 Naiman 1982 Quebec, Canada 50° N 56 Naiman 1962 uregon, UoA Tri<:ka et al 1982 Oregon, USA 45° N 1111 Triskaetal. 1982 Tennessee, USA 36° N 106 MulhoUand et al. 1985 North Carolina, USA 35° N 71 Webster etal. 1990 North Carolina, USA 35° N 137 Wallace etal. 1995 Florida, USA 29° N 37 Present study

PAGE 52

42 ONDJFMAMJJ 2000 2001 H Leaves fl Epiphytes [y] Wood I Reproductive structures Figure 3-7. Percent by dry mass of constituents of organic matter stored on the Possum Creek floodplain, Florida, from October 2000 through November 2001 . Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction.

PAGE 53

43 Figure 3-8. Organic matter storage on the Possum Creek floodplain, Florida, from October 2000 through November 2001 . Error bars represent ± one standard error.

PAGE 54

44 often the case in low-order streams such as Possum Creek. These factors, combined with the broader area of storage of floodplains as compared to the stream channels, result in greater floodplain storage of CPOM. Some processing occiu-s on the floodplain by terrestrial invertebrates, bacteria, and fungi (Merritt and Lawson 1992), but perhaps the greatest importance of floodplains for stream biota is as a source of CPOM (Cuf&iey 1988, Smock 1990). CPOM is brought into streams by spates and lateral wind deposition. As a result, CPOM can be present year round, although its addition to the stream channel is mediated by storm events (Hill and Brooks 1996, Ractliffe et al. 1995). Floodplain CPOM is present year round for Possum Creek, although quantities decreased during the summer wet-season as a result of both decreased deposition and increased processing due to higher temperatures and wetter conditions (Peterson and Rolfe 1982, Singh 1984) and transport to the stream chaimel during floods. Most in stream organic matter was stored in debris dams (Figure 3-9). This storage component was positively related to leaf fall (r^ = 0.46, p < 0.001), with peak storage (0.16 m^ -100 m stream length'') occurring during late December 2000. Leaf packs were present only during and immediately after peak leaf fall, with peak storage (0.081 m -100 m stream length' ) occurring during early December 2000 (Figure 3-9). Debris dam organic matter was dominated by leaves from early October 2000 through early January 2001, but by small diameter woody debris for much of the remainder of the year until leaf fall again increased during late September 2001 (Figure 3-10). Leaf packs gradually became dominated by small-diameter woody debris after mid-November, but

PAGE 55

45 O) 0.25 n c e 2000 2001 Figure 3-9. In-stream organic matter storage in leaf packs (white) and debris dams (black) per 100 m of stream length from October 2000 through November 2001 in Possum Creek, Florida.

PAGE 56

I 46 Figure 3-10. Percent by dry mass of constituents of in leaf packs (upper) and debris dams (lower) from October 2000 through November 2001 in Possum Creek, Florida. Gaps in data represent periods where no leaf packs were present. Reproductive structures include nuts, seeds, blossoms, and other structures associated with floral reproduction. |:

PAGE 57

47 by May almost completely consisted of woody debris, and disappeared altogether after that. Debris dam storage is an important reservoir of CPOM in streams, especially in high velocity streams where storage on the streambed is small (Angradi 1996, Bilby 1981, Bilby and Likens 1980, Bretschko and Leichtfried 1988, Gonzalez and Pozo 1996, Smock 1990, Smock et al. 1989). CPOM storage in debris dams is greatest during senescence, when most leaves are entering the stream and decreases as they are processed and fragmented. CPOM in temperate stream debris dams can be available all year, as those leaves in the center of the dam are protected against processing because of low dissolved oxygen (Reice 1974). As outer leaves are processed, the oxygenated environment proceeds towards the inner leaves allowing them to be processed (Reice 1974). I However, despite an extended period of leaf fall, CPOM was not present in Possum Creek throughout the year, and very little organic matter was present during summer. Elevated annual water temperatures in subtropical streams such as Possum Creek, as compared to temperate systems, facilitate more rapid processing of organic matter (Benstead 1996) and, subsequently, the presence of little in stream organic matter during summer. Even during venter, mean stream temperatures rarely dropped below 10° C, allowing fungal processing of organic matter to occur at higher rates than would occur in cooler temperate areas during the same time period (Suberkropp and Chauvet 1995). Leaf packs can be either an important or unimportant store of CPOM and this is predicated by discharge (Angradi 1996, Smock 1990, Smock et al. 1989). In streams with low discharge, leaf packs can form throughout the streambed and remain intact

PAGE 58

48 unless disturbed by a spate. Conversely, higher discharge streams have fewer areas (generally pools and backwaters) where leaf packs may form. As with debris dams, leaf packs can sustain stream communities by providing CPOM throughout the year. High discharge in Possum Creek limited leaf packs to only the most quiescent areas (i.e. deep pools and backwaters). Benthic CPOM peaked at 7.4 g AFDM m'^ in the surface sediments at the time of peak leaf fall during October 2000 and approximated 3 g AFDM m"^ for the remainder of the year (Figure 3-11). FPOM was consistently an order of magnitude lower than CPOM, about 0.5 g AFDM m"^ throughout the year. Spates can cause the substrate to be either a source or sink of CPOM (Metzler and Smock 1990). During high leaf fall, spates will generally bury more leaf litter relative to that released. Conversely, spates generally release CPOM to the stream channel. Low processing rates in the oxygen-deprived BOM shifts its utilization to when it is brought to the sediment surface by spates (Strommer and Smock 1989). Along with green leaf fall associated with storm events, this appears to be the dominant manner in which CPOM in Possum Creek is made available to the stream community, even though relatively little organic matter was stored in the sediment. Transport from the floodplain is also of some importance, but the low flood frequency limited its value as a CPOM source to the two flooding events during the study. Buried CPOM or benthic organic matter (BOM) is often the most important reservoir for CPOM in streams (Cummins et al. 1983, Metzler and Smock 1990, Rounick and Winterboum 1983, Smith and Lake 1993, Smock 1990). Most streams have an ample supply of CPOM stored in their sediments for processing throughout the year, and

PAGE 59

49 Figure 3-11. In-stream benthic organic matter storage as CPOM (circles) and FPOM (triangles) from October 2000 through November 2001 in Possum Creek, Florida. Error bars represent ± one standard error.

PAGE 60

50 V25 n o "e 20 S Q liE 15 S 5; 10 ra I = c CO P) n ° NDJFMAMMJJASO 2000 2001 Figure 3-12. In-stream organic matter transport as FPOM (white) and CPOM (black) from October 2000 through November 2001 in Possum Creek, Florida.

PAGE 61

Figure 3-13. Gage height at Possum Creek, Florida, from September 2000 to October 2001 . Asterisks (*) represent instances when water height .exceeded bankfuU stage.

PAGE 62

52 Table 3-4. Organic matter export from representative first and second-order streams along a latitudinal gradient Location Export (kg yr"') Citation Alaska, USA (65° N) 1 AT 1 1071 Irons and Oswood 1 997 Alaska, USA (65° N) 1133 Irons and Oswood 1 997 Oregon, USA (45 N) J / insKaetai. lysz Oregon, USA (45° N) 245 Triskaetal. 1982 New Hampshire, USA (44° N) 1700 Fisher and Likens 1973 Virginia, USA (37° N) 13,751 Jones and Smock 1991 North Carolina, USA (35° N) 4,326 Webster etal. 1990 Florida, USA (29° N)) 2,506 Present study Puerto Rico (18° N) 211 McDowell and Asbury 1994

PAGE 63

53 this is a more stable source of CPOM than either leaf packs or debris dams, which are rather dependent on seasonal leaf abscission (Smock 1990). However, the sand sediment of Possum Creek contained little organic matter as compared to debris dams, leaf packs, and floodplain storage. Smock (1990) found that organic matter in a sand substi-ate was very high, although no storage occurred in clay substrates. Possum Creek appears to be intermediate in that although the substrate is predominantly sand, it overlies a clay layer 0-60 cm below the sand. This clay may have limited the organic matter storage potential of the sand substrate as the relatively homogenous particle size of the sand substrate may have acted as an impediment for trapping of organic matter. Transport CPOM and FPOM transport were related to both leaf fall and stream discharge. Two peaks in transport were evident, one from mid-November through mid-December (peak 22 kg AFDM m"^ day"^) resulting from leaf litter inputs during that period, and another from early June through late July (peak of 16 kg AFDM m'-^ day"') (Figure 3-12). The summer peak was related to high discharge during the Florida surmner wet-season (Figvire 3-13). Transport values are highly variable across North America (Table 3-4), and Florida is intermediate. Organic matter is positively correlated to discharge (Webster and Meyer 1997), and comparing transport rates is difficult among streams that vary considerably in gradient, order, and precipitation. Downstream transport of both CPOM and FPOM is of great importance in the energy dynamics of stream ecosystems (Vannote et al. 1980). Although studies have shown that discharge and CPOM are poorly correlated, they are sensitive to the effects of both spates and maximum leaf fall (Bilby and Likens 1979). Coarse particulate organic

PAGE 64

54 matter export was not related to discharge over the entire study (r^ = 0.051, p = 0.278), but leaf fall partially obscured this. Looking only at CPOM export during MarchAugust 2001, there was a very strong correlation (r^ = 0.816, p < 0.001). From SeptemberFebruary there was a strong correlation between CPOM transport and litter fall (i^ = 0.557, p < 0.02). Peak CPOM transport occurred during the period of maximum leaf fall and again during the summer wet-season. Storms occurred almost daily from late May through late July and kept the discharge of Possum Creek higher than base flow (Figure 3-13). FPOM transport is more accurately predicted from discharge (Wallace et al. 1991). This is the case in Possum Creek, as FPOM export was related to discharge (r^ = 0.196, p = 0.03). This was particularly the case from MarchAugust 2001 when the effect of leaf fall could be factored out (r^ = 0.60, p < 0.001). Higher FPOM export during summer spates may also be influenced by lower retention in depleted debris dams (GoUaday et al. 1989). In many temperate streams, macroinvertebrate shredders have been linked to FPOM generation and, consequently, transport (Cuffney et al. 1990, Wallace et al. 1991). However, the relative lack of shredders in Possum Creek (Chapter 4) suggests that they are of minor importance in FPOM generation and transport, and this may also be the case in other subtropical streams with low shredder densities (Cuffriey and Wallace 1987a, Dudgeon 1994, McArthur et al. 1994). This is partially evident as the lack of correlation between litter fall and FPOM (r^ = 0.01, p = 0.636). Organic matter inputs and storage in leaf packs and debris dams were greatly influenced by the extended period of leaf fall as compared to temperate systems. This relation was less pronounced for lateral deposition, benthic storage, and floodplain

PAGE 65

55 Storage. Spatial floodplain organic matter storage, however, appeared to be related to moisture and flooding events associated with the Florida summer wet-season. Although rainfall data were not available from this study site, floodplain leaf litter was noticeably more moist during the summer than during the winter dry-season. Downstream organic matter transport was also related to increased discharge during the wet-season, as well as to leaf fall. This information is critical to understanding the sources and periodicity of organic matter inputs into streams, and consequentiy, the food resources for in stream fauna. Alteration in either the timing of leaf fall or ability to store organic matter effectively could have implications for the stream biotic community by reducing inputs during periods to which macroinvertebrates may have synchronized their life histories.

PAGE 66

CHAPTER 4 LITTER PROCESSING Introduction Many macroinvertebrates in low-order streams are dependent on seasonal inputs of leaf litter as an energy source. Although studied in great detail in the temperate zone, a paucity of information exists for the subtropics on litter processing (Dudgeon 1982; Elder and Cairns 1982; Hauer et al. 1986; Cuffhey and Wallace 1987; McArthur et al. 1994). Many temperate areas are characterized by leaf fall limited to a relatively brief autimmal period corresponding to the end of the growing season, and macroinvertebrate life cycles are adapted to this seasonality of organic matter inputs (Cummins 1974). However, in many Southern Hemisphere streams, macroinvertebrate life cycles are not synchronized to leaf fall that is spread over much of the year (Campbell and Fuchshuber 1994). The subtropics differ from temperate areas in terms of both leaf fall seasonality and length of growing season. North-central Florida is located near ihe northern boundary of the subtropics in North America and is characterized by a continuous growing season and leaf fall that occurs from October through May (Chapter 2). The input of leaves during this period occurs over a considerably longer period than further north, and may have considerable influence on stream macroinvertebrate life cycles. Although timing of leaf fall can structure macroinvertebrate commimities, the nutritional content and physical structure of each leaf species will help dictate how 56

PAGE 67

57 rapidly each is processed and, as a result, how long a species may remain available to stream macroinvertebrates (Petersen and Cummins 1974). Processing is the biological and physical processes by which coarse particulate organic matter (CPOM) is utilized by stream biota to yield fine particulate organic matter (FPOM) and dissolved organic matter (DOM) (Petersen and Cummins 1974). Both physical and biotic means are involved in leaf processing. CPOM is physically fi-agmented by abrasive forces governed by current. Biological processing initially involves fungi and bacteria (Benfield et al. 1977), and later in many systems, invertebrate shredders (Cummins et al. 1973). Leaves with low nutritional value may not become available for macroinvertebrate processing until they are first conditioned by fungi and bacteria, and it is this microbial community that can provide much of the energy requirements for macroinvertebrates (Anderson and Sedell 1979; Cummins and Klug 1979). This study will determine the rates at which leaves of five common riparian trees are processed and examine the temporal patterns of abundance for stream macroinvertebrates in relation to sequential leaf organic matter additions. It is expected that the processing rates of the species used in this study and macroinvertebrate colonization rates reflect their nutrient content and cuticle development. Finally, the trophic structure of subtropical stream macroinvertebrate communities will be examined to determine the abundance of functional feeding groups. Methods Study Site Possum Creek is a small, second order stream (approximately 4.8 km in length) in the larger Hogtown Creek watershed, Alachua County, Florida. The study area was in a

PAGE 68

58 nature preserve approximately 1 km upstream from the confluence with Hogtown Creek. The floodplain has few obvious anthropogenic alterations with the exception of a sewer line that at places parallels and bisects the streambed. Despite few outward appearances of anthropogenic activity, at least 15 exotic plants inhabit the floodplain. Like many Florida creeks, the streambed is predominantly sand, although outcroppings of clay are present where the stream cuts through the Hawthorne layer. The channel is incised, and the streambed is about a meter below the surrounding floodplain. Dommant canopy tree species include box elder (Acer negundo), red maple (A. rubrum), sweet gum {Liquidambar styraciflua), American elm {Ulmus americana), ironwood {Carpinus caroliniand), water oak (Quercus nigra), swamp chestnut oak (Q. michauxii), laurel oak (Q. laurifolid), Carolina ash (Fraxinus caroliniand), and magnolia (Magnolia grandiflord). Experimental Approach Leaves of individual tree species were collected for a mesh-bag study once each species became abundant in litter traps. Ten litter traps were placed randomly throughout the floodplain within 20 m of the stream channel, and contents of each trap were collected biweekly beginning in October 2000. Five species were selected based on ease of collection and their relative timing of leaf fall. Acer rubrum, Liquidambar syraciflua, and Ulmus americana were selected to represent the early abscising taxa, Q. michauxii was the mid-season-abscising taxa, and Magnolia grandiflora represented the only lateabscising taxon. One hundred g (dry mass) of freshly abscised leaves from each species were collected from the floodplain during peak leaf fall and placed into plastic bags. Leaves were air-dried for 48 hours to standardize for water content.

PAGE 69

59 Initial nutrient content for leaves of each species was analyzed for nitrogen, phosphorus, and recalcitrant compounds (including phenolics). For nitrogen and phosphorus analyses, samples were digested using a modification of the aluminum block digestion procedure (Gallaher et al. 1975). Sample mass was 0.25 g, the catalyst used was 1.5 g of 9:lK2S04:CuS04, and digestion was conducted for at least 4h at 375°C using 6 ml of H2SO4 and 2 ml H2O2. Nitrogen and phosphorus in the digestate were determined by semiautomated colorimetry (Hambleton 1977) (Appendix 1). In-vitro organic matter digestion (IVOMD) was performed by a modification of the two-stage technique (Moore and Mott 1974) (Appendix 2). The IVOMD measures those leaf constituents that are easily digestible by ruminant gut microflora. Material that is not digested, such as tannins and lignins, is recalcitrant and of limited use to processing microflora. Each 1 cm-mesh bag contained 5 g of leaf material of a single species. These bags measured 45 cm^, leaves placed in the center, comers folded to the center, and cinched with rigid plastic ties. Seventy-five bags were constructed for each species and placed in the stream, tethered to steel chains stretched across the channel in five randomly selected areas displaying approximately the same velocity (0.25-0.30 m s'^ at base flow). Bags were placed in the stream corresponding to the period of leaf fall for a particular species: A. rubrum, L. styraciflua, and U. americna in mid December, Quercus michauxii in early January, and M grandiflora in early April. Bags were collected every 100, 200, 300, 500, 700, 900 degree-days (sum of mean daily air temperatures) to standardize for temperature effects. One or two bags of each species were collected randomly from each steel chain. Remaining bags were removed from the stream after

PAGE 70

60 900 degree days: A. rubrum, L. styraciflua, and U. americana in late February, Q. michauxii in mid March, and M grandiflora in late May. For the first three sampling periods, five bags each were collected, and for the final three periods, ten bags each were collected. Bags were collected more frequently initially because of anticipated rapid processing of the material, and more samples were collected later in the study to reduce variability among bags. After collection, bags were immediately brought to the laboratory for processing. In the laboratory, samples were gently cleaned to remove macroinvertebrates and the contents washed into a 125 |am sieve. The collected material was then transferred to bottles containing 70% ethyl alcohol and stained with the vital stain Rose Bengal to aid in sorting. All macroinvertebrates were removed manually, transferred to vials containing 70% ethyl alcohol, and identified to lowest practical taxonomic level using Pennak (1989), Epler (1995), and Merritt and Cummins (1996). Macroinvertebrate identification primarily took place under a 4.5x (with lOx oculars) Meiji stereoscope, though chironomid dipterans were placed in CMC10 mounting medium on a clean glass slide and viewed under a Fisher Scientific Micromaster CK compound microscope (4x to lOOx with lOx oculars). Those samples with large numbers of macroinvertebrates were subsampled on a gridded plate under a stereoscope, and at least 200 individuals were removed for identification. Leaf material separated from the macroinvertebrates was placed into drying ovens for 24 hours. Initially, unprocessed 5 g leaf samples were placed into the muffle ftimace at 550 °C for 1 hour to determine initial ash-free dry mass (AFDM). AFDM was also determined for leaf material remaining at each collection date.

PAGE 71

61 Data Analysis The rate coefficient for leaf pack AFDM (k) for each leaf species was estimated by regressing mesh bag AFDM against time using the exponential decay model (Petersen and Cummins 1974): Where WM AFDM after time / and Wi is the initial AFDM. To control for temperature effects, / was expressed in degree-days. One-way ANOVA's with Scheffe's post-hoc tests were performed to test the null hypothesis that leaf processing does not differ among species (a = 0.05), and that macroinvertebrate abundance does not differ among leaf species. Spearman rank correlations were used to test the null hypothesis that initial invertebrate abundance and leaf processing rate were unrelated. Simple linear regressions were used to relate macroinvertebrate abundance to organic matter storage (Chapter 4). Organic matter quantity was interpolated assuming a linear change in value between sampling dates. Results Mean air temperatures were as follows: December (9.7 °C), January (9.7 °C), February (16.5 °C), March (17.8 °C), April (18.9 °C), and May (21.4 °C). Acer rubrum and Ulmus americana leaf packs were processed most rapidly (-k = 0.0021 degree days"' and 0.0022 degree days"', respectively), followed closely those of by Liquidambar styraciflua (k = 0.0019 degree days"'). Quercus michauxii and Magnolia grandiflora leaf packs were processed much more slowly (-k = 0.0009 degree days"' and -k = 0.0008 degree days"', respectively). As a result of the increasing stream temperatures from December-May, M. grandiflora was processed over a shorter time period than A. rubrum.

PAGE 72

62 U. americana, L. styraciflua, and Q. michauxii. Nevertheless, the standardization to degree-days controls for temperature statistically. Significant interspecific differences (p < 0.01) were noted in leaf content for each of phosphorus, nitrogen, and recalcitrant compounds, as all species were significantly different from one another for each variable. Liquidambar styraciflua had highest percentage as phosphorus (0.32%), and M grandiflora had the least (0.07%) (Figure 41). Nitrogen was highest in U. americana (0.97%) and lowest in^. rubrum (0.44%) (Figure 4-1). Q. michauxii had the highest percentage of recalcitrant compounds (74%), and M grandiflora had the lowest (46%) (Figure 4-1). Neither P, N, nor recalcitrant compounds were correlated to processing rates, although cuticle development was marginally correlated (r = 0.32, p = 0.052). Over the course of the study, 34 macroinvertebrate taxa were collected from leaf packs in Possum Creek. Macroinvertebrate abundance decreased through time for y4. rubrum, L. styraciflua, and U. americana leaf packs, but peaked midway through processing for Q. michauxii and M. grandiflora (Figure 4-2). Few shredders were collected from leaf packs, and were represented primarily by Tipula sp. and secondarily Hyalella azteca and some chironomids (Table 4-1). Tipula sp. were collected frequently on^. rubrum, L. styraciflua, and U. americana leaf packs, but were less frequent on Q. michauxii and especially M grandiflora (Figure 4-3). No trend could be seen over the course of processing for any leaf packs. Collector-gatherers were dominated by Stenonema sp. and Microcylloepus sp., and some chironomids (Table 4-1). Stenonema sp. abundance increased somewhat over the course of processing for ^. rubrum, L. styraciflua, and U. americana, but remained

PAGE 73

Liquidambar styraciflua Quercus michauxii Ulmus americana Acermbrum Magnolia grandiflora Ulmus americana Quercus michauxii Liquidambar styraciflua Magnolia grandiflora Acerrubrum Quercus michauxii Liquidambar styraciflua Ulmus americana Acermbrum Magnolia grandiflora 0 10 20 30 40 50 60 70 80 Percent of dry weight Figure 4-1 . Percent of dry weight of phosphorus, nitrogen, and refractory compounds in the senescent leaves of five tree species collected from October 2000May 2001 . Error bars represent ± one standard error.

PAGE 74

64 Br §j c a "a c o c c o X CO H cx4 3 DO 00 C •3 d) c3 C o c 3 c o X C3 ^ C ^ £ ^ 5 "S 5 * ^ C3 ^ M S 00 OB 00 is o i 2 2 2 ^ S o ° t) o o 5 ig t; u u u C u — o — s zis 5 o ~ o o o — a o ca U 00 o u u IO OK 0.0. CO cU c3 •a o 3 T3 •a ° 2 2 2 •a u M B M O iJ -O T3 u u u u: i-T i-T t-T u 53 53 O 4> *0 "O "O tj 73 " O.CI.&.VI crt c* O o V2 fcCO Or; S IT! »« crt R 2 -S 2 f8 o S i « -Si a. 3 2 o •S « cu p o a A X a ;o "E o 00 o o. o 3 .e-u Q .— U I 5 5 a S5 S ° •Si 6 a •S Q. S s a ^ I a. 00 g o. 00 ^ -« a §1 -5 In S §-g Q. 5 o § S " -r I 5 § I Q, 5 q; E Co a u u ss 00 T3 a 00 t2 00 o o. s T3 CD T3 T3 U a •o S 'E •2 £:> 5i •§ I 2 8-2P -§ < u u o

PAGE 75

65 12/1212/19 1/7 1/16 2/3 2/16 2/27 4/8 4/13 4/17 4/23 5/4 5/14 5/22 0 100 200 300 500 700 900 0 100 200 300 500 700 900 12/1212/19 1/7 1/16 2/3 2/16 2/27 0 100 200 300 500 700 900 Figure 4-2. Macroinvertebrate abundance and leaf carbon content over time collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack"' and white bars represent macroinvertebrates g C"'-leaf pack''. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error.

PAGE 76

66 relatively stable for Q. michauxii and M. grandiflora leaf packs (Figure 4-3). Microcylloepus sp. was rarely collected from leaf packs during December-January, but made up a considerable portion of non-chironomid abundance from February-May (Figure 4-3). Because A. rubrum, L styraciflua, and U. americana leaf packs were placed into Possum Creek in December and removed in late February, few Microcylloepus sp. were noted until 700-900 degree days. Conversely, Microcylloepus sp. were abundant on M. grandiflora leaf packs and became more so through the course of processing. There was a marginal relationship between Microcylloepus sp. Abundance and organic matter storage (r^ = 0.21, p 0.06), and Stenonema sp. was strongly related to organic matter storage (r^ = 0.67, p < 0.001). Cheumatopsyche sp. and Simulium sp. were the most frequent collector-filterers associated with leaf packs (Table 4-1). Cheumatopsyche sp. abundance increased over the course of processing in^. rubrum, L. styraciflua, and U. americana, but remained relatively stable in Q. michauxii and M. grandiflora, although lower than peak levels in the former three species (Figure 4-4). Simulium sp. were frequently collected from leaf packs from December through early May, but virtually disappeared thereafter As a result, Simulium sp. was rarely found on M. grandiflora leaf packs after 200 degree days (Figure 4-4). The abimdance of both taxa was positively related to organic matter storage {Cheumatopsyche sp., r^ = 0.34, p = 0.03; Simulium sp., = 0.49, p 0.006) Predators such as Corydalus sp. and Argia sp. were not frequently collected from Possum Creek leaf packs until February (Figure 4-5), thus lagging behind prey abundance. In general, not only did predator abundance increase from December-May, but also over the coarse of processing for each leaf species (Figure 4-5). The abundance

PAGE 77

67 Acer rubrum Acer rubrum 4 -r Acer rubrum 3 + rh 2 1 I 12n9 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 20 Liquidambar styraciflua jril i 12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 10 J Liquidambar Styraciflua 8 6 4 2 12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 .Liquidambar styraciflua 70 60 50 40 30 20 10 12/19 1/7 1/16 2/3 2/16 2/27 100 200 3C 500 700 900 Ulmus americana 12A9 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Ulmus americana I 12/19 1/7 1/16 2/3 2A6 2/27 100 200 300 500 700 900 Quercus michauxii 12/19 1/7 1/16 2/3 2A6 2/27 100 200 300 500 700 900 Ulmus americana 12/19 1/7 1A6 2/3 2/16 2/27 100 200 300 500 700 900 Quercus michauxii 1 T 13/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Quercus michauxii 1/16 i/25 2^ 2/16 2/27 3/11 100 200 300 500 700 900 1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900 10 J Magnolia grandiflora 25 j Magnolia grandiflora 03 0^ 020.15 0.1 0.05 1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900 Magnolia grandiflora 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Stenonema (Ephemeroptera) 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Microcylloepus (Coleoptera) Tipula (Diptera) Figure 4-3. Dominant shredders and collectorgatherers collected from Possum Creek from December 2000 to May 2001 . Gray bars represent macroinvertebrates-leaf pack'^ and white bars represent macroinvertebratesg C"'-leaf pack"'. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error.

PAGE 78

I' 68 25 T 20 15 10 5 0 u n o. u B A o B S3 X! < Acer rubntm 16 14 12 10 8 6 4 2 0 14 12 10 8 6 4 2 0 12/19 1/7 1/16 2/3 2^6 2/27 100 200 300 500 700 900 Liquidambar styraciflua 12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Ulmus americana 12A9 1/7 1/16 2^ 2^6 2/27 100 200 300 500 700 900 Quercus michauxii 5 4 3 2 1/16 1/25 2/3 2^6 2/27 3/11 100 200 300 500 700 900 Magnolia grandiflora 350 300 250 200 150 100 35 30 25 20 15 10 5 0 600 500 400 300 200 100 0 30 25 20 15 10 5 0 Acer rubrum 1 12A9 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Liquidambar styraciflua 12^9 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Ulmus americana 12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Quercus michauxii 1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900 Magnolia grandiflora 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Cheumatopsyche (Trichoptera) 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Simulium (Diptera) Figure 4-4. Dominant filter-feeders collected from Possum Creek from December 2000 to May 2001 . Gray bars represent macroinvertebrates-leaf pack'' and white bars represent macroinvertebrates-g C'-leaf pack'\ Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error.

PAGE 79

I 69 Acer rubrum Acer rubrum 12^9 1/7 1/16 2/3 M6 2/27 100 200 300 500 700 900 5 J Liquidambar styraciflua 4 3 2 + 6 5 + 4 3 2 1 12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Ulmus amerlcana -+60 50 40 30 20 10 0 14 12 10 8 6 4 2 0 12/19 1/7 1/16 2^ 2/16 2/27 100 200 300 500 700 900 Liquidambar styraciflua 12/19 1/7 1/16 2/3 2^6 2/27 100 200 300 500 700 900 Ulmus amerlcana 12/^9 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 2 T Quercus michauxii 12/19 1/7 1/16 2/3 2A6 2/27 100 200 300 500 700 900 Quercus michauxii 1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900 3 y Magnolia grandlflora 2.5 1/16 1/25 2/3 2/16 2/27 3/11 100 200 300 500 700 900 Magnolia grandlflora 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Corydalus (Megaloptera) 4A3 4/17 4/23 5/4 5A4 5/22 100 200 300 500 700 900 Argia (Odonata) Figure 4-5. Dominant predators collected from Possum Creek from December 2000 to May 2001. Gray bars represent macroinvertebrates-leaf pack'' and white bars represent macroinvertebrates-g C"'-leaf pack"'. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error.

PAGE 80

70 of both Corydalus sp. (r^ = 0.01, p = 0.7) nor Argia sp. (r^ = 0.02, p = 0.82) was not related to organic matter storage. Chironomids were the most frequently collected macroinvertebrates from all leaf packs. They were far more abundant during December-February than later and their abundance was strongly related to organic matter storage (r^ = 0.66, p < 0.001). Peak abundance per gram carbon of A. rubrum, L. styraciflua, and U. americana was on degree-day 900, although absolute abundance was highest earlier and decreased though the coiiTse of processing (Figure 4-6). However, absolute abundance in Q. michauxii and M grandiflora leaf packs peaked around degree-day 300. The abundance of chironomids was strongly related to organic matter storage (r^ = 0.66, p < 0.001). Eleven chironomid taxa were collected from leaf packs from December-May (Table 4-1). CricotopuslOrthocladius sp. was not collected from leaf packs until early February, and only infrequently after early May (Figure 4-7). Pentaneura inconspicua, the dominant predaceous chironomid, Rheochcotopus robacki, and Tanytarsus sp. were dominant throughout the study for all leaf species, with numbers remaining relatively stable (Figure 4-7). Corynoneura sp. was a dominant chironomid through late January, but became a minor part of the chironomid community, then decreased and remained stable through May (Figure 4-7). Rheocricotopus robacki and Corynoneura sp. displayed similar patterns, but imlike Corynoneura, Rheotanytarsus was collected from M grandiflora leaf packs (Figure 4-7). Polypedilum illinoense gp. was fairly common during December-early January in A. rubrum, L. styraciflua, and U. americana, but were virtually absent for the rest of the

PAGE 81

71 Acer rubrum 800 700 600 500 400 300 200 100 0 12/19 1/7 1/16 2/3 2A6 2/27 100 200 300 500 700 900 Liquidambar styraciflua 12/19 1/7 1/16 2/3 2/16 ^27 100 200 300 500 700 900 Ulmus americana 12A9 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 Quercus michauxii 1/16 1/25 2/3 2A6 2/27 3/11 100 200 300 500 700 900 Magnolia grandiflora 4/13 4/17 4/23 5/4 5/14 5/22 100 200 300 500 700 900 Figure 4-6. Chironomidae (Diptera) collected from Possum Creek from December 2000 to May 2001 . Gray bars represent macroinvertebrates-leaf pack"' and white bars represent macroinvertebrates-g C'-leaf pack''. Both sampling date and cumulative degree days are shown. Error bars represent ± one standard error.

PAGE 82

12/19 ^^ 1/16 2/3 2/16 2/27 12/19 1/7 1/16 2/3 2/16 2/27 100 200 300 500 700 900 100 200 300 500 700 900 Ulmus americana Quercus michauxii Figure 4-7. Chironomidae (Diptera) relative abundance over time collected from Possum Creek from December 2000 to May 2001. Both sampling date and cumulative degree days are shown.

PAGE 83

73 processing period for these three species (Figure 4-7). No P. illinoense gp. were collected from Q. michauxii leaf packs, but they were fairly abundant throughout in M. grandiflora leaf packs. Discussion Leaf Chemistry Leaf litter varies in the degree to which it can be processed. This depends oncharacteristics of both the organic matter and stream including: the tree species from which it originates (Petersen and Cummins 1974; Webster and Benfield 1986), whether it has been conditioned on the floodplain prior to entrainment in the stream (Merritt et al. 1984; Merritt and Lawson 1992), water velocity (Reice 1974), water temperature (Petersen and Cummins 1974; Webster and Benfield 1986), and chemical constituents of the stream water, primarily nitrogen (Kaushik and Hynes 1971; Howarth and Fisher 1976), phosphorus (Howarth and Fisher 1976), and pH (Griffith and Perry 1993). The present study controlled for these factors except tree species by placing all leaf packs in the same stream under similar conditions and using degree days to normalize temporal temperature differences. After initial leaching of soluble compounds from leaf litter, colonization by microorganisms increases both the amount of nitrogen and protein and degradation of structural components such as cellulose, thus making them more palatable for shredders (Kaushik and Hynes 1971; Suberkropp et al. 1976; Anderson and Sedell 1979). This process, termed conditioning, occurs to all leaves, but at varying rates, depending on leaf chemistry and structure. As a result, shredders generally utilize leaves only after they have been conditioned (Petersen and Cummins 1974; Suberkropp et al. 1983).

PAGE 84

74 Initial leaf quality is often the factor determining processing rates of leaves (Kaushik and Hynes 1971; Merritt et al. 1984). Individual species contain different concentrations of nutrients as well as both chemical and physical deterrents, and these influence processing rates. Chemical defenses are of greatest importance in many species, but thick, waxy cuticles may also retard processing in others (Bunn 1988; Canhoto and Gra9a 1995). The fact that L. styraciflua, A. rubrum, and U. americana had the highest initial macroinvertebrate abundance, especially chironomids, may very much reflect the fact that these three species had the highest processing rates of those tested. Conversely, low initial invertebrate abundances on Q. michauxii and M grandiflora may be indicative of the well-developed cuticles of these leaves. There was a very strong correlation between initial macroinvertebrate abundance and processing rate. The fact that recalcitrant compounds differed considerably between the two species, as did nutrient concentrations, suggests that none of these variables were the primary determinant of their processing rates. Declining chironomids and Tipula and increasing Microcylloepus sp. and Corydalus from December through May may be related more to biotic or abiotic factors than strictly nutritional quality or cuticle development of Q. michauxii and M grandiflora. Few taxa encountered in Possum Creek were shredders, and as such, neither quantity nor quality of leaf litter entering the stream should be the primary factor controlling abundance of individual taxa. However, slow-processed leaves result in slower generation of smaller CPOM fractions, FPOM, and DOM, and this could affect collector-gatherer and coUector-filterer communities that depend on such energy sources.

PAGE 85

75 Leaf packs and debris dams in Possum Creek appear to function more as habitat than an energy source, aUhough trapping of FPOM in these structures likely attracts those organisms that utilize this energy source. The extent of insect herbivory and/or tree health may alter the chemical constituents of leaves from expected levels (Bryant et al. 1983). As a result of a severe drought, some trees on the Possum Creek floodplain dropped their leaves during summer 2000, and it is likely that those that retained their leaves may have had altered leaf chemistry compared to non-drought years. Water availability can be a factor influencing not only individual trees, but also entire stands where soil water deficits occur. The effect of the drought on energy sources in Possum Creek and its floodplain is unknown, but both quality and quantity of leaf litter may have been affected. Minoletti (1994) found higher P and N levels in Viburnum acerifolium during a drought year, although Pilon (1996) found that drought did not alter nutrient concentrations in sugar maple {Acer saccharum) leaves. The quantity of leaf litter may also be affected by drought. Melaleuca quinquenervia on an Australian floodplain contributed 1 0% less leaf litter in a drought than a wet year (Green way 1994). Nitrogen content of leaves is among the most important chemical constituents determining leaf quality (Bunn 1988; Sinsabaugh and Linkins 1990). It provides the highest nutritive value of leaves for detritivores, thus initial concentration is positively correlated with processing rate (Cummins and Klug 1979). Growth rate and survivorship are heavily influenced by N intake, and those leaves high in N should be consumed preferentially (Anderson and Sedell 1979). Despite this, of the three fast-processed leaves, although both L. styraciflua and U. americana were high in N, ^4. rubrum had the

PAGE 86

76 least N of the five species used in this study. Thus, N does not appear to be of much predictive value for processing rates of taxa used in the study. Phenolic chemicals, as measured by the IVOMD procedure used in the present study, are relatively recalcitrant compounds that also dictate processing rates (Chew and Rodman 1979; Bryant et al. 1983; Bunn 1988). Many phenols are water-soluble and can either have little influence on processing rates (Suberkropp et al. 1976), or may indeed slow it (Edwards and Heath 1975; Barlocher and Oertli 1978; Bunn 1988). Phenols interfere with the breakdown of proteins (Walker 1975) and carbohydrates, including cellulose (Kirk et al. 1977; Zucker 1983), and may have antibiotic capability (Haslam 1989). Although not specifically measured in this study, tannins and lignins are the most commonly encountered recalcitrant compounds, and lignins are especially important in limiting leaf processing and may be more important than N or P in dictating processing rates (Melillo et al. 1982). These compoimds are likely the most important of the recalcitrant compounds measured by IVOMD employed in this study. However, although Q. michauxii and M. grandiflora were the slowest processed leaves in this study, Q. michauxii had the highest levels of recalcitrant compounds and M. grandiflora the least. As such, levels of recalcitrant compounds do not appear to serve as predictors for processing rates for taxa in this study. Processing rates are a reflection of many variables. The very well-developed cuticle of M grandiflora may be dictate its slow processing, but in the similarly slowprocessed Q. michauxii, high levels of recalcitrant compounds and a well-developed cuticle may be of greater importance. The faster-processed A. rubrum, L. styraciflua, and

PAGE 87

77 U. americana display variability in N, P, and recalcitrant compound levels, but all have less-developed cuticles than either Q. michauxii or M. grandiflora. As such, no single chemical constituent appears to govern the processing of these leaf species, but cuticle development may have a considerable role. Timing of litter fall Along with chemical content of leaves, timing of leaf fall is an important factor governing leaf processing. In the temperate zone, leaf fall is limited to a relatively brief autumnal period when bulk of allochthonous energy inputs occur (Cummins 1974). In the subtropics and tropics, leaf fall generally extends over a longer time period and often occurs at times other than during autumn. In north-central Florida, leaf fall begins in early October and can continue through early June (Chapter 2). Much of the total leaf fall in the Possum Creek floodplain falls from October through December, but additional leaf inputs from oaks (Quercus spp.) in January-February and M. grandiflora from midMarch through early June also add considerable carbon (Chapter 3). Differences in timing of leaf fall results in predictable sources of fresh litter in Possum Creek for over half the year. As a result of temporally staggered leaf fall, macroinvertebrates in Possum Creek have a somewhat constant energy supply entering the stream from October through May, that supplements the highly variable inputs from lateral deposition and flooding of the floodplain. As a result of lower temperatures and subsequent slower processing (Chergui and Pattee 1990), organic matter in temperate streams remains available for macroinvertebrates for a period of time far surpassing the period of leaf fall. Possum Creek displays higher temperatures than temperate streams, and if not for the extended

PAGE 88

78 period of leaf fall, it is likely that organic matter would be less available temporally for macroinvertebrate conmiunities. Macroinvertebrates The paucity of shredders in Possum Creek is a departure from most temperate systems. Most shredders are amphipods of insect larvae in the orders of Plecoptera and Trichoptera and the dipteran family Tipulidae. No Plecoptera were collected from Possum Creek, and the only trichopteran was Cheumatopsyche sp., a coUector-filterer. Tipulids and amphipods were collected and represent the only shredders in Possum Creek, although several chironomid taxa may be partially fit into this classification. A relative lack of shredders has been noted in other subtropical streams (Cuffney and Wallace 1987; Dudgeon 1994; McArthur et al. 1994), although this is not a universal phenomenon (Hauer et al. 1986). It is likely that shredding organisms are not important leaf processors in many subtropical streams. Rather, bacterial and fiingal processing and physical abrasion may be of greatest importance. Both of the dominant predators, Corydalus sp. and Argia sp., increased numerically over the course of the study, corresponding to decreased abundance of many macroinvertebrates during this same period. The fact that predators are important in controlling prey population levels is clear (Caswell 1978). Fish are generally the most important predators controlling prey populations in streams (Allan 1982), but macroinvertebrates may also play a large role in structuring macroinvertebrate prey communities (Obemdorfer et al. 1984; Wallace and Webster 1995). Despite the fact that both taxa are collector-gatherers, Microcylleopus sp., as opposed to Stenonema sp. was not regularly collected from leaf packs until late in the

PAGE 89

79 study. By the time Microcylleopus sp. abundance peaked in early May, in stream organic matter was low compared to earlier in the study. As a result, it does not appear that increased Microcylleopus sp. abundance was associated with leaf litter inputs. Instead, it is likely more closely related to life history patterns and/or abiotic factors. The decreased abundance of both Stenonema sp. and the coUector-filterer Cheumatopsyche sp. may be related to a reduction in leaf pack and debris dam habitat during this period (Chapter 3) as may have predation, life history, and abiotic factors. Collector-filterers feed primarily on FPOM and do not appear to be heavily influenced by leaf fall. The drop in Simulium sp. abundance from December through May likely occurred independently of changes in either quality or quantity of leaf fall. The increase in temperature during this period may have indirectly contributed to their decrease in abundance by increasing the metabolism of predators such as mosquitofish (Gambusia holbrooki), Corydalus sp., and Argia sp.. Simulium sp. appears relatively prone to predation, as they attach to the surface of in stream structure and are thus readily available to predators. Metamorphosis into adults during this period of time may also have contributed to their in stream decline, but no adults were seen. Changes in chironomid dominance from December-May were fairly consistent among leaf types. This suggests that leaf chemistry had less to do with chironomid community composition than other factors. This is particularly evident with chironomids, as one might conclude, based on changes in chironomid abundance for .4. rubrum, L styraciflua, and U. americana, that certain taxa such as CricotopuslOrthocladius and Rheotanytarsus became dominant as processing continued. However, chironomid dominance on M. grandiflora packs showed that although this

PAGE 90

80 conclusion may be valid for CricotopuslOrthocladius, it would appear to be incorrect for Rheotanytarsus, which maintains population levels throughout processing. The same can be said of Pofypedilum illinoense gp. and Corynoneura, both of which decreased in abundance during A. rubrum, L. styraciflua, and U. americam processing, but were present on M. grandiflora packs throughout its processing. Although leaf chemistry can play a considerable role in determining macroinvertebrate species composition of leaf packs (Anderson and Sedell 1979; Cummins and Klug 1979), at least in Possum Creek, it did not appear to be as important in dictating macroinvertebrate, specifically chironomid, dominance. This is particularly the case with those taxa that only depend indirectly on leaf litter as an energy source (e.g., collectors and predators). Life histories of some macroinvertebrates in Possum Creek may be related to peak seasonal inputs of leaf litter. However, separating the effects of seasonal litter accession, warming water temperatures, and predation is difficult within the design of this study. Furthermore, the fact that shredders are generally those macroinvertebrates whose life histories are most synchronized with leaf litter inputs (Cummins 1974), and yet are so uncommon in Possum Creek and other Northern Hemisphere subtropical streams, makes it difficult to assess the extent to which extended period of leaf fall affects their life histories. The Northern Hemisphere subtropics likely occupy a niche different from those of both northern temperate and both tropical and temperate of the Southern Hemisphere streams. Northern temperate streams are limited to a brief period of autumnal leaf fall that occurs just prior to the coldest period of the year (Fisher and Likens 1973). Conversely, in Southem Hemisphere streams, leaf fall occurs over an extended period of

PAGE 91

81 time, but during the wannest time of the year (Lake and Barmiita 1986; Stewart and Davies 1990). Possum Creek is intermediate in that leaf fall begins just prior to the coldest time of year, yet the period of leaf fall extends far beyond autumn. The fact that macroinvertebrate life histories in north temperate streams are synchronized to seasonal leaf inputs (Cummins 1974), Southern Hemisphere streams are not (Campbell 1994), and the possibility that Possum Creek macroinvertebrate life histories are only related to the habitat afforded by leaf aggregations further reinforces the uniqueness of Northern Hemisphere subtropical streams.

PAGE 92

CHAPTER 5 CONCLUSIONS Organic matter inputs in subtropical north-central Florida are available from September through early Jime. These inputs are reflected in in stream storage, predominantly in debris dams and leaf packs. Sequential additions of organic matter resulting from trees with different abscission periods not only result in an extended period of leaf fall as compared to the Northern Hemisphere temperate zone, but in fresh organic matter being made available for stream organisms. Although in Northern Hemisphere temperate streams, organic matter is processed by both shredding macroinvertebrates and microbes, shredders in Possum Creek were uncommon, suggesting that they played a lesser role in processing as compared to those in Northern Hemisphere temperate streams. This research may have implications for management and restoration of riparian areas in north-central Florida. Both natural and anthropogenic disturbances have altered riparian vegetation in many areas. Severe fires can alter the quantity of inputs to sfreams (Minshall et al. 1989), and vegetation succession after such an event may vary the quality and timing of leaf inputs, consequently affecting invertebrates (Minshall et al. 1995). Logging, particularly clear-cutting, is similar to fires in that an initial decrease of leaf quantity is observed (Fisher and Likens 1973). Subsequent succession can have implications for leaf litter quality and accession (Griffith and Perry 1991), which in turn can affect invertebrate communities (Benfield et al. 1991). Combining knowledge of 82

PAGE 93

83 what species are likely to colonize a riparian system after disturbance, from pioneer species to climax conummities, with the information in the present studies, will enable land managers and scientists to determine temporal availability of organic matter based on tree species present. Restoration of degraded riparian habitats is increasing (Kondolf and Micheli 1995). In many cases, these habitats have little to no woody vegetation, or the vegetation may be undesirable. Knowledge of tree associations can increase the probability for long-term success by planting trees from each of the four groupings classified in this study. By maximizing nutrient content, levels of recalcitrant and inhibitory compounds, and timing of leaf fall, organic matter can be available for extended periods of the year and more closely resemble communities in undisturbed areas.

PAGE 94

REFERENCES Abelho, M., and M. A. S. Gra9a. 1996. Effects of eucalyptus afforestation on leaf litter dynamics and macroinvertebrate community structure of streams in central Portugal. Hydrobiologia 324: 195-204. Allan, J. D. 1982. The effects of reduction in trout density on the invertebrate community of a mountain stream. £co/ogv 63:1444-1455. Anderson, N. H., and J. R. Sedell. 1979. Detritus processing by macroinvertebrates in stream ecosystems. Rev. Entom. 24: 351-377. Angradi, T. R. 1996. Inter-habitat variation in benthic community structure, function, and organic matter storage in three Appalachian headwater streams. J. N. Am. Benthol. Soc. 15:42-63. Barlocher, F., and J. J. Oertli. 1978. Inhibitors of aquatic hyphomycetes in dead conifer needles. Mycologia 70:964-974. Benfield, E. F. 1997. Comparison of litterfall input to streams. J. N. Am. Benthol. Soc. 16: 104-108. Benfield, E. F., D. S. Jones, and M. F. Patterson. 1977. Leaf pack processing in a pastureland stream. Oikos 29:99-103. Benfield, E. F., J. R. Webster, S. W. Golladay, G. T. Peters, and B. M. Stout. 1991. Effects of forest disturbance on leaf breakdown in southern Appalachian streams. Verh. Internal. Verein. Limnol. 24: 1687-1690. Benke, A. C. and J. L. Meyer. 1988. Structure and function of a blackwater river in the Southeastem U.S.A. Verh. Internal. Verein. Limnol. 23:1209-1218. Benstead, J. P. 1996. Macroinvertebrates and the processing of leaf litter in a tropical stream. Biolropica 28: 367-375. Bilby, R. E., and G. E. Likens. 1979. Effect of hydrologic fluctuations on the transport of fine particulate organic carbon in a small stream. Limnol. Oceanogr. 24: 69-75. Bilby, R. E. 1981. Role of organic debris dams in regulating the export of dissolved and particulate matter from a forested watershed. Ecology 62: 1234-1243. 84

PAGE 95

85 Bilby, R. E. 1980. Importance of organic debris dams in the structure and function of stream ecosystems. Ecology 61 : 1 107-1 113. Boulton, A. J., and P. I. Boon. 1991. A review of methodology used to measiire leaf litter decomposition in lotic environments: Time to turn over an old leaf? Aus. J. Mar. Freshwat. Res. 42: 1-43. Bretschko, G., and M. Leichtfried. 1988. Distribution of organic matter and faima in a second order, alpine gravel stream (Ritrodat-Lunz study area, Austria). Verh. Internat. Verein. Limnol. 23: 1333-1339. Brinson, M. M., A. E. Lugo, and S. Brown. 1981. Primary productivity, decomposition and consumer activity in freshwater wetlands. Ann. Rev. Ecol. Syst. 12: 123-161. Bruce, K. A., G. N. Cameron, and P. A. Hacombe. 1995. Initiation of a new woodland type on the Texas Coastal Prairie by the Chinese tallow tree (Sapium sebiferum (L.) Roxb.). Bull. TorreyBot. Club 122: 215-225. Bryant, J. P., F. S. Chapin, III, and D. R. Klein. 1983. Carbon/nutrient balance of boreal plants in relation to vertebrate herbivory. Oikos 40:357-368. Bunn, S. E. 1988. Processing of leaf litter in two northern jarrah forest streams, Westem Australia: 2. The role of macroinvertebrates and the influence of soluble polyphenols and inorganic sediment. Hydrobiologia 162:21 1-223. Campbell, I. C. 1994. Ecology of Australian detritus-based streams A note in response to Gra9a. Limnologica 24:291-293. Campbell, 1. C, and L. Fuchshuber. 1994. Amount, composition and seasonality of terrestrial litter accession to an Australian cool temperate rainforest stream. Arch. Hydrobiol. 130:499-512. Canhoto, C, and M. A. S. Gra9a. 1995. Food value of introduced eucalypt leaves for a Mediterranean stream detritivore: Tipula lateralis. Freshwat. Biol. 34:209-214. Caswell, H. 1978. Predator-mediated coexistence: A non-equilibrium model. Amer. Nat. 112:127-154. Chergui, H., and E. Pattee. 1990. The influence of season on the breakdown of submerged leaves. .4rc/2. Hydrobiol. 120:1-12. Chew, F. S., and J. E. Rodman. 1979. Plant resources for chemical defense. Pages 271307 In G. A. Rosenthal and D. H. Janzen (eds.) Herbivores: their interaction with secondary plant metabolites. New York, Academic Press.

PAGE 96

86 Conners, M. E., and R. J. Naiman. 1984. Particulate allochthonous inputs: Relationships with stream size in an undisturbed watershed. Can. J. Fish. Aquat. Sci. 41: 14731484. Cuffhey, T. F. 1988. Input, movement and exchange of organic matter within a subtropical coastal blackwater river-floodplain system. Freshwat. Biol. 19: 305320. Cuffhey, T. F., and B. J. Wallace. 1987. Leaf litter processing in coastal plain streams and floodplains of southeastern Georgia, U.S.A. Arch. Hydrobiol, Suppl. 76: 124. Cuffhey, T. F., J. B. Wallace, and G. J. Lugthart. 1 990. Experimental evidence quantifying the role of benthic invertebrates in organic matter dynamics of headwater streams. Freshwat. Biol. 23: 281-299. Cummins, K. W. 1974. Structure and function of stream ecosystems. Bioscience 24:631641. Cummins, K. W., and M. J. Klug. 1979. Feeding ecology of stream invertebrates. Annu. Rev. Ecol. Syst. 10:147-172. Cummins, K.W., M.J. Klug, G.M. Ward, G.L. Spingler, R.W. Speaker, R.W. Ovink, D.C. Mahan, and R.C. Petersen. 1981. Trends in particulate organic matter fluxes, community processes and macroinvertebrate functional firoups along a Great Lakes drainage basin river continuum. Verh. Internat. Verein. Limnol. 21: 841849. Cummins, K. W., R. C. Petersen, F. O. Howard, J. C. Wuycheck, and V. I. Hoh. 1973. The utilization of leaf litter by stream detritivores. Ecology 54: 336-345. Cummins, K. W., J. R. Sedell, F. J. Swanson, G. W. Minshall, S. G. Fisher, C. E. Gushing, R. C. Petersen, and R. L. Vannote. 1983. Organic matter budgets for stream ecosystems: problems in their evaluation. Pages 299-354, In J. R. Barnes and G. W. Minshall (eds). Stream Ecology: Application and Testing of General Ecological Theory. Plenum, New York. Cummins, K. W., M. A. Wilzbach, D. M. Gates, J. B. Perry, and W. B. Taliaferro. 1989. Shredders and riparian vegetation. Bioscience 39: 24-30. Day, F. P., Jr. 1982. Litter decomposition rates in the seasonally flooded Great Dismal Swamp. Ecology 63: 670-678. Dudgeon, D. 1982. An investigation of physical and biological processing of two species of leaf litter in Tai Po Kau Forest Stream, New Territories, Hong Kong. Arch. Hydrobiol. 96:1-32.

PAGE 97

87 Dudgeon, D. 1989. The influence of riparian vegetation on the functional organization of four Hong Kong stream communities. Hydrobiologia 179: 183-194. Dudgeon, D. 1994. The influence of riparian vegetation on macroinvertebrate community structure and functional organization in six New Guinea streams. Hydrobiologia 294:65-85. Edwards, C. A., and G. W. Heath. 1975. Studies in leaf litter breakdown. III. The influence of leaf age. Pedobiologia 15:348-354. Elder, J. F., and D. J. Cairns. 1982. Production and decomposition of forest litter fall on the Apalachicola River flood plain, Florida. Alexandria, VA, U.S. Geological Survey. Elwood, J. W., J. D. Newbold, A. F. Trimble, and R. W. Stark. 1981. The limiting role of phosphorus in a woodland stream ecosystem: Effects of P enrichment on leaf decomposition and primary producers. Ecology 62: 146-158. Epler, J.H. 1995. Identification Manual for the Larval Chironomidae (Diptera) of Florida. Florida Department of Environmental Protection, Tallahassee, FL. Fisher, S. G., and G. E. Likens. 1973. Energy flow in Bear Brook, New Hampshire: an integrative approach to stream ecosystem metabolism. Ecol. Monogr. 43:421-439. Fisher, S. G., L. J. Gray, N. B. Grimm, and D. E. Busch. 1982. Temporal succession in a desert stream ecosystem following flash flooding. Ecological Monographs 52: 93110. Gallaher, R. N., C. O. Weldon, and J. G. Futral. 1975. An aluminum block digester for plant and soil analysis. Soil Sci. Soc. Amer. Proc. 39:803-806. Gallardo, A., and J. Merino. 1999. Control of leaf litter decomposition rate in a Mediterranean shrubland as indicated by N, P and lignin concentrations. Pedobiologia 43: 64-72. Gessner, M. W., and J. Schwoerbel. 1989. Leaching kinetics of fresh leaf-litter with implications for the current concept of leaf-processing in streams. Arch. Hydrobiol. 115: 81-90. Golladay, S. W., J. R. Webster, and E. F. Benfield. 1989. Changes in stream benthic organic matter following watershed disturbance. Holarc. Ecol. 12: 96-105. Gonzalez, E., and J. Pozo. 1996. Longitudinal and temporal patterns of benthic coarse particulate organic matter in the Aguera stream (northern Spain). Aquat. Sci. 58: 355-366.

PAGE 98

88 Greenway, M. 1994. Litter accession and accumulation in a Melaleuca quinquenervia (Cav.) S.T. Blake wetland in south-eastern Queensland. Aust J. Mar Freshwat. Res. 45:1509-1519. Griffith, M. B., and S. A. Perry. 1991. Leaf pack processing in two Appalachian mountain streams draining catchments with different management histories. Hydrobiologia 220: 247-254. Griffith, M. B., and S. A. Perry. 1993. Colonization and processing of leaf litter by macroinvertebrate shredders in streams of contrasting pH. Freshwat. Biol. 30:93103. Gurtz, M. E., G. R. Marzolf, K. T. Killingbeck, D. L. Smith, and J. V. McArthur. 1988. Hydrologic and riparian influences on the import and storage of coarse particulate organic matter in a prairie stream. Can. J. Fish. Aquat. Sci. 45: 655-665. Haefher, J. D., and J. B. Wallace. 1981. Shifts in aquatic insect populations in a firstorder southern Appalachian stream following a decade of old field succession. Can. J. Fish. Aquat. Sci. 38: 353-359. Hambleton, L. G. 1977. Semiautomated method for simultaneous determination of phosphorus, calcium and crude protein in animal feeds. J. Assoc. Off. Anal. Chem. 60:845-852. Haslam, E. 1989. Plant Polyphenols. Cambridge, Cambridge University Press. Hauer, F. R., N. L. Poff, and P. L. Firth. 1986. Leaf litter decomposition across broad thermal gradients in southeastern coastal plain streams and swamps. J. Freshwat. Ecol. 3:545-552. Hill, A. R., and A. M. Brooks. 1996. Coarse particulate organic matter inputs to a headwater swamp stream. Arch. Hydrobiol. 137: 25-38. Howarth, R. W., and S. G. Fisher. 1976. Carbon, nitrogen, and phosphorus dynamics during leaf decay in nutrient enriched stream microecosystems. Freshwat. Biol. 6:221-228. Hynes, H. B. N. 1975. The stream and its valley. Verh. Internat. Verein. Limnol 19: 1-15. Irons, J. G., and M. W. Oswood. 1997. Organic matter dynamics in 3 subarctic stt-eam.s of interior Alaska, USA. J. N. Am. Benthol. Soc. 16: 23-28. Jones, J. B. 1997. Benthic organic matter storage in streams: influene of dettital import and export, retention mechanisms, and climate. J. N. Am. Benthol. Soc. 16: 109119.

PAGE 99

89 Jones, J. B., and L. A. Smock. 1991 . Transport and retention of particulate organic matter in two low-gradient headwater streams. J. N. Am. Benthol. Soc. 47: 588594. Kaushik, N. K., and H. B. N. Hynes. 1971. The fate of dead leaves that fall into streams. Arch. Hydrobiol. 68:465-515. King, J. M., J. A. Day, B. R. Davies, and M. P. Henshall-Howard. 1987. Particulate organic matter in a mountain stream in the south-western cape, South Africa. Hydrohiologia 154: 165-187. Kirk, T. K., W. J. Connors, and J. G. Zeikus. 1977. Advances in imderstanding the microbiological degradation of lignin In F. A. Loewus, and V. C. Runeckles, eds. Recent Advances in Phytochemistry. Vol. 11. The Structure, Biosysnthesis and Degradation of Wood. New York, Plenum. Kondolf, G. M., and E. R. Micheli. 1995. Evaluating stream restoration projects. Environ. Manage. 19: 1-15. Lake, P. S., and L. A. Barmuta. 1986. Stream benthic communities: Persistent presumptions and current speculations. Monogr. Biol. 61: 263-276. Lester, P. J.,. S. F. Mitchell, and D. Scott. 1994. Effects of riparian willow trees {Salix fragilis) on macroinvertebrate densities in two small central Otago, New Zealand, streams. N. Z. J. Mar. Freshwat. Res. 28: 267-276. Liu, S., H. Riekerk, and H. L. Gholz. 1996. Leaf litterfall, leaf area index, and radiation transmittance in cypress wetlands and slash pine plantations in north-central Florida. Wetl. Ecol. Manage. 4: 251 -21 \. Lockhart, C. S., D. F. Austin, W. E. Jones, and L. M. Downey. 1997. The invasion of carrotwood (Cupaniopsis anacardioides) in natural areas. Final study report. Pages 24. Habitat Specialists, INC. for Florida Department of Environmental Protection, Tallahassee, FL. Lugo, A. E., and C. P. Zucca. 1983. Comparison of litter fall and turnover in two Florida ecosystems. Fla. Sci. 46: 101-111. Lugthart, G. J., and J. B. Wallace. 1992. Effects of disturbance on benthic functional structure and production in mountain .streams. J. N. Am. Benthol. Soc. 11: 138164. Lyons, J., S. W. Trimble, and L. K. Paine. 2000. Grass versus trees: Managing riparian areas to benefit streams of central North America. J. Am. Water Res. Assoc. 36: 919-930.

PAGE 100

90 Maity, S. K., and V. C. Joy. 1999. Impact of antinutritional chemical compounds of leaf litter on detritivore soil arthropod fauna. J. Ecohiol. 11: 193-202. McArthur, J. V., J. M. Aho, R. B. Rader, and G. L. Mills. 1994. Interspecific leaf interactions during decomposition in aquatic and floodplain ecosystems. J. N. Am. Benthol. Soc. 13:57-67. McClaugherty, C. A., J. Pastor, J. D. Aber, and J. M. Melillo. 1985. Forest litter decomposition in relation to soil nitrogen dynamics and litter quality. Ecology 66: 266-275. McDowell, W.H., and C.E. Asbury. 1994. Export of carbon, nitrogen, and major ions from three tropical montane watersheds. Limnol. Oceanogr. 39:111-125. Melillo, J. M., J. D. Aber, and J. F. Muratore. 1982. Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 63:621-626. Merritt, R. W., and D. L. Lawson. 1992. The role of leaf litter macromvertebrates in stream-floodplain dynamics. Hydrobiologia 248: 65-77. Merritt, R. W., W. Wuerthele, and D. L. Lawson. 1984. The effect of leaf conditioning on the timing of litter processing on a Michigan woodland floodplain. Can. J. Zool. 62:179-182. Metzler, G. M., and L. A. Smock. 1990. Storage and dynamics of subsurface detritus in a sand-bottomed stream. Can. J. Fish. Aquat. Sci. 47: 588-594. Minoletti, M. L., and R. E. J. Boemer. 1994. Drought and site fertility effects on foliar nitrogen and phosphorus dynamics and nutrient resorption by the forest understory shrub Viburnum acerifolium L. Am. Midi. Nat. 131:109-1 19. Minshall, G. W. 1996. Organic matter budgets. Pages 591-605 in F. R. Hauer and R. H. Lamberti, eds. Methods in Stream Ecology. Academic Press, San Diego, CA. Minshall, G. W., R. C. Petersen, K. W. Cummins, T. L. Bott, J. R. Sedell, C. E. Gushing, and R.L. Vannote. 1983. Interbiome comparison of stream ecosystem dynamics. Ecol. Monogr. 53: 1-25. Minshall, G. W., K. W. Cummins, R. C. Petersen, C. E. Cushing, D. A. Bruns, J. R. Sedell, and R. L. Vannote. 1985. Developments in stream ecosystem theory. Can. J. Fish. Aquat. Sci. 42: 1045-1055. Minshall, G. W., J. T. Brock, and J. D. Varley. 1989. Wildfires and Yellowstone's stream ecosystems. Bioscience 39: 707-715.

PAGE 101

91 Minshall, G. W., C. T. Robinson, T. V. Royer, and S. R. Rushforth. 1995. Benthic conununity structure in two adjacent streams in Yellowstone National Park five years after the 1988 wildfires. Great Basin Naturalist 55: 193-200. Moore, J. E., and G. O. Mott. 1974. Recovery of residual organic matter fi-om in vitro digestion of forages. J. Dairy Sci. 57:1258-1259. Mulholland, P. J. 1981. Organic matter flow in a swamp-stream ecosystem. Ecol. Monogr. 51:307-322. Mulholland, P. J., J. D. Newbold, J. W. Elwood, L. A. Ferren, and J. R. Webster. 1985. Phosphorus spiralling in a woodland stream: seasonal variations. Ecology. 66: 1012-1023. Naiman, R. J. 1982. Charactristics of sedimant and organic carbon export fi:om boreal forest watersheds. Can. J. Fish. Aquat. Sci. 39: 1699-1718. Nelson, G. 1994. The trees of Florida. Pineapple Press, Saratota, FL. Nelson, G. 1996. The shrubs and woody vines of Florida. Pineapple Press, Sarasota, FL. Obemdorfer, R. Y., J. V. McArthur, and J. R. Barnes. 1984. The effect of invertebrate predators on leaf litter processing in an alpine stream. Ecology 65:1325-1331. Petersen, R. C, and K. W. Cummins. 1974. Leaf processing in a woodland stream. Freshwat. Biol. 4:345-368. Peterson, D. L., and G. L. Rolfe. 1982. Nutrient dynamics and decomposition of litterfall in floodplain and upland forest of central Illinois. For. Sci. 28: 667-681. Pilon, C. E., B. Cote, and J. W. Fyles. 1996. Effect of an artificially induced drought on leaf peroxidase activity, mineral nutrition and growth of sugar maple. Plant Soil 179:151-158. Pozo, J., A. Basaguren, A. Elosegui, J. Molinero, E. Fabre, and E. Chauvet. 1998. Afforestation with Eucalyptus globulus and leaf litter decomposition in streams of northern Spain. Hydrobiologia 373: 101-109. Ractliffe, S. G., B. R. Davies, B. A. Stewart, and C. D. Snaddon. 1995. The influence of discharge on entrainment of bank litter in a headwater stream. Arch. Hydrobiol 134: 103-117. Reice, S. R. 1974. Environmental patchiness and the breakdown of leaf litter in a woodland stream. Ecology 55: 1271-1282. Rounick, J. S., and M. J. Winterboum. 1983. Leaf processing in two contrasting beech

PAGE 102

92 forest streams: Effects of physical and biotic factors on litter breakdown. Arch. Hydrobiol. 96: 448-474. Singh, J. 1984. Effects of temperature, rainfall and soil moisture on soil respiration and litter decomposition in a sub-tropical humid forest ecosystem. Acta Bot. Indica 12: 167-173. Sinsabaugh, R. L., and A. E. Linkins. 1990. Enzymic and chemical analysis of particulate organic matter from a boreal river. Freshwat. Biol. 23:301-309. Smith, J. J., and P. S. Lake. 1993. The breakdown of buried and surface-placed leaf litter in an upland stream. Hydrobiologia 271: 141-148. Smock, L. A. 1990. Spatial and temporal variation in organic matter storage in lowgradient, headwater streams, ^rc/z. Hydrobiol. 118: 169-184. Smock, L. A., G. M. Metzler, and J. E. Gladden. 1989. Role of debris dams in the structure and functioning of low-gradient headwater streams. Ecology 70: 764775. Smock, L. A., and C. E. Roeding. 1986. The trophic basis of production of the macroinvertebrate community of a southeastern U.S.A. blackwater stream. Holarc. Ecol. 9: 165-174. * Songwe, N. C, F. E. Fasehun, and D. U. U. Okali. 1988. Litterfall and productivity in a tropical rain forest, Southern Bakundu Forest Reserve, Cameroon. J. Trop. Ecol 4: 25-37. Spain, A. V. 1984. Litterfall and the standing crop of litter in three tropical Australian rainforests. J. Ecol. 72: 947-961. SPSS. 1998. SPSS for Windows v. 9.0. SPSS hic, Chicago, IL. Stewart, B. A., and B. R. Davies. 1990. Allochthonous input and retention in a small mountain stream, South Africa. Hydrobiologia 202: 135-146. Strommer, J. L., and L. A. Smock. 1989. Vertical distribution and abundance of invertebrates within the sandy substrate of a low-gradient headwater stream. Freshwat. Biol. 22: 263-274. Suberkropp, K., T. L. Arsuffi, and J. P. Anderson. 1983. Comparison of degradative ability, enzymatic activity, and palatability of aquatic hyphomycetes grown on leaf litter. ^/>p/. Environ. Microbiol. 46:237-244.

PAGE 103

93 Suberkropp, K., G. L. Godshalk, and M. J. Klug. 1976. Changes in the chemical composition of leaves during processing in a woodland stream. Ecology 57:720727. Tobe, J. D., K. C. Burks, R. W. Cantrell, M. A. Garland, M. E. Sweeley, D. W. Hall, P. Wallace, G. Anglin, G. Nelson, J. R. Cooper, D. Bickner, K. Gilbert, N. Ayrmond, K. Greenwood, andN. Raymond. 1998. Florida Wetland Plants, An Identification Manual. Florida Department of Environmental Protection, Tallahassee, FL. Triska, F. J., J.R. Sedell, and S.V. Gregory. 1982. Coniferous forest streams. Pages 292332 In R.L. Edmonds (ed.) Analysis of coniferous forest ecosystems in the western United States. Hutchinson Ross, Stroudsburg, PA. Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Gushing. 1980. The River Continuum Concept. Can. J. Fish. Aquat Sci. 37: 130-137. Walker, J. R. L. 1975. The Biology of Plant Phenolics. London, Arnold. Wallace, J. B., and J. R. Webster. 1995. The role of macroinvertebrates in stream ecosystem fimction. Ann. Rev.Entomol. 40:1 15-139. Wallace, J. B., M. R. Whiles, S. Eggert, T. F. Cuffney, G. J. Lugthart, and K. Chung. 1995. Long-term dynamics of coarse particulate organic matter in three Appalachian Mountain streams. J. N. Am. Benthol. Soc. 14: 217-232. Webster, J. R., and E. F. Benfield. 1986. Vascular plants breakdown in freshwater ecosystems. Ann. Rev. Ecol. Systemat. 17:567-594. Webster, J. R., S. W. Golladay, E. F. Benfield, D. J. D'Angelo, and G. T. Peters. 1990. Effects of forest disturbance on particulate organic matter budgets of small streams. J. A'^^ Am. Benthol. Soc. 9: 120-140. Webster, J. R., and J. L. Meyer. 1997. Organic matter budgets for streams: a synthesis. J. N. Am. Benthol. Soc. 16: 141-161. Weigelhofer, G., and J. A. Waringer. 1994. Allochthonous input of coarse particulate organic matter (CPOM) in a first to fourth order Austrian forest stream. Verhand. Int. Rev. Gesamt Hydrobiol. 79: 461-471. Woodall, S. L. 1982. Herbicide tests for control of Brazilian pepper andmelaleuca in Florida. 10 pp. USDA Forest Service, Washington D.C. Zucker, W. V. 1983. Tannins: Does structtire determine function? An ecological perspective, ^mer. Nat 121:335-365.

PAGE 104

APPENDICES Appendix 1 . Procedures for determination of nitrogen and phosphorus in leaf tissue. 1. Mix ground sample thoroughly. 2. Weigh out 0.25 g of sample and place into 75 ml tubes. 3. Add 1 .5 g of catalyst (9:1 K2S04-CuS04) to each tube. 4. Place tubes into digestion block in fume hood. 5. Dispense 4.5 ml H2SO4 into each tube and vortex until mixed thoroughly. 6. Add 1 ml of H2O2 to each tube, wait a moment for foaming to subside, and add a second ml. 7. Add de-ionized water to each tube and fill until level reaches the point where the tube narrows. 8. Place tubes in metal rack and cap with rubber stopper. 9. Allow tubes to cool for 1 5 minutes in water. 10. Bring volume of the tube up to the 75 ml mark with de-ionized water. 1 1 . Recap tubes, place rack on mixer, and invert 5 times to mix. 12. Uncap tubes and pour into numbered scintillation vial. 1 3 . For nitrogen determination, add ammonia-salicylate complex and alka) ine medium (pH 12.8-13.0) into each vial, place into autoanalyzer, and measure absorbance at 660 nm. 94

PAGE 105

95 For phosphorus determination, add molybdovanadophosphate comple acidic medium into each vial, place into auto analyzer, and measiire absorbance at 420 nm.

PAGE 106

it 96 Appendix 2. Procedures for determination of nitrogen and phospriorus in leaf tissue. 1 . Place 0.5 g sample into a numbered centrifuge tube. 2. Prepare buffer: 9.8gNaHC03 7.0 g Na2HP04-7tfO (3.71 g anhydrous) 0.57 g KCl i 0.47gNaCl 0.12gMgSO'*-7H20 3. Add 1 ml 4% CaCb per liter of buffer to create McDougall's artificial saliva and place in 39° C water bath. 4. Bubble CO2 through artificial saliva. 5. Add 2 ml distilled water to all centrifuge tubes and wet all sample using test tube mixer. 6. Check to make sure pH of artificial saliva is 6-8-7,0. 7. Add one part rumen fluid inoculum to four parts artificial saliva and allow to mix for 10 minutes through bubbling with CO2. 8. Add 50 ml of inoculum-saliva medium to each centrifuge tube. 9. Place tubs into water bath and sparge samples by passing C02 over each tube for 15 seconds and quickly stopper tubes with Bunsen valve. 1 0. Transfer tubes to incubator at 39° C. 1 1 . After one-hour incubation, swiri tube contents. Repeat twice the first day and three times the second day. 1 2. After 48 hours incubation, remove stoppers and wash any particles adhering to the stoppers into the tube with distilled water.

PAGE 107

13. Add 1 ml 20% HCI and swirl tubes. Add another 1 ml 20% HCl and swirl. Add 4 ml 40% HCl and swirl. 14. Check to see the pH is between 1 -2. 1 5. Add 2 ml 5% pepsin, swirl thoroughly, replace stoppers, and return tubes to incubator at 39° C. Swirl twice the first day and three times the second day. 1 6. Prepare gooch crucibles by forming a glass wool mat from 3-4 pieces of glass wool. 1 7. After 46 hours pepsin digestion, transfer contents of centrifuge tubes to the gooch crucibles using the crucible holder and filter with low filter. Wash all residue from tube into crucible with hot distilled water. Wash gooch crucibles three tunes with hot distilled water. 1 8. Place gooch crucibles in drying oven at 1 05° C and dry to constant weight (overnight), cool in desiccator, and weigh. 1 9. Place crucibles in muffle fiimace at 500° C for 3 hours, cool in desiccator and weigh.

PAGE 108

BIOGRAPHICAL SKETCH Christopher Richard Roberts was bom in Willoughby Hills, Ohio, on August 23, 1972. Soon thereafter, he and his family moved to his parents' hometown of Youngstown, Ohio. Chris graduated from Boardman High School in 1990 and later that year began his studies at The Ohio State University. At Ohio State, Chris had a double major in in Zoology and Natural Resources, with a specialization in fisheries management and gained the basic knowledge of aquatic ecology that would serve as the foundation for his later work. One week after obtaining his degrees from Ohio State in June 1996, Chris married Stephanie Jane Bosh and two months later, both moved to Gainesville, Florida. The University of Florida Department of Environmental Engineering Sciences accepted Chris for graduate studies beginning in the Fall of 1996. While at the university, he worked with the Center for Wetlands on a variety of research topics dealing with benthic macroinvertebrates, fish, and zooplankton, and their intereactions with the environment. In May 2000, Chris received his Master of Science degree fi-om the University of Florida Department of Environmental Engineering Sciences. Upon receiving his degree, he worked on a variety of research projects in both streams and wetlands dealing with organic matter and macroinvertebrates. 98

PAGE 109

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy Thomas L. Crisma Professor of Environmental Engineering Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fiilly adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. William R. Wise Associate Professor of Environmental Engineering Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy Frank G. Nordlie Professor Emeritus of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fiilly adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy iren'j. Cfiapman Associate Professor of Zoology

PAGE 110

This dissertation was submitted to the Graduate Faculty of the College of Engineering and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May 2002 A Pramod P. Khargonekar / Dean, College of Engineering Winfred M. Phillips Dean, Graduate School