WWWMWWWWIll WLIHHI 1/ /’

we 3 1293 107075

UBRARY
Michigan State ‘
University

 

 

 

 

This is to certify that the
dissertation entitled
THE ROLE OF EMERGENT MACROPHYTES TO

NITROGEN AND PHOSPHORUS CYCLING IN A
GREAT LAKES MARSH

presented by

James C. Kelley
has been accepted towards fulfillment

of the requirements for

Ph . D . degree in ECOLOGY

 

Major professor

Date 2/39/35

MS U is an Affirmative Action/Equal Opportunity Institution 0-1277!

 

MSU

LIBRARIES
g3;—

 

 

f

RETURNING MATERIAL§5
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

 

j“; 2 § via—3'7

(‘3’ -' " ~
. win—.4 ‘19:“):
5‘33““ i". b \

r A

, qy r' find
ig’ffl'n

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

THE ROLE OF EMERGENT MACROPHYTES TO NITROGEN

AND PHOSPHORUS CYCLING IN A GREAT LAKES MARSH

BY

James C. Kelley

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Fisheries and Wildlife

1985

ABSTRACT

THE ROLE OF EMERGENT MACROPHYTES TO NITROGEN
AND PHOSPHORUS CYCLING IN A GREAT LAKES MARSH

BY
James C. Kelley

Marshes may function by regulating nutrient flux
between many aquatic and terrestrial ecosystems. Few
studies have quantified the cycling and storage of nitrogen
and phosphorus in freshwater marshes. Plant nutrient
uptake and litter mineralization are two of the primary
mechanisms a wetland might control nutrient cycles and
water quality. The importance of these processes relative
to the total flux of nutrients through a wetland is rarely
known. This study quantified the flux and retention of N
and P through a marsh, and evaluated the role of emergent
plants in controlling nutrient retention.

Hydrologic, non-gaseous N, and P mass balance data
were developed from studies conducted at Pentwater Marsh,
Pentwater Twp., MI USA (430 46' N, 860 W). On an
annual basis, about 10% of the inorganic-N inputs were
removed by the marsh. An export of organic-N (27.6%)
occurred, but organic-N export was not significantly
greater than inorganic-N retention. Total-P and reactive-P
were removed by the marsh (30.4% and 13.8% respectively).

For both N and P, the greatest retention occurred during

the growing season, suggesting biological control of
removal processes.
Biomass production in 6 emergent and wet meadow

communities (Sparganium eurycarpum, Scirpus validus, Typha

 

latifolia, Calamagrostis canadensis, Carex stricta, and

 

Carex aquatilis) was studied using harvest techniques.

 

Tissue N and P concentrations and the biomass data were
used to evaluate the role of vegetation in marsh nutrient

cycles. Maximum shoot N content ranged from 4.8 g m"2 to

2, and P content from 0.6 g m-2 to 2.0 g m-z.

7.9 g m—
Average production and nutrient uptake rates were lower in
the meadow than the emergent zone. Annually, only 2% of
the N and 15% of the P inputs are accounted for in the
August shoot biomass. However, shoot biomass could account
for 25% of the inorganic-N and 59% of the total-P removed
from the rivers.

N and P mineralization rates were greatest in the
emergent zone. In emergent areas, as much as 88% of N and
P in litter may be mineralized annually. Nutrient

mineralization in the meadow zone can be only 20% that of

the emergent zone.

ACKNOWLEDGEMENTS

This research was supported in part by a grant to Dr.
Thomas Burton from the Michigan Sea Grant Program (Project
R/CW—S) of NCAA, US Department of Commerce. My thanks are
extended to Dr. Niles Kevern and the Department of
Fisheries and Wildlife for financial assistance.

I wish to express sincere appreciation to Dr. Thomas
Burton for his patience and guidance during this study.
Many thanks are extended to the members of my guidance
committee (Dr. Peter Murphy, Dr. Stephen Stephenson, Dr.
Cal McNabb, and Dr. Ted Baterson) for sharing their time
and expertise. I am grateful Dr. Charles Liston for the
use of his equipment and facilities. I also thank William
Enslin and Mat Krogulecki of the Center for Remote Sensing
for their generous assistance with the aerial photography.

I am very much indebted to the understanding and
assistance of many friends and colleagues. In particular,
I want to extend my sincere thanks to Dr. Karl Ulrich and
family, Martha Case, and Dr. Christopher Carmichael for
their special contributions. Finally, I express gratitude
and love to Lucile McCook for her generous assistance,

cooperation and love.

ii

TABLE OF CONTENTS

page
LIST OF TABLES ......................................... v
LIST OF FIGURES. ....................... . .............. viii

I 0 INTRODUCTION 0 O O O O O O O O O O O I I O O O O O O O O O C O O O I C O O O I O O O O O 0 1
OBJECTIVES O O O O O O ....... O O O I O O O O O O O O O O O O O O O O O O I 0 O O O 2
DESCRIPTION OF THE STUDY SITE..................... 3

 

II. LITTER DECOMPOSITION AND NUTRIENT MINERALIZATION.. 9
INTRODUCTION............. ........ ................. 9
OBJECTIVES........................................ 10
METHODS........................................... 11
RESULTS........................................... 16

Weightloss................................... 17
Nitrogen changes............................. 23
Phosphorus changes........................... 30
1980-1982 Studies............................ 34
DISCUSSION........................................ 38
Weight loss.................................. 38
Nutrient changes............................. 41

 

 

 

 

 

 

III. PLANT PRODUCTION AND PLANT NUTRIENT CYCLES....... 45
INTRODUCTION...................................... 45
‘VEGETATION OF PENTWATER MARSH..................... 50
OBJECTIVES........................................ 51
METHODS........................................... 52
Field Procedures............................. 52
Laboratory Procedures........................ 57

Image analysis and community mapping......... 58

Data analysis................................ 61
RESULTS........................................... 64
Plant biomass................................ 64

Plant nutrient content....................... 69
Litter biomass and nutrient content.......... 84

Image analysis............................... 96
DISCUSSION........................................ 96

 

 

 

 

 

 

IV. NUTRIENT MASS BALANCE STUDIES.......... ....... ....120
INTRODUCTION......................................120
OBJECTIVES........................................122
METHODS...........................................122

Hydrologic budget............................122
Chemical sampling............................124

 

 

iii

Chemical analysis............................126
Nutrient Mass Balance......... .............. .127
RESULTS...........................................129
Hydrologic budget............................129
Water chemistry.. ...... ......................134
Mass Balance............ ..... ....... ......... 138
DISCUSSION.................. ....... . ..... . ........ 147

 

 

 

 

V. SUMMARYOCCCOCCOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 0.0.0155

LIST OF REFERENCES. ..... OOOOOOOOOOOOOOOOOOO ...... O ..... 162

iv

LIST OF FIGURES

Figure 1.1 The Pentwater River drainage basin.

Figure 1.2 Major vegetation zones of Pentwater
Marsh (1983).

Figure 2.1. Weight loss from 1983 litterbag
studies.

Figure 2.2. Changes in the total N and P mass
in the 1983 litterbag studies.

Figure 2.3. Changes in the % N and % P in the
1983 litterbag studies.

Figure 2.4. Changes in the % weight, % N mass
and % P mass remaining in 1980-1982 litterbags.

Figure 3.1 Changes in Lake Michigan water
levels (annual mean), 1960—1983 (NCAA/NOS,
1984). Vertical bars indicate the range of
monthly means. Arrows indicate years of
photographic analysis.

Figure 3.2. Location of transects sampled for
shoot and litter biomass during 1980 and 1981.

Figure 3.3. Location of transects sampled for
shoot and litter biomass during 1982 and 1983.

Figure 3.4. Monthly changes in live biomass in
the six communities sampled (1982).

Figure 3.5. Monthly changes in tissue nitrogen
concentrations in shoots of six major plant
species (1982).

Figure 3.6. Monthly changes in tissue P
concentrations in shoots of six major plant
species (1982).

18

24

31

35

49

54

56

67

73

74

Figure 3.7. Monthly changes in the N mass in
the shoot biomass of the six communities sampled
(1982).

Figure 3.8. Monthly changes in the P mass in
the shoot biomass of the six communities sampled
(1982).

Figure 3.9. Monthly changes in litter biomass
of the six communities sampled (1982).

Figure 3.10. Monthly changes in tissue nitrogen
concentrations in the litter of six major plant
species (1982).

Figure 3.11. Monthly changes in tissue P
concentrations in the litter of six major plant
species (1982).

Figure 3.12. Monthly changes in the N mass in
the litter biomass of the six communities
sampled (1982).

Figure 3.13. Monthly changes in the P mass in
the litter biomass of the six communities
sampled (1982).

Figure 3.14. Estimated changes in the coverage
of the major vegetation zones of Pentwater
Marsh, 1965-1983.

Figure 3.15. Changes in the distribution of the
major vegetation zones of Pentwater Marsh,
1965-1983.

Figure 3.16. Nitrogen and phosphorus cycling in
six emergent and meadow communities.

Figure 3.17. Estimated changes in the August
nitrogen storage in shoot and litter biomass,
1965-1983.

Figure 3.18. Estimated changes in the August
phosphorus storage in shoot and litter biomass,
1965-1983.

Figure 4.1. Locations of stream gaging stations
and water sampling sites.

Figure 4.2. Precipitation (Hart, MI) and
streamflow during 1983 sampling periods.

vi

82

83
86

88

89
94
95
97
98
105

116

125

131

Figure 4.3. Changes in water velocity during a
seiche cycle (15 September 1982) at Pentwater
Marsh outlet.

Figure 4.4. Nitrate-N and Total-N
concentrations in the North Branch, South
Branch, and marsh outlet (1983).

Figure 4.5. Total reactive—P and total-P
concentrations in the North Branch, South
Branch, and marsh outlet (1983).

Figure 4.6. Nitrate-N concentrations measured
at the Pentwater Marsh outlet, 13-14 June, 1983.

Figure 4.7. Export and retention of total-P,
Nitrate-N, chloride and organic-N during 1983
sampling intervals.

Figure 4.8. Seasonal and annual (1983) nitrogen
budget of Pentwater Marsh.

Figure 4.9. Seasonal and annual (1983)
phosphorus and chloride budgets of Pentwater
Marsh.

Figure 4.10. Chloride concentrations in the
North Branch, South Branch and marsh outlet
(1983).

Figure 4.11. Diurnal 02 changes (6 September,
1983) (A), and changes in 0 concentrations
across flooded marsh communities (B).

vii

135

137

139

140

141

143

145

148

153

LIST OF TABLES

Table 2.1. Summary of litterbag studies
conducted in Pentwater Marsh.

Table 2.2. Significance (p (.05) of % weight, N
and P remaining treatment differences between
flooded and non-flooded litterbags (+ = flooded
> non-flooded, - = non-flooded >flooded, m =
missing data, ns = not significant).

Table 2.3. Significance of between species
comparisons of mean weight, nitrogen, and
phosphorus remaining in September, 1984
litterbag samples (+ = significance (p <0.05);
values along diagonals are mean % remaining).
Species abbreviations are the same as those used
in Table 2.2.

Table 2.4. Mean N and P concentrations (% by
dry wt) in 1983-1984 litterbag studies
(underscored values are not significant at p
<.05, # indicates significance between treatment
means on the same sampling date). The initial
treatment (Nov—Apr) was suspended 30 cm above
the substrate.

Table 2.5. Decomposition coefficients (k
units/year) for litterbag studies conducted
between 1980 and 1983 (k calculated after 1 year
of decay).

Table 3.1. Aerial photography used for
preparation of vegetation maps (CIR = color
infrared).

Table 3.2. Signature characteristics of wetland
cover types identified from 1983 true color,
1:8,600 transparencies.

Table 3.3. Mean September % N and % P mass

remaining in litterbags (Chapter 2) after 8, 12
and 21 month intervals.

viii

13

20

21

26

37

59

6O

63

Table 3.4. Estimated coverage of the dominant
plant communities in Pentwater Marsh (1983).

Table 3.5. Percentage contribution to the total
maximum shoot biomass by the 7 major species
(1980-1983).

Table 3.6. Significance (p < .05) of 1982
monthly shoot biomass samples (underlined means
are not significant, only significance between
sequential dates indicated).

Table 3.7. Significance (p <.05) of 1982
maximum biomass between community types, and
1983 maximum biomass (see Table 2.3 for species
abbreviations).

Table 3. 8. Significance (p <. 05) of monthly
root biomass samples (x, sx, n).

Table 3.9. Percentage of biomass contribution.
(1982) to the 6 community types by species (see
Table 2.2 for species abbreviations).

Table 3.10. Monthly N content (in % dry weight)
in shoot tissues of the 7 dominant plant species
in Pentwater Marsh.

Table 3.11. Monthly P content (in % dry weight)
in shoot tissues of the 7 dominant plant species
in Pentwater Marsh.

Table 3.12. Nitrogen concentrations (% by dry
weight) in root/rhizome tissue during 1983.

Table 3.13. Phosphorus concentrations (% by dry
weight) in root/rhizome tissue during 1983.

Table 3.14. Percentage of August shoot nutrient
mass attributable to the six major community
types.

Table 3.15. Significance (p < .05) of 1982
monthly litter biomass samples (underlined means
are not significant, only significance between
sequential dates not indicated).

Table 3.16. Monthly N content (in % dry weight)

in litter biomass of the 7 dominant plant
species in Pentwater Marsh.

ix

65

66

68

7O

71

72

76

78

8O

81

85

87

90

Table 3.17. 'Monthly P content (in % dry weight)
in litter biomass of the 7 dominant plant
species in Pentwater Marsh.

Table 3.18. Comparison of community biomass
production values for Pentwater marsh to
literature values.

Table 3.19. Comparison of community biomass
production values for Pentwater marsh to
literature values (gm m' ).

Table 3.20. Comparison of total N and P mass in
live biomass_£or Pentwater marsh to literature
values (gm m ).

Table 3.21. Percent of maximum nutrient mass
(summed by community type) remaining after 21
months of decay (estimates based on litterbag
mineralization studies).

Table 3.22. Comparison of vegetation changes in
Pentwater Marsh and the Betsie River Marsh.

Table 3.23. Nutrient cycling characteristics of
marsh vegetation.

Table 3.24. The distribution of N and P (kg/ha)
in shoot biomass of emergent and wet meadow
communities during August, 1983.

Table 3.25. The distribution of N and P (kg/ha)
in litter biomass of emergent and wet meadow
communities during August, 1983.

Table 4.1. Laboratory analysis used for
nutrient determinations of water samples.

Table 4.2. The water budget and chloride mass
balance for Pentwater Marsh (January 1,
1983-December 31, 1983).

Table 4.3. Percentage of time outflow from
Pentwater Marsh was measured at the marsh outlet
(July 1, 1983 - March 16, 1984).

Table 4.4. Nutrient concentrations (mg/L) of
the input and output streams of Pentwater Marsh,
January 1, 1983 -December 31, 1983 (mean and
standard error).

92

101

102

104

108

110

128

130

133

136

Table 4.5. The nitrogen mass balance for
Pentwater Marsh (January 1, 1983 - December 31,
1983).

Table 4.6. The phosphorus mass balance for
Pentwater Marsh (January 1, 1983 - December 31,
1983).

Table 4.7. Drainage areas and water yield from
watersheds adjacent to Pentwater Marsh (January
1, 1983-December 31, 1983).

Table 5.1. Soil characteristics of the six
vegetation types in Pentwater Marsh.

xi

144

146

140.

156

I . INTRODUCTION

Wetlands are diverse and complex ecosystems
characterized by intense rates of production and material
processing (i.e. energy and nutrient transformation). The
ecological functioning of these systems is controlled by
numerous interactions between biological, chemical, and
physical properties. There are, however, few wetland
systems for which these processes have been adequately
quantified. Some of the most thoroughly studied wetlands
are the Atlantic coast salt marshes (Pomeroy and Wiegert,
1981), and even they are not thoroughly understood (Nixon,
1980). Research on ecosystem structure and function in
Great Lakes marshes has been less detailed and the current
data base is more fragmented.

Ecosystems are 'linked' by interactions between the
biogeochemical functioning of adjacent systems. These
linkages control the internal nutrient cycles and the flux
of nutrients across ecosystem boundaries (Likens, 1975).
Since wetlands are situated between aquatic and terrestrial
systems, they are thought to function as important linkages
between these environments and to regulate the movements of
water, nutrients and other materials between them.

This dissertation research focused on several aspects
of the ecosystem function in a Great Lake riverine marsh in
an attempt to quantify some of the 'linkages' between

adjacent aquatic systems. The studies are presented in

2
three chapters, each with a review of the pertinent
literature, a summary of the procedures, and a presentation
and discussion of the results. The final chapter

synthesizes the research findings.

OBJECTIVE
The overall objective of this study was to assess and
quantify the effects of a marsh on the hydrologic export of
nitrogen and phosphorus. Nitrogen and phosphorus were
selected for this study since they are most likely to be
the limiting nutrients in marshes and other aquatic
ecosystems (Valiela, 1978; Wetzel, 1983). The major

hypothesis of the study was that a riverine marsh retains N

and P, at least seasonally, and that this retention could

be partially controlled by the nutrient uptake and release
patterns of emergent plants. Specific objectives were:

1. To quantify nitrogen, phosphorus and hydrologic inputs
into a Lake Michigan rivermouth marsh and to quantify
outputs of these parameters from the marsh.

2. To quantify emergent plant production and nutrient (N
and P) uptake and release patterns in marsh vegetation

over an annual cycle.

3. To evaluate the potential role of emergent macrophytes
in controlling water quality in this marsh.

4. To identify the effects of Lake Michigan waterlevel
fluctuations on the cycling of N and P by emergent and
wet meadow vegetation.

3
DESCRIPTION OF THE STUDY SITE

Riverine marshes or wetlands are defined as wetlands
adjacent to or near rivers or streams where the water in
the river or stream is the principle inflow to the wetland
(Tchobanoglous and Culp, 1980). Surrounding the Laurentian
Great Lakes, riverine wetlands are common and account for a
large percentage of the total wetland area directly
bordering the lakes. In particular, along the eastern
shore of Lake Michigan a series of geomorphologically
similar marshes form the dominant wetland type (Jaworski et
al., 1978). The marshes occur in delta regions where third
to fifth order rivers enter the head of small bays or
barrier lakes of Lake Michigan. Some examples of these
wetland systems include those found at the mouths of the
following rivers:

Muskegon River, Muskegon Co.

White River, Muskegon Co.

Pentwater River, Oceana Co.

Pere Marquette River, Mason Co.

Manistee, Little Manistee Rivers, Manistee Co.

The barrier lakes and bays along the eastern Lake
Michigan coast are of post-glacial origin. Their formation
dates to the Lake Chippewa Stage (6,500 - 6,000 y.b.p.) of
the Lake Michigan Basin when the Lake Michigan water level
was some 70 m above sea level, 76 m below its present stage
(Bough, 1958). Rivers draining into Lake Chippewa
underwent considerable downcutting, forming deep ravines

along the edge of the present shoreline. More recent lake

stages flooded these ravines forming bays, most of which

have been partially isolated from Lake Michigan by
shoreline processes resulting in the formation of barrier
lakes. The present wetland systems have developed at the
head of these basins where water depths are suitable for
the growth of aquatic vegetation.

Pentwater Marsh, Pentwater Twp., Michigan, USA (430
46' N, 860 24' W) was selected for this study as
representative of Lake Michigan riverine wetlands. The
marsh basin is well defined, being separated from adjacent
upland areas by steep slopes rising up to 25 m above the
marsh on the north and south sides. The predominant
hydrologic features of the basin are the North and South
Branches of the Pentwater River which join in the marsh to
form the Pentwater River (Figure 1.1). The Pentwater river
flows through the marsh into Pentwater Lake. The 423 km2
of watershed is primarily forest, abandoned farmland, or
orchard. There is a small power dam and reservoir at Hart,
a small community on the South Branch. There is also some
secondary sewage input into th South Branch from this
community.

Pentwater Lake is connected to Lake Michigan by a
jettied channel. Since most of the marsh basin, Pentwater
Lake and Lake Michigan are at the same elevation, the basin
is technically a bay of Lake Michigan. The marsh is
subjected to both long and short term water level
fluctuations of Lake Michigan. At the marsh outlet, seiche

(atmospheric generated waterlevel occilations) activity of

 

kn: 5
SCALE
WATERSHEDS: LAKES:
A North Branch a Pentwater Lake
8 South Branch b Hart Lake
C Lambricks Creek c Lake Michigan
D Watsons Creek

fl'l

Figure 1.1

Ungaged

The Pentwater River drainage basin.

6
Lake Michigan is dampened considerably by Pentwater Lake
(Seeling and Sorenson, 1977) but the water level changes
are great enough to cause current reversals during periods
of average or below average discharge. Seiche influences
are greater at above average Lake Michigan water levels.
The flow reversals affect the retention of water from the
marsh and in this way, the wetland shares similarities to
marine wetlands subjected to astronomical tides.
The total area of the marsh basin is 118 ha and 62% of

this area is dominated by emergent or wet meadow

vegetation. Sparganium eurycarpum Engelm. and Typha

 

latifolia L. dominate the emergent zone, while Carex

stricta Lam. and Calamagrostis canadensis (Michx.) Beauv.

 

dominate the less frequently flooded wet meadow zone
(Figure 1.2). Along several abandoned channels Scirpus
validus Vahl. is the dominant plant community. In addition

to these communities, a shrub zone composed of Alnus rugosa

 

(Du Roi) Spreng. and Salix L. spp. borders much of the
basin. Portions of the river channels, several old river
channels, and a ponded area near the marsh outlet are

dominated by submergent macrophytes (Potamogeton crispus

 

L., Myriophyllum spicatum L. and Anacharis canadensis

 

(Michx.) Rich.)

Pentwater marsh was selected by the Michigan Sea Grant
Program as a site for several interdisciplinary studies of
ecological processes which control energy flow and material

processing in a rivermouth marsh. These studies have

 

 

VEGETATION
PENTWATER MARSH
1 983

H

DUPLAND .

. EEMERGENT
. mwer MEADOW

.Ensnaue
-WATER/SUBMERGENT

-LAKE

-ROADS

 

 

Figure 1.2 Major vegetation zones of Pentwater Marsh
(1983).

8
included avian utilization, larval fish production, fish
population dynamics, and sedimentation studies.
Specifically, for this research project, the site appeared
to satisfy several important criteria for the developement
of mass balance data (Chapter 4). The diverse emergent and
meadow communities provided an opportunity to study some
important aspects of the processing of N and P by marsh
vegetation. Plant nutrient uptake studies are presented in
Chapter 3 and decomposition studies are discussed in

Chapter 2.

II. LITTER DECOMPOSITION AND NUTRIENT MINERALIZATION

INTRODUCTION

Decomposition results in the dissipation of energy and
the release of the mineral nutrients stored in organic
matter. The dynamics of plant litter decomposition are
important processes which can control other aspects of
ecosystem function, in particular the retention and
accumulation of nutrients and organic matter in ecosystems.
In marshes, mineralization releases nutrients stored in
plant litter to the soil and surface water where they
become available to producer organisms or microbial
populations. Additionally, these nutrients may be
exchanged (through absorption, ion exchange, and diffusion
processes) with the marsh sediments, chemically
transformed, or exported from the marsh. When plant
production rates exceed decomposition rates, nutrients
accumulate in the organic matter and sediments of the marsh
basin, and/or may be exported from the marsh as detritus.
In a marsh basin, any net storage of nutrients in the
litter pool could result in an overall reduction of
nutrient exports compared to nutrient inputs.

Some decomposition may be accomplished by physical-
chemical processes such as fragmentation, leaching or

autolysis, but the dominant pathway of decay is through

10

biochemical processes involving microbial and/or
detritivore populations (Puriveth, 1980).

Decomposition rates are variable between species, and
depend in part upon plant morphology, the chemical
composition of the litter, and site environmental
conditions. Aspects of plant morphology such as leaf and
stem thickness, shape, or size can affect fragmentation and
the surface area available for colonization. Some of the
most important chemical attributes that control
decomposition rates of plant litter are the total fiber
content (Godshalk and Wetzel, 1978), lignin content (Aber
and Melillo, 1982), nitrogen content (Coulson and
Butterfield, 1978; Enwezor, 1976; Day, 1982), and the ratio
of some of these components (e.g. C:N) (Berg and Ekbohm,
1983). The most important site conditions affecting decay
rates are temperature, oxygen status and nutrient
availability (Day, 1982; Boyd, 1970; Frasco and Good, 1982;
Godshalk and Wetzel, 1978). These environmental parameters
are frequently controlled, or strongly influenced by the

hydrologic regime of a wetland.

OBJECTIVES
Decomposition processes were studied at Pentwater
marsh to identify the storage and dynamics of N and P in
plant litter. The specific objectives of these studies ‘

were:

11

1. To identify differences in the decomposition rates
between species and different sites.

2. To determine which plant communities were more
conservative of N and P.

3. To predict what effect changes in species dominance or
distribution might have on overall mineralization
rates in the marsh.

4. To evaluate the significance of nitrogen and

phosphorus storage and flux in the litter pool to the
N and P cycles of the marsh.

METHODS

Litterbag techniques are frequently used to study
decomposition processes ig_§itg by enclosing known amounts
of plant material of known chemical composition in mesh
bags and measuring the weight and chemical changes of the
confined litter (Chamie and Richardson, 1978; Davis and van
der Valk, 1978). In these studies, weight loss represents
nutrient mineralization, the release of soluble organic
compounds and dissolved inorganic nutrients as well as the
loss of detrital particles smaller than the mesh size
(Boyd, 1970). Decay rates are found to increase with
increasing mesh size indicating that either particulate
losses are significant in larger mesh bags and/or that the
smaller mesh bags eliminate invertebrate populations which
facilitate decomposition (Brinson et al., 1981).
Regardless of these drawbacks, the mesh bag simulates the
decay of the fallen litter component (Davis and van der
Valk, 1978) and is a replicable assay of decomposition.

The litterbag technique was used in this study to

12

determine decomposition rates for seven of the dominant
plant species found in Pentwater marsh. The studies were
conducted over a four year period from October 1980 to
December 1984 (Table 2.1). During this period
decomposition and nutrient losses from litter of 7 emergent
and wet meadow macrophytes were examined.

For each species, senescent, standing litter of the
current seasons production was collected in early October.
The litter was cut to 20-25 cm lengths and partially dried
at 200 C for 36 hr before 14-16 g of material was packed
into litterbags. This temperature was selected to minimize
temperature induced chemical changes in the litter (Chamie
and Richardson, 1978). Subsamples of this litter were then
dried at 650 C to a constant weight and retained for
chemical analysis. From these subsamples, a conversion
factor was calculated and all weights are expressed
relative to 65° C. Litterbags were constructed from
nylon mesh netting sealed along the edges with a soldering
iron.

Initially, litterbags were collected monthly, but as
the studies progressed, and during the winter months,
longer samples intervals were choosen. For each species,
and sampling interval, 4-6 bags were collected. Upon
collection, the bags were rinsed with river water to remove
colonizing animals, sediments and particles smaller than
the mesh size (Boyd, 1970). In the laboratory the bags

were opened and further cleaned of macroscopic organisms

13

Table 2.1. Summary of litterbag studies conducted in
Pentwater Marsh.

 

 

 

 

 

 

 

 

 

 

 

 

 

Year Species Mesh (mm)
Nov. 1980 Sparganium eurycarpum 3.2
thru Typha latifolia 3.2
June 1983 Carex stricta 3.2
Carex rostrata 3.2
Nov. 1981 Sparganium eurycarpum 1.5
thru Typha latifolia 1.5
Sept. 1983 Carex stricta 1.5
Calamagrostis canadensis 1.5
Nov. 1982 Scirpus validus 1.5
thru Carex aquatilis 1.5
June 1984
Nov. 1983 Sparganium eurycarpum
thru Typha latifolia
Dec. 1984 Scirpus validus

 

Carex stricta

Carex aquatilis

Carex rostrata
Calamagrostis canadensis

 

 

Add—3.3.4...)
o o o o o o o
NNNNNNN

 

14
(animals, seedlings, and invading shoots and/or roots).
The litter samples were then dried to a constant weight at
650 C. Samples were weighed to the nearest 0.1 g and
tested for total nitrogen and total phosphorus using the
wet digestion procedure described in Chapter 3.

The initial nitrogen and phosphorus mass in each bag
was calculated from the initial sample weight and the
initial tissue nutrient concentration (%N or %P by dry
weight) measured in the reference subsamples. The final
nutrient mass in each bag was determined from the final
sample weight and the tissue %N or %P concentration. For
comparative purposes, the final nutrient mass was expressed
as a percentage of the original mass present. The
significance between species and treatment means were
determined using an analysis of variance’and Duncan's
multiple range test (Steele and Torrie, 1980). Decay
constants (k units/yr) assuming a negative exponential
model (Olson, 1963) were used to compare annual decay
losses between years.

In 1983, overwintering losses from standing litter and
the effect of flooded vs. non-flooded environments on decay
processes were quantified. Senescent shoot material of 7
dominant species (Calamagrostis canadensis, Carex stricta,
Carex aquatilis, Carex rostrata, Sparganium eurycarpum,
Scirpus validus and Typha latifolia) were collected in
mid-October and used to fill a total of 480 bags which were

placed in the marsh by mid-November, 1983. Bags containing

15
litter of each Species were initially suspended 30-40 cm
above the substrate to simulate overwintering standing
litter. In April, 1984, the remaining suspended litterbags
were moved to flooded and non-flooded substrates,
simulating spring litter fall into these two environments.
Where possible, bags were placed in sites with their
respective species present, but dry stands of the emergent

species and flooded stands of Carex stricta, Carex rostrata

 

 

and Calamagrostis were not found. The bags containing

 

litter of the emergent species were placed in a mixed
meadow stand and those representing meadow species were
placed in flooded stands of mixed emergents. Suspended
bags were sampled in February and April, 1984. Bags of the
other treatments were sampled in June, August, September,
and December, 1984. After collection, the suspended
litterbags were returned to the laboratory and dried
without further preparation.

During 1980, studies were initiated to examine long
term (2.5 years) weight and nutrient changes. The
litterbag mesh size in these studies was 3.2 mm x 3.2 mm
and a total of 200 bags were constructed to study four
species (Typha latifolia, Sparganium eurycarpum, SEES!
rostrata, and 93325 stricta). All bags were placed in the
marsh during November on non-flooded substrates in stands
of their respective species.

Studies initiated in 1981 provided replication of the

1980 treatments of Typha latifolia and Carex stricta. In

16

addition, Sparganium eurycarpum bags were placed in
continuously flooded stands and decay of Calamagrostis
canadensis was studied on non-flooded substrates. Mesh
size in these studies was 1.5 mm x 1.5 mm and a total of
180 bags were constructed. In 1982, decay of two
additional species were studied. Eighty bags containing
litter of Scirpus validus and Carex aguatilus were placed
on flooded substrates in stands of their respective

species.

RESULTS

The 1983-1984 decomposition studies included the seven
dominant plant species in the marsh and allow direct
species by species comparisons of decomposition rates and
the nutrient dynamics in litter under different treatments.
These data are discussed first.

Suspended bags were buried in snow between
mid-December and late-February while bags placed in flooded
stands remained flooded by 25 - 100 cm of water during the
entire study. ~Due to above average Lake Michigan water
levels, bags transferred to non-flooded sites became
flooded with 1 - 25 cm of water in mid-June and generally
remained flooded until mid-September. Despite this period
of inundation, these bags were subjected to different

environmental conditions than were the continuously flooded

17

treatments. For convenience this treatment will be

referred to as the non-flooded treatment.

Weight loss

The pattern of weight loss during the 1983-1984 study
was generally similar for all species investigated, but the
magnitude of loss varied between species and treatments
(Figure 2.1). Overwintering weight loss was significant (p
<0.05) only for Calamagrostis canadensis and Sparganium

eurycarpum and was less than 10% of the original weight for

 

all species. Rates of loss were greatest during the early
summer months and somewhat slower during the autumn. By
the end of the study, total weight loss varied between 24%
- 83% depending upon the species and treatment.

In September and December, significant (p (0.05)
differences between species and treatments existed (Table
2.2, Table 2.3). Of the 21 possible between species
comparisons, 18 in the flooded and 15 in the non-flooded
treatments were significantly different. All species
except Carex aquatilis showed significant (p <0.05)
differences between the flooded and non-flooded treatments.
Typha latifolia was the only species with greater weight
loss in the non-flooded treatment (39%) vs. the flooded
treatment (28%). The remaining five species, in September,
(averaged 27% greater weight loss in the flooded vs. the
.non—flooded treatments. There were also differences in the

:redative decay rates of the species between treatments.

Carex aquatilis

'—(

i—E

 

§

I 1
ID 0
h to

bugugawai m 95

Carex stricta

      

a non-flooded

o suspended
I flooded

.25-4

 

r-fi
1—0
-2
-O
—¢f)
r-(

1—5

 

 

 

1
O O
00

(Magnum: w

I
no
N
%

—k'

D
—Q'
~06)

2v-

date

1984

1983

18

    

Carex rostrata

   

l—lLQ
L3,.
.09,

2?;

 

 

I l l
8 2 8 a
.-

bugugamai w 1.

Calamagrostis canadensis

 

 

 

l 1 l

o n
8 “'3 an N

' bugugawa: 3M 95

Weight loss from 1983 litterbag studies.

Figure 2.1.

    

Scirpus validus

bfip

~01")

 

2v-

 

     

Sparganlum eurycarpum

 
   

6111111311191 m

e suspended
flooded

A non-flooded

date

’5
-:E
-<
-:
u—gV
1.3.-

hon

 

 

100

I
I)
:5

0
no
Bugugemea 3M 95

 

Zh-

19

Typha latifolia

 

--s
1—0
1-2
--0
—m

P(

l
J
date

 

 

*__

to: 2 ..
00101811191 w

l

l
O

l
D
N

%

(Cont‘d.).

Figure 2.1.

20

Table 2.2. Significance (p <.05) of % weight, N and P
remaining treatment differences between flooded
and non-flooded litterbags (+ = flooded >
non—flooded, - = non-flooded > flooded, m =
missing data, ns = not significant).

 

 

September
Sp, Cc Cs Cr Ca Tl Sv
WT + + + + + - +
N + + + + ns ns +
P ns + ns + + + +
December

Sp Cc Cs Cr Ca Tl Sv

WT (m) + + + + + (m)
N (m) + + + + ns (m)
P (m) + + + + + (m)

 

1Sp z Sparganium eurycarpum, Cc = Calamagrostis
canadensis, Cs = Carex stricta, Cr = Carex rostrata,
Ca = Carex aquatilis, T1 = Typha latifolia, Sv =
Scirpus validus

 

 

 

 

21

 

 

mu mm? 00 mm?

no me new Ha m: we

HE + + mm? mu m: m: mor

mo + + m: mo? mu + + + Pm

mm + + + + mm no + + + m: we

>m + + + + m: we mm + + + m: m: mm

00 + + + + m: m: mm >m + + + + m: we Pv
mu H0 H9 m0 ow >w 00 He m0 m0 no om >m 00
.pmpoo~mncocv mCACAMEoH z m Apopoonv mficwmemu z w

mm Nb He on

He + me no + me

no + on so mm + + be

MU + m: mm mm >m + + m: we

no + + + m: mm mo + + m: m: we

>m + + + m: m: Fm 00 + + + + + cm

00 + + + + + + mm HO + + + + + + mw
mo mm HE m0 nu >m 00 He no mm >m mu 00 no

Apmpoonucoc. mcflcflmeou usmflos w

Apopoonv mcflcflmeou unmwm3 w

 

mcofiumfi>ounnm mowooam

a. oocmowmacmflm
can .comouufl: .usmfimz some mo mcomfiumaeoo mwfloQO :mm3umn mo mocmonHcmam

.~.m manna a“ come omonu we deem on» one
.AmCACAmEou w some one mamcommflo macaw monam> “.mo.v
u +V moamemm menuouufia «mar .uoneouamm a“ mcficflmEou msuonamona

.m.m £ng

22

 

 

Apoooonncoc. mcflcflmeou m w

He hr mu VF

mu + om? He + N—

ew + + Nor mo + + hm

>m + + + ms nu + + m: om

mu + + + m: we >m + + on m: we

DO + + + m: m: we DO + + w: m: m: ov,

mm + + + + + + mm mm + + m: we m: m: ow
HE mo no >m mo 00 mm mo He no HO >m 00 mm

Aomooon. mficflmemu m w

.poSCAucou

.m.~ OHQMB

23
For instance, while Carex stricta decomposed relatively
rapidly in the flooded environment, it showed the smallest

weight loss of any species in the non—flooded treatment.

Nitrogen changes

The nitrogen mass in all suspended litterbags except

Carex stricta decreased during the winter months. This

 

(decline averaged 32.6% in February and 28.6% in April and
(was attributable primarily to the decrease in the tissue
riitrogen concentrations during this period (Figure 2.2,
{Table 2.4). The nitrogen decline was significant (p (0.05)
for all Species except Carex aquatilis and Carex stricta.
Exollowing this decline, nitrogen concentrations increased
tdiroughout the growing season in both the flooded and
non-f looded treatments .

All species showed some increase in the absolute
ltitter N content following the minimum observed levels.
Tfiuis increase was similar in both flooded and non-flooded
‘tereatments and averaged 114%. Calamagrostis was the only
Slpecies with no nitrogen accumulation over initial content
liri either of the treatments. Both the flooded and
nGin-flooded treatments of Typha latifolia, Scirpus validus,
.Carex stricta, and Sparganium eurycarpum showed net

accumulations in excess of the initial levels. The

remaining species, Carex rostrata and Carex aguatilis,

exhibited N accumulations above initial content in the

non-flooded treatment only.

—~’

211

.mwflpsum
mmnntUAH mma_ man an mmme a new 2 Hmuou one an mwmcmno .~.N masons

2
(u

 

 

 

 

 

 

 

 

 

 

 

 

  
   

 

Bee 1:... 32

U comp now, _. o 2 O m ( a. q. 2 < 2 u a O 2
020m<aa2<2uqoz .rrhprpbrrpri.o
___LPPL~_L_LLO

an
E\. .. “N. ,3...
13.0
a.
Q .d
4 965 fie W
M
I.
P h h h h r b b b b i- b h b o
i! b b L L (P P h b P b b P o
as
iii. [no
Inc an N
m
. N to; M
.6. Q 1
. e
mm—Ouzfl— U£Q>P . IMO mm—zflaaa X0550 n.—
m. But 1:... one.

Sec «3.2:... qonm<aaz<zuaoz
so20m<qaz<2uaoz _P___.._r._r»_
p___ptrr___p_bo e

um
peace: I lmod
I no.0 % oevoo:ucoc d d
evacuees» e a.
d A
. 19.0M
1.0qu ‘0
A
U
M fora
fife...
.11 b b P P b 1» L F b P PL by a r .P b h tr b (P b i? (P b b P b o
Ind % u a wee an
26:9, 39.6w N 5398330 63cm Se m W
19. ml .6. A
w
w
-.. rm.
row

 

25

...o.ucoov .m.~ mnsmhm

 

Set 322::
4onn<qsaaauqoz
pb_p..p.r._bLFo

13.6%
d
liq
Eek
M
rec:

FP—pbphprPppb

 

28.2: .830

Sat 8st

 

In..."

     

 

 

no venue: l
ace 1:2.
vegans: « Ind %
N
to. M
20:.» .830 1.... w «.3323 3.352.530 re

 

26

 

 

 

 

 

 

 

 

 

 

 

 

 

pmpoon Nrp.o ooP.o vmo.o hvo.o
m w a a
popoon co: woe.o «mo.o who.o Poo.o nvo.o mmo.o Pno.o
empoon mom.r mum.o omm.o mvm.o
z w *
cocoon co: mmv.w mmm.o who. Nam.o mom.o emv.o vow.o
mfimcopmcmo mwumoummEmHmo
popooHu mNP.o mmw.o mmo.o hmo.o
m w
empoon cos .2. SPF.o obo.o mmo.o mmo.o moF.o mer.o
popcon mmn.r mum.w omN.P emm.e
z w
topoon no: as. hum.w mmm.o omm.o vm>.o Frm.o omv.P
Esmumoxuso Eaficmmummm
Bzmzedmme 0mm Bmmm 03¢ mzoh and mum >02

 

o>onm EU on coflmcwamsm mos Auadn>ozv unoEumouu Hmfluflcfi one

.oumuumnom on»
.Imume

mcflamEmm meow ecu co memos unoEumouu coosuon oocmoflwficmflm moumoapcfi
* .mo.v a pm ucmofiuficmfim you one mwoam> pwuoomuoncsv moflozum

mmnumuuga ammP1mmmP an .us sum up a. mcoflumuuewocoo a one 2 new:

.v.N magma

27

 

 

 

 

 

 

 

 

 

 

 

 

emooon mpp.o emr.o va.o moo.o
m w a a
cocooaw co: mFP.o sor.o who.o mmo.o mmo.o mmo.o Poo.o
concede oeo.F emm._ mom.P orm.o
z w *
popooHu co: ¢m~.P noo.F mnn.o Pub.o vmm.o mem.o Nme.o
mumuumou xouou
cocooau mow.o aPP.o moF.o mmo.o
m w a e a 1
cocoon co: nrp.o Pno.o nmo.o mmo.o vmo.o bmo.o wmo.o
popcon mmm.— mm~.P nem.e wam.o
z w e *
omooon co: mwv.P omF.F mmo.o qmn.o mom.o mmm.o mwm.o
muowhum XOHMU
ezmzecmme omo .emmm 0:4 mzou mmm mmm >oz
poscaucoo .v.~ canoe

28

 

 

 

 

 

 

 

 

 

popoon ovo.o heo.o mmo.o hmo.o.//
w .
popoon co: ho—.o who.o Poo.o mmo.o Nmo.o mno.o mmo.o
popoon moo.F omm.o emm.o Pme.o
m t V
cocoon co: mam." mmb.o mom.o hmo.o mpv.o Pmm.o mpv.o
. announumH mamas
cocoon Aev mmo.o .Ev ~mo.o
w e . e
popoon co: .5. who.o Nmo.o mno.o mvo.o ~wo.o who.o
concede Ia. mmm.o As. «mw4d//////
w e
poooon coc As. nFm.o new.o mmw.o vmv.o hbv.o emm.o
mfiaaumnmm xoumo
ezmze<mme omo emmm oo< mzon «ma mum >oz
.poocfiucou .v.~ wanes

29

 

vuo.o

 

 

 

 

cocoon omF.o ~mo.o mmo.o
s. e
popoon co: As. bop.o vso.o mmo.o ~mo.o moo.o mmo.o
popoon mmb.P me.P mov.F 5mm.o
e a
popcon :0: As. w>~.F omm.o mmm.o mom.o hem.o hmp.w
mspfiam> msmeom
Bzmzaflmme Omn Baum 0:4 mzan mm< mum >02

.eoscaucoo .¢.~ magma

30

The greatest N accumulation occurred in Sparganium
eurycarpum litter where the June nitrogen content was 2.4
times the initial content. By December, only the
non-flooded treatments of Typha, and the three Egggx
species continued to retain N at greater than 100% of the
initial N content. Typha latifolia was the only species in
the flooded treatment to retain N at greater than 100% the
initial values. Between species differences in September
were significant for 13 of the flooded comparisons and 16
of the non-flooded comparisons (Table 2.3). As with weight
loss, the species ranking by %N remaining changed between

the dry and flooded treatments.

Phosphorus changes

 

Trends in the P content of the litter samples
generally followed those of N, though the overall retention
of P was less than that of N (p (0.05) (Figure 2.3).
Initially P was lost more rapidly than N, but two species,
Typhg latifiolia and Carex stricta, showed a net increase
in absolute P content during the suspended treatment.

The accumulation of P averaged 30% greater than the N
accumulations and was 129% greater than the observed
minimum P content of the litter. Three species, Egggx

canadensis, Carex rostrata, and Scirpus validus did not

 

accumulate P above the initial mass in either of the
treatments. Both treatments of Typha latifolia, Sparganium

eurycagpum, and Carex stricta and the non-flooded treatment

Carex stricta

e suspended
e non—flooded

I Needed

p.5
.0
1—2
—O
.10
L.<
h—fi
r—fi
I—I
p-<
I—S

unu'

date

—‘5"

ptbn
o

 

:-

 

200*!
‘50‘
00

Bugugawei N 1,

Calamagrostis canadensis

 

Bugugeum d 1,

P5
—0
1—2
--0
1.10
l—(
P52
bfig
-2
—<
r—!
u-u'
-q—
u-On

 

I I

s a

bugugamea

I

O

o
N

I,

 

150i
0..
30‘

bugugewea d N.

 

31

 

 

 

r r-
m
8 > r
a . _
>
n P p
3
9 » _
8 . _
I- P-
. )-
b h
b )—
r I.
P r-
b
I l T 1—1 l l
O O
O O O O O O m
'2 9. "’ .“2 2
bquIeuIeJ N g, bugugeum d 95

NDJFMAMJJASONDJ
date

1963

1984

the % N and % P in the 1983 litterbag studies.

1n

Changes

Figure 2.3.

32

1.0

-'Z

See

-o
._(n
- <
~‘fi

.,

32:3 29¢

- I

-<
r—z

I-LL'

"'80-

~00

2:!

Too-

lon—

ugugawe: d 95

I
O
a)

[06'

tom-

 

loom

dugugetuei N 95

.~.o.ucoo.

 

 

 

 

 

33 v2: n3.
nonm<sss<ausoz
rLr_.____..._.o

[on
.8.
683: .- Ion.
concerto: e
neoceaaae e
:03
F? b . p b h p F L p b p b 0
r3
TOO—
you.
5:93:30 1 3...
Sagacenm
roam

.m.m gunman

Ougugewea N 95

Ougugewea d 95

Carex rostrata

 

 

.2
E
e
a
a
a
X
o
h
a
0
j: l l
O
3 3 n
v- '-

fiugugewei N 95

 

33

.,
[a
)-2
Lo
..m
—(

t-fi

 

 

I l
8 8 °

DugugauieJ d a,

773

'-Z

i—(fl

-<

—s

1004

 

l l I l l
N D J 5 II A I4

1983

 

(WMMMUN %

150—

100-
50
O

Dugugewei d 95

date

(Cont'd.).

Figure 2.3.

34

of Carex aquatilis accumulated P in excess of the initial

 

content. There were fewer (11) between species differences
in the %P remaining in the flooded vs. the non-flooded
treatments (18) (Table 2.3). By December, only Typha

latifolia and Carex aquatilis (both treatments), and the

 

 

non-flooded treatments of Carex stricta and Carex rostrata

 

retained P in excess of initial values. The %P remaining
during September and December, as with percent weight and
%N remaining, were significant for most treatment

differences (Table 2.2).

1980-1982 studies

 

The trends found during the first year of the long
term studies initiated between 1980 and 1982 are generally
similar to the 1983 studies (Figure 2.4). Initial weight
loss from the 1980-82 studies were somewhat greater than
the 1983-84 studies (Table 2.5) possibly due to the larger
mesh size used in the earlier studies. Subsequent to
initial losses, nitrogen and phosphorus accumulations
occurred during the spring and summer months. The initial
nutrient decline, for most species, was greater than the

initial weight loss. Except for Carex stricta litter, the

 

immobilization of nitrogen exceeded P immobilization, at
least initially. In the latter stages of decay (after 12
months) the nutrient losses began to exceed or parallel

weight loss.

35

.mmmnnmuufig mmmF1ommP ch

mcflcflmewu mmmE m m can mmme z w .unmwm3 w mzu Ca mmmcmnu .v.~ musmfih

 

 

 

 

 

eaeo
coup now? «our
Iclmwozoudswldiua-l
b-Pprppth...rrrp -
Ion
% V %
8 18— H
e e
... m
M :8. M.
u u
8 6
Icon
«Incubate... ManotueEEeo f can £=.e:ee 330 I a..."
300 Sea
000' «no. pco— 000' «cap poop coop
Idluw-IouCQQIdluan..u¢aal¢l.a.l Iclmfinxoutsnlnlu sol-n1 Haidlmsnt
FFPP-PF_PpPpP-pb.bppp.prrpppL_ .— I.
Ian In
an
I: I: H
a. m
H 0
Ian m 9.2.2:».ele I2. m.
I neg-632$. W
W... usages... .3 fl 4 5
1.2m. :8.
6
13- 83.2.! .330 I as—
203» .830 f3—

 

   

 

 

 

.36

.A.p.u:oov .v.m musmwm

0.00

NDG, .Oa—OOG-
loucquldlnQDIOncqwlclsqnl

 

 

 

    

LlELI—lEEEI—l—thltlile
Inn
%
w
N
I w
a 1:. .w
I'—
Eenseubae Sic-acne” —1
an.
Sen
3. v3. 32 «oo—
Iclgqnlcn¢96I¢I5QII
a ...F..L_bubbkp...b .
. nu I 8
u
0 e
W 05:36.... a I e w
v 2. fl. ocEfiEo. 2 I 4 I 3. w.
m. 0552:: .l i e N
u m
a
I 8. 1 .N
r $2 2.5:: «33.5% fl
32:2 23: .c

 

 

37

 

 

 

 

 

 

 

 

 

 

 

xuo mpw.o ov¢.o mflmcmomcmo mfiumoummemamo
xuc Pmo.o opo.P mumuumou xmumo
cmooon mnm.o mFm.o mflaflumamm xoumu
who mv¢.o mmn.o muofiuum xmumu
nmcoon mno.P omo.~ mscfldm> msmfluom
>p© mpm.o omm.o mfiHOMflumH mnmma
nwcoon FmP.P hom.P Esmumowusm Esmcwmummm
>u© mom.o mmo.P ESQWmUNusw Enacmwummm
ucmEumwuu mmmp Nmnowmp mmwummm
..>mow© mo mama P uwumm Gmumasoamo xv mmmp
cam 0mm. cmwsuwn cwuosccoo mwfiosum mmnumuufla
uOm .umm>\mufics xv mucmwofiwmmoo coHufimoaeoomo .m.~ magma

38

DISCUSSION

Weight loss

 

The weight loss during the overwintering period was
probably due to the leaching of soluble compounds from the
litter as well as the fragmentation and loss of litter
particles smaller than the mesh size. Fragmentation was
likely of minor importance for species with relatively

coarse litter (such as Typha, §parganium, Scirpus, and

 

Carex rostrata). These litter types appeared resistant to
phySical breakage.‘ Rates of weight loss increased
substantially during the summer months. By this time the
litter was soft and easily fragmented suggesting
particulate losses were potentially an important component
of this weight loss. The annual rates of decomposition
were generally similar to literature values (Brinson et
al., 1981).

Puriveth (1980) described two components to weight
loss. The first being the leaching of soluble organic
compounds during the winter months, followed by the
physical and biological breakdown of plant litter occurring
during the spring and summer. Initial leaching in these
studies was somewhat smaller than literature values
(Puriveth, 1980; Davis and van der Valk, 1978) probably
because of different phenological or physiological states

of the initial litter sample. For instance, the varing

39

degrees of senecence (between species and years) or
exposure to rain (between years) could affect the pool
readily leached compounds present in litter at the start of
the studies.

Typha and Carex aguatilis were the only species to
have a greater weight loss in the dry treatment versus the
flooded treatment (Figure 2.1). This anomaly may be
related to differences in the dissolved 02 content of the
water column at the treatment sites for these species.
These litterbags were placed in flooded stands
approximately two hundred meters from the river channels.
In these peripheral areas, the surface water was stagnant
or exchanged very slowly with O2 saturated water of the
river channels. The shallow water column in the stands was
low in dissolved O2 (<1.5 mg L-1) probably due to the
biological oxygen demand of the submersed litter and the
organic sediments. In emergent stands adjacent to the
river channels 02 concentrations were 4 - 8 mg L'1.
Nutrient inputs to the peripheral areas from streamflow are
likely low, and the exogenous nutrient supply could limit
decay processes in these areas.

Oxygen availability has been shown to control
decomposition processes and under anaerobic conditions, the
decomposition of dissolved organic carbon (DOC) is reduced
(Godshalk and Wetzel, 1978). DOC lost from litter is less
likely to be mineralized under anaerobic conditions and

decay estimates in anaerobic sites (such as the Typha and

4O

Carex aquatilis litterbags) are more likely to be over

 

estimates than those in aerobic sites.

During the three month period that the non-flooded
treatment was inundated, the differences in decay rates
could be related to oxygen levels as discussed above.
During non-flooded periods, moisture and/or nutrient
availability likely reduced decomposition rates in the
non-flooded treatments. Flood water provides litter and
associated microbial populations with the inorganic
nutrients required for mineralization of nutrient poor
substrates. Nutrient immobilization in non-flooded
environments must depend on atmospheric precipitation and
the diffusion of nutrients to the litter from the soils.
These inputs may not proceed fast enough to meet microbial
nutrient requirements, imposing nutrient limitations on
decay processes.

For most species, a substantial amount of the energy
stored in the litter component is mineralized, lost as
dissolved organic carbon, or particulate organic matter
after one year of decay. In the aerobic, flooded stands
adjacent to the river channels, mineralization can be rapid
and accumulations of organic matter and nutrients in the
sediments are likely to be slow. In non-flooded
environments, after as long as 2.5 years, as much as 25% of
the original litter mass remained (Figure 2.4). In these
stands or under anaerobic conditions, slower rates of

mineralization lead to greater annual accumulation of

41

organic matter, nitrogen and phosphorus in soils.

Nutrient changes
The accumulation of the phosphorus concentrations in

suspended litterbags of Typha and Carex aquatilis may be

 

the result of precipitation inputs (perhaps associated with
the accumulation of snow around and over the bags). Tissue
nutrient concentrations declined in the litter of other
species during the winter months (Figure 2.2). The percent
nutrient decline was larger than the weight loss during the
same period, and probably represents the leaching of
soluble compounds from the litter.

The nitrogen and phosphorus concentrations of the
litter of all species and treatments increased during the
growing season with the highest nutrient concentrations
generally occurring in species and treatments with the
highest decay rates (such as Sparganium eurycarpum and
Carex rostrata). In most samples, the nitrogen and
phosphorus mass of the litter increased over minimum
levels, indicating an immobilization of these nutrients
from the environment. This immobilization generally
resulted in a nutrient mass exceeding initial levels.
Immobilization was greatest in species and treatments with
relatively slow rates of weight loss. Over the study
period, a net retention of nitrogen and phosphorus in
litter biomass is indicated as the percentage of these

nutrients remaining always exceeded the percentage weight

42
remaining.

Mellilo et al. (1982) have identified a three
component model describing the nutrient dynamics in litter
(leaching, accumulation, and release phases). These data
support this model, but the studies presented here appear
to have been terminated during the release phase.

Decomposition rates in these studies did not appear to
be closely linked to initial nitrogen and phosphorus
levels. The lack of correlation to nutrient content was
likely due to an overwintering period unfavorable to rapid
microbial growth which permitted the leaching of N and P.
Leaching during the winter and early spring months reduced
the mean tissue nutrient concentrations by as much as 70%.
Leaching was greatest for species with the greatest initial
nitrogen or phosphorus content and as a result, the spring
nutrient concentrations were quite different from those of
the preceeding autumn. Between species variation in the
spring nutrient concentrations were smaller than those of
the fall.

Other studies have demonstrated that decomposition and
mineralization rates are dependent upon initial nutrient
content (especially C:N ratios) (Godshalk and Wetzel, 1978;
Coulson and Butterfield, 1978; Berg and Ekbohm, 1983).
Those species with low C:N ratios undergo greater
mineralization than species with higher C:N ratios. The
initial conditions in this study apparently did not favor

microbial colonization. The importance of the initial

43
nitrogen status of the litter may have been negated by
leaching.

The nutrient content of decomposing plant litter is
controlled in part by micrbial populations as well as
physical-chemical processes. In nutrient poor litter,
initial increases in the N and P content of litter is
typically attributed to the accumulation of microbial
proteins (Puriveth, 1980; Godshalk and Wetzel, 1978).
Substrate nutrients are retained by microbes as carbon is
mineralized resulting in increased nutrient concentrations
in the litter. When substrate nutrient levels are
insufficient to sustain microbial requirements, additional
immobilization from the environment occurs and the nutrient
content of the litter is further increased, frequently
above the initial content (Berg and Staff, 1981). Nitrogen
may also be immobilized independent of microbial
utilization through the formation of humus compounds
(Melillo et al., 1984).

For most species, rates of nutrient immobilization
were greatest during the early growing season when the
plant nutrient demands were greatest. The immobilization
of nitrogen and phosphorus during the spring and early
summer suggests that a competition between the producer and
microbial populations may exist. This competition could
increase any existing N or P limitations on primary
production, reducing overall marsh production. The

greatest nitrogen and phosphorus mineralization occurred in

44

mid-summer when retention by macrophytes was less likely.
In summary, decomposition rates vary as a result of
interspecifc differences and environmental factors.
Decomposition rates are greater in the emergent zones.
Under typical field conditions, for the species studied,

decomposition of Sparganium > Scirpus > Calamagrostis >

 

 

Carex rostrata > Carex aquatilis > Carex stricta > Typh .

 

 

Nutrient losses from litterbags followed a similar ranking.
The reduced litter and nutrient turnover in the meadow
communities suggest that this vegetation type is more
important to the storage and accumulation of nutrients in
the marsh than most of the emergent stands. Factors such
as low water levels, which favor developement of meadow
over emergent communities, will alter ecosystem function by

altering decay processes.

III. PLANT PRODUCTION AND PLANT NUTRIENT CYCLES

INTRODUCTION

Emergent plant stands are typically some of the most
productive temperate plant communities (Keefe, 1977;
Brinson et al., 1981) and as such, they likely play a
important role in the nutrient cycles of the marshes they
inhabit. The role of plant nutrient uptake and release
patterns in controlling the water quality of a marsh is
dependent upon the hydrologic flux of nutrients relative to
the total nutrient flux through the vegetation components.
Few estimates of nutrient cycling by emergent macrophytes
have been made for a marsh basin with concurrent
measurements of hydrologic flux.

Production processes, nutrient uptake rates, and the
nutrient standing stocks in freshwater wetland plants are
probably some of the better known facets of marsh
ecosystems. Emergent plants generally derive their
nutrients from the rooting substrate and translocate them
to storage and shoot tissues for growth processes.
Considerable amounts of N and P accumulate in emergent
plant biomass during the growing season. At peak standing

1

crop, values range from 40-460 kg ha' for N and 2 to 27

1

kg ha' for P (e.g., Auclair, 1982, Auclair et al. 1976;

45

46
de la Cruz and Gabriel, 1974; Lindsley et al., 1977; Boyd,
1978; Davis and Van der Valk, 1978; Hopkinson and
Schubauer, 1984; Klopatek, 1975, 1978; Odum and Heywood,
1978; Prentki et al. 1978; Richardson et al., 1978; Simpson
et al., 1978; Ulrich, 1984; Ulrich and Burton, 1985).

At the time of plant senescense, the nutrient content
of shoots is reduced from a late summer peak due to foliar
leaching and the translocation (Shaver and Melillo, 1984)
of nutrients to overwintering root and rhizome tissues.
Still, upon death, a considerable quantity of both N and P
are transferred to the litter component. The decay of this
litter results in the release of nutrients to either the
water column or the substrate surface.

Submerged plant species often derive much of their
nutrient requirement from sediments (Barko and Smart, 1983;
Bristow and Whitcombe, 1971; Wallsten, 1980). These
nutrients are frequently released to the surrounding
surface waters. Because of this cycling pattern, submerged
vegetation has been thought to function as a "nutrient
pump" (McRoy, et al., 1972; Wallsten, 1980). The
importance of submerged vegetation to nutrient cycling in
lakes and reservoirs is recognized (Boyd, 1971; Wetzel,
1983). Emergent macrophytes also derive their nutrients
from the substrate, and the nutrients leached from living
plant shoots as well as the leachate derived from litter
may enter surface waters directly. Thus, emergent

communities also function as "nutrient pumps" (Klopatek,

47
1978; Hopkinson and Schubaur, 1984), transferring nutrients
from the rhiZOSphere to shoot standing crop, and ultimately
releasing a portion of them to the surface water or
sediment surface.

The functioning of emergent plants as nutrient pumps
is largely dependent upon rates of nutrient mineralization.
If decomposition rates are slow, much of the annual
nutrient uptake may be returned to the sediments.

Microbial populations and humus formation can result in the
immobilization of nutrients from the water or soil for long
periods of time. This nutrient accumulation in litter is a
retention mechanism and suggests that the concept of
emergent vegetation acting as "nutrient pumps" may be too
general.

Variations in the hydrologic regime occur in many
wetland ecosystems and are to varying degrees
characteristic of most marshes. The hydrologic regime of a
wetland includes the source, retention, velocity and timing
of water movements through the system (Gosselink and
Turner, 1978). Especially important to aquatic vegetation
is the period of inundation, as this factor controls many
of the chemical and physical properties of the substrate.
These properties include 02 concentrations, the chemical
species of nutrients present, nutrient availability, and
and the presence of toxins such as H28 (Gosslink and
Turner, 1978).

The chemical properties of flooded soils require

48

special morphological and physiological adaptations of the
plants colonizing them (Williams and Barber, 1961; Fitter,
1981; Smirnoff and Crawford, 1983). Since these
adaptations may not be suited for successful growth in
non—flooded areas, the composition and structure of aquatic
communities is closely linked to the nature and extent of
water level fluctuations (Sculthorpe, 1967). Changing
environmental conditions brought on by flooding or
drainage, impose changes in community structure which can
alter the rates and patterns of energy flow and nutrient
cycling. For example, waterlevel changes may effect
production and decOmposition rates by altering nutrient,
moisture, and oxygen availability between sites (Klopatek,
1975; Auclair, 1979; Brinson et al., 1981). In Great Lakes
marshes, the potential effect of hydrologic fluctuations on
ecosystem properties have not been throroughly evaluated.

Water level fluctuation of Lake Michigan can be as
great as 1.75 m over 10-12 year periods (Figure 3.1).
During the early 1960's, Lake Michigan water levels were
much below normal, and in 1964 the second lowest annual
mean water level was recorded. From 1965 to 1973, the
water levels of the lake increased and the 1973 mean water
level is the historic maximum (NCAA/NOS, 1984). As a
result of the low relief of the marsh basin and an
elevation close to the mean lake level, large areas of
marsh are drained or submerged by these water level

changes.

49

.mfimxamcm

oflLQMHGODOLQ m0 mumm> mumoflpcfi m3ouu<

.mcmwe xacucoe mo mocmu mcu mumofipcfl numb

Hmogunw> .xemmP .moz\<<oz. mmmw-owmr .Acmme
Hmsccm. mam>ma umumz Gamecon wxma :fl momcmno F.m musmam

 

 

 

m<m>
no? 82 m3. 2.: 32 82
P — P b . _
Ind:
“M
m
13: 3
H
1
9 a
A
caoE '''''''''''''''' II J a
_V .I
13:
)
W
a
I.
a
1
ll 9
a » a as: I.

 

nfihw

4w>mn_mm._.<>> Z<0_IO=)_ mx<._

50

VEGETATION OF PENTWATER MARSH

The aquatic vegetation of Pentwater Marsh is
characterized by recognizable species assemblages that
appear to be primarily related to the inundation regime and
amount of soil moisture. The relationships of these
communities to the hydrologic regime in wetlands of the
Great Lakes region of North America have been discussed by
Harris (1977), Van der Valk (1980, 1979), and Tessier et
al. (1984).

In Pentwater Marsh, the major vegetation zones consist
of submergent, emergent, wet meadow, and shrub communities
(Figure 1.2). The submergent plant beds are composed

primarily of Potamogeton crispus L., Anacharis canadensis

 

(Michx.) Rich., Myriophyllum spicatum L., with the floating

 

leaved Nuphar advena (Ait.) Ait. f. and Nymphaea odorata

 

Ait. occasionally dominate. Monodominant emergent plant
stands of Sparganium eurycarpum occur in water 10-50 cm
deep, and in some shallow areas, these stands sometimes mix

with Scirpus validus. Monodominant stands of Scirpus are

 

found along some of the old river channels. Typha
latifolia generally occurs as a codominant with other
emergent species and/or meadow species in transitional
areas. Meadow communities are found in more peripheral
areas that usually flood only during periods of above~
normal water levels. These communities consist of £3325

aquatilis - g; rostrata stands, Carex stricta -

 

 

51
Calamagrostis canadensis stands and monodominant
Calamagrostis stands. On elevated sites which rarely

flood, shrub communities consisting of Alnus rugosa and

 

Salix spp. dominate.

OBJECTIVE

Vegetation studies were designed to determine the N
and P uptake and storage patterns in emergent and wet
meadow biomass. These processes can then be compared to
overall flux of these nutrients through the marsh. The
general hypothesis was that the seasonal nutrient uptake
and release patterns by the marsh vegetation are important
to the storage and flux of N and P through the marsh. It
was expected that the quantities of nutrients cycled
annually in vegetation would represent a large percentage
of the nutrient retention or export by the marsh.

The nutrient transfers identified in this study were
root/rhizome storage, aboveground assimilation and release
from litter biomass. The seasonal variations in nutrient
concentrations in shoot and root/rhizome biomass were
identified as well as the seasonal importance of nutrient
storage in the various biomass components.

An additional objective was to evaluate the effect of
Lake Michigan water level fluctuations on plant zonation in
Pentwater Marsh. This information was then used to

estimate the effects of hydrologic disturbances on the

52
quantity of N and P stored in the shoot and litter biomass

of the emergent and wet meadow zones.

METHODS

Field procedures

 

Plant biomass studies were restricted to stands of
emergent and wet meadow vegetation. These zones
represented a large proportion of the marsh basin (Figure
1.2), and since they are typically some of the most
productive wetland communities, their contribution to the
total organic matter production of the marsh basin was
likely to be proportionately greater than the submerged and
shrub zones (Wetzel, 1983; Westlake, 1982; van der Valk and
Bliss, 1971).

Emergent vegetation occurs on substrates flooded
during normal water levels. During above-average water
levels much of the meadow stands are inundated. These
inundated stands interact chemically with the water column
of the marsh basin. Other life forms such as floating
leaved and submergent macrophytes, because of their
physiology and habit, interact more intimately with the
water column but they were not studied because of their
relatively limited occurrence in Pentwater Marsh. Since
the shrub communities of the marsh are situated in areas
without direct interaction with the surface waters, they

were excluded from this study.

53

Community composition, shoot and litter biomass, and
the N and P pools in the living plant biomass of these
communities were quantified from field studies conducted
between 1980 and 1983, and with the aid of aerial
photographs taken during the 1982 and 1983 growing seasons.
During the 1980 and 1981 growing seasons, plant biomass was
sampled randomly along 8 transects crossing the marsh basin
perpendicular to the river channels (Figure 3.2). These
transects were selected to sample the central marsh basin
as well as the upstream areas along the North and South
Branches of the Pentwater River. Along each transect, 1O
sampling points were randomly established in areas of
emergent or meadow vegetation. Monthly, 0.25 m2 plots
were harvested within 100 m2 of the permanent points.

The sampling procedure involved carefully setting the
quadrat frame over the vegetation and removing all stems
rooted within the area of the plot. Live and standing
litter biomass was clipped at the substrate surface.

Before fallen litter was removed, the perimeter of the plot
was clipped so that the fallen litter removed represented
only that which occurred within the area of the plot.
Litter fragments as small as 2 cm were collected from the
sediment or water surface. The plot samples were bagged
and transported to the laboratory for further analysis.

2, centered on the plot

Once sampled, an area 1 m
location, was excluded from further sampling.

The transect locations were altered during subsequent

54

Bus 3:

 

 

Lannie” Crook '

Figure 3.2. Location of transects sampled for shoot and
litter biomass during 1980 and 1981.

55
years to take advantage of the strong patterning of the
vegetation types and to increase the sampling of specific
plant communities. Sampling during 1981 — 1983 was
conducted in six major community types identified from the
previous years data. A total of 65 samples were collected

in stands dominated by Sparganium eurycarpum, Typha

 

latifolia, Scirpus validus, Carex stricta, Carex aquatilis
or Calamagrostis canadensis. Five plots were randomly
selected within a 10 m radius of points randomly
established along each of two 100 m transects (Typha was
sampled along three transects) in two stands of each
vegetation type (Figure 3.3). The stands were visually
selected to be representative of each community type.
Samples were obtained approximately monthly from the time
of shoot emergence (early-May) through leaf senescence
(mid-October).

These sampling procedures included most of the site
variation apparent in the marsh. For instance, most
transects ran perpendicular to the river channel and thus
crossed gradients of water depth and substrate texture.
This sampling procedure did reduce sampling effort in the
ecotonal areas between community types, but permitted
reliable extrapolation of stand biomass data to the entire
marsh basin based on estimates of community coverage
obtained from aerial imagery.

In October of 1981, and monthly through 1982 and 1983,

root cores were taken concurrently, and centered in each

56

   
  

Pontwalev Lake

Long Budq
"d- BUS 31

Vegelanon Typo

 
  

Spugamum
500 molars

   

Calamagrostis
Cnrox stricta

bun;

Carox aquatilis

5 Tynhl
e Scivpun

anbricks Croat I

Figure 3.3. Location of transects sampled for shoot and
litter biomass during 1982 and 1983.

S7

plot after shoot and litter samples had been collected.
Root cores were taken with a 15.2 cm diameter aluminum
irrigation pipe. The sampling end was sharpened and
serrated to cut cleanly through the root/rhizome mat as the
core was twisted into the substrate. Roots were sampled to
a depth of 40 cm including nearly 100% of the root biomass
(most roots and rhizomes occurred within the top 20 cm of
the sediments). As with the shoot biomass samples,
root/rhizome samples were stored at 1-30 C until further

processing.

Laboratory procedures:

Shoot and litter samples were sorted by species and
dried to a constant weight at 65 oC. Shoot samples were
classified as litter when they were entirely senescent, or
when only 3-5 cm of the stem or leaf base remained green.
Root cores were washed free of soil particles over a 1
mm2 sieve and live materials (roots, rhizomes, stem
bases) were separated from dead material based on visual
appearance. Live belowground tissue was identified by it's
light color, firm and fresh appearance (as opposed to the
softer, discolored dead material).

Sorted biomass samples (if less than ca. 50 g) or a
representative subsample (for larger samples) were ground
in a Wiley mill to pass a 1 mm2 sieve and retained for
nutrient analysis. The ground tissue was thoroughly mixed

and a 0.2 g subsample was digested using a modified

58
micro-Kjeldahl digestion (Nelson and Sommers, 1973; Ulrich,
1984). Lithium sulfate was substituted for potassium
sulfate, and included in the digestion as a reducing step
which converted the nitrites and nitrates to ammonia to
include them in the analysis. Nitrogen and phosphorus were
determined colorimetrically using a Technicon Autoanalyzer
II system. Nitrogen was determined by the phenate method
(Method 351.1, U.S. EPA, 1979), and phosphorus was
determined with an ascorbic acid method (Method 365.1, U.S.
EPA, 1979). The digestion procedures and nutrient assays
were checked using U.S. National Bureau of Standards'
standard plant material (pine needles) of known elemental
composition as a control. The procedures resulted in
average recoveries of 96% N (sx = 2.1%) and 103% P (sx =

1.8%).

Image analysis and community mapping

 

 

Aerial photographs and standard remote sensing
techniques were used to document the 1983 coverage of the
six vegetation types and to estimate the changes in
community distribution over a 17 year period (Table 3.1).
The large scale and exceptional quality of the 1983 images,
coupled with the opportunity for ground verification
permitted detailed vegetation mapping for this year.
Signature characteristics based on color, texture, tone,
and height (useful with stereo images) were developed for

each community type from true color transparencies (Table

59

 

mmfiocmumamcmuu uoHoo “mHU ooo.wmuP hash

 

wwwocwummmcmnu HOHOO «mHU ooowwup wash
mwflocmumamcmuu “CHOU «mHU oop.vup >H5h mmme
mmfiocmumamcmuu uoHoo «mHO oop.vuP wash Nome
wwflocmummmcmuu mHU ooo.omur umnEmuQmm wrap
moflocmumamcmuu mHO ooo.mmuP uwQODUO mum?
mmfiocmummmcmuu mHU ooo.oour HODEODQOm mom?
magnum z a m ooo.o~"P sass momP
wows mmmsH mamom nucoz “mow

 

.Aomumuucfl uoHoo u mHU. mama
coflumuwmo> wo coflumumamua new 0mm: >LQMHOODO£Q Hmwumd .p.m magma

60

 

 

 

 

 

 

 

 

 

 

o :uOOEm combo Esflpme ©w>mma mcfiumon
cwwumnxwum xumo
o nuoosm unsoun “cmoum (Esfipme ucmmumensm
xoman ucmmumensm
o nDOOEm “c3oun xumo umumz \Hmumz
cwwum
EFV wcflw I3OHHO> unwed mflumoummemHmU
cwwum
Epv mafia (mafia Esflome mflawumsmw 4m.
cmwum 300mm:
E—v mcflw nsoHHm> Esflpwe muofluum xwumu um:
comma
EPA snapwe u>wum Esflome mnmxe
Esflbme csoun Enacmmummm
EPV umcflu Ismflcmwum xumn umsmuflom
Esfipwe cwmumu3ofiam> xHMU
EPV nwcflm “compo (endows Esflcmmummm ucmmumem
Emem: mmaexme mam mzoe wows mm>ou mzoN
.mmfiocwumawcmuu oow.mup .HoHou msuu meme
scum cmflmflucwpfl mmaxu um>oo pcmHum3 mo moaumflumuomumnu musumcmfim .m.m magma

61
3.2). The signature characteristics were modified using
1983 color infrared (CIR) transparencies for use in the
analysis of older CIR images (1969, 1975, 1978). Image
scale and quality as well as the lack of ground
verification limited the analysis of the historical
photographs to mapping only the major vegetation zones
(shrub, wet meadow, emergent, open water/submergent) rather
than specific community types.

Transparencies were enlarged 5 - 10 times (to a scale
of 1:6,400) with a Bausch and Lomb variable magnification
tracing projector. This scale was adequate to map the
smallest units desired (ca. 50 m2). The images were
traced onto mylar sheets and the resulting map was
digitized using a Calcomp electronic digitizing table and
an Earth Resources Digital Analysis System (ERDAS) computer
and software package. The area of each classification unit
was calculated from the digital data. The plot samples
obtained throughout the marsh served as ground truth data
for the 1983 aerial photographs but additional field
reconnaissance was obtained for selected areas by a visual
determination of community type based on the species
present. The map designations were checked against the

field observations and corrected as necessary.

Data Analysis

 

Mean standing crOp of the biomass components and the N

and P masses in plant biomass were calculated for the

62

quadrat samples, root core data, and litterbag

decomposition studies. These data were then used to

estimate the annual production and seasonal fluxes of

nutrients in the communities sampled. The following

equations were employed:

[1] Net production = (maximum biomass) - (minimum
biomass).

[2] Accumulation of nutrients in aboveground biomass =
(maximum nutrient stock in above ground tissue).

[3] Over-wintering losses of N and P to soil = (autumn
nutrient mass in roots) - (spring nutrient mass in
roots).

[4] Annual uptake of nutrients from soil = (maximum
nutrient stock in live biomass) — (minimum nutrient

stock in live biomass).

[5] Losses from leaching and mineralization = (initial
nutrient mass in litter) - (final nutrient mass in
litter).

The percent of the original mass and nutrients
remaining in the litterbags (described in Chapter 2) were
used to estimate releases of N and P from the September
living biomass. Calculations were made using litterbag
estimates at 8, 12 and 21 months after plant senescence
(Table 3.3). The nutrient flux calculated for the litter
component thus represented changes occurring in a single
years' litter production. Estimates of remaining N and P
masses were made individually for each of the major species
contributing to the shoot or litter biomass of each
community type and the calculated sum thus represents an

average reflecting the species composition of each

 

 

 

 

 

 

 

 

 

mm mm om, mm? oer mm mscnam> msmwnum
mo mm Pa mm, PmF om anaemfiuma mamas
mm an em we, ma ma mflaflumswm xmumo
om mm mm mm ma Pm mumuumou xmumo
ma we om? mm mm we muonuum xmumo
sq Pm me we om mo? mnmcmcmcmo mfiumoummsmamo
he em mm me one OPP emmumowusm Esficmmummm
aw zw aw zm my 2m

mzns mmmzmemmm quma mmHommm

 

.mam>umucfl LDCOE Pm can we
mmmnuwuufla :fi mcficflmewu mmmE m m com 2 w umnewuamm com: .m.m OHDME

.m umumm .m umbamno.

64

community type.

RESULTS

Plant biomass

 

The emergent and wet meadow vegetation zones accounted
for 61% of the marsh basin and 69% of the marsh area
dominated by vascular plants (Table 3.4). About 86% of the
area dominated by non-woody macrophytes was included in
this study. Of the six emergent and meadow communities
studied, Carex stricta covered the largest area of the

marsh (31 ha) while Calamagrostis canadensis was the most

 

restricted cover type (3 ha). Ranked in terms of total

biomass production, in 1982, Carex stricta was the.most

 

important community producing 42%, of the total August
shoot biomass (Table 3.5). Sparganium eurycarpum stands
were the most important emergent community, representing
about 22% of the total production.

Living shoot biomass increased rapidly from near zero

values in May to maximum values of 430 to 725 gm m2

by
mid summer (Figure 3.4, Table 3.6). The August biomass
peak was significantly different (p <0.05) from other
monthly values only in Sparganium and Typha communities.
In the other communities summer (July — September)
measurements were only significantly different from the

late spring and fall samples. Carex stricta was the only

August sample mean significantly (p <0.05) different from

 

 

 

 

 

 

 

 

<mm< AdBOB

 

mm mm azmwmmzmam\mma<3
he om .QQm xwamm .mmomsu mocH< mommm
mm 9v sconce um; Hmuoe
m m mflmcmpmcmo mmumoummemHmu
m h mflaaumawm xwumo
om Pm muuHuum xwumo zoadmz an:
mm «m ucmmumew Hmuoe
v m Esflcmmwmmmnmsmuflom
w or MAHOMADMH mnmNH
a? m— Esmumomusw Eswcmmummw Bzmummzm
w .9: .
¢mm< mmwa weHzazzoo mzoN

 

.Amwmpv smumz umumzucmm cfl
mmfluflcseeoo ucmHQ ucmcflEOU mcu mo mommm>oo pmumefiumm .v.m OHDMB

66

 

vhv

«mm

mme

A -e.m. aaeoe

 

 

 

 

 

 

 

 

N

0.9 h.m F.N v.9 Hmnuo
o.a m.m m.w P.v mumuumou xmumu
m.m m.v m.~ N.P mandam> msmufium
$.99 o.m m.mp or mflaflumswm xmumo
m.m— o.PF ~.n m.o mfiHOHADMH MLQNB
N.oP w.mw m.mP P.mp mwmcmpmcmo mflumoummemHmo
P.mm m.m~ m.~m m.¢m Esmumomusw Esficmmummm
v.—~ m.mm v.P~ o.mm muofluum anmu
mam, mam? Pam. mwnommm

 

.immm.-omm.. mmflomam “ohms a man an mmmeofln noonm

Eseflxme Hmuou on» on cofiusnfluucoo mmmucmoumm .m.m manna

J Sparganium eurycarpum .1 Scirpus validus
d

 

 

 

 

 

TYPhO latifolia Carex aquatilis

Ll

 

 

 

 

 

Biomass (g - m'2)

 

 

 

 

 

Calamagrostis canadensis Carex stricta

9‘ 4 l
5 non j _
g - 4
3 4m14 _
a A -
E 200 _] _ _]
.2 A an
m 0 j T T l I l "

l J .l I 8 II I I

Date

 

Figure 3.4. Monthly changes in live biomass in the six
communities sampled (1982).

68

 

 

 

 

 

 

 

 

 

 

 

 

 

mm mme mom pom Pmm Fm mscAHm> msauflom
as com mos «Fm arm NF mfiaomfiumfl mnawe
am omm amp amo own a mwagumsmm xwumo
oer adwlllwdwllldmw. mmm om muonuum xmumo
c was men ems mmm as mnmcmomcmo muumoummsmHmo
sq mme are omm com m esmumosusm enucmmmmmw
eoo emmm one show mznn w<z [Nuficsssoo

 

.prumoflpcfi mounp Howucmsvwm
.ucmoflwacmam uoc
mum memos pocflaumpcsv moaaemm mmmeofln Doonm
v as mocmonuflcmfim .o.m magma

cmmzumn mocmoflwflcmflm >Hco

snnucoe mam, no .mo.

69

other community means (Table 3.7). Except for Sparganium

 

and Scirpus communities, whose peak standing crops in 1983
were 60% and 36% respectively below 1982 estimates, the
1982 biomass estimates were statistically similar to both
1981 and 1983 data.

Root and rhizome biomass was generally 3 - 4 times the
maximum shoot biomass (Table 3.8). Root biomass was

greatest in the Carex aquatilis and Typha communities. For

 

most species, root biomass appeared to increase during the
growing season but there was a large variation in the root
core data. Month to month trends for most species were not
significant (p (0.05). There was no significant variation
in belowground biomass between 1982 and 1983.

Seven plant species accounted for more than 97% of the

emergent and meadow plant biomass. In 1982, Sparganium

 

eurycarpum and Carex stricta were the most abundant species

 

and each accounted for about 29% of the total shoot biomass
(Table 3.9). Calamagrostis and Typha each contributed less
than 16% and the remaining species, about 5% each. Stands

of Sparganium and Calamagrostis were monodominant, with

 

these species contributing greater than 85% of the total
biomass to their respective communities. In the other

communities, 2 - 3 species shared dominance (Table 3.9).

Nutrient content

 

The nitrogen and phosphorus content of most species was

greatest in May (Figure 3.5, Figure 3.6, Table 3.10, and

70

Table 3.7. Significance (p <.05) of 1982 maximum
biomass between community types, and 1983
maximum biomass (see Table 2.3 for species
abbreviations).

 

1982 Maximun

 

Sp Cc Cs Ca Tl Sv
- 0 + 0 0 0 Sp
- + 0 0 0 Cc
_ + + + C5
- 0 0 Ca
’ — 0 T1

1982 vs. 1983 maximum

 

 

Sp Cc Cs Ca Tl Sv
+ + O + + O Sp
0 0 + 0 0 0 Cc
+ + O + + + Cs
0 O + O O 0 Ca
0 0 + O 0 0 T1
0 0 O 0 O + Sv

 

71

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m a, OF or m m
owe owe omv mom sea com

comm seam comm meop seam com. mannam> msmunom
m or or or m m

om. cam]; omq ore cam one

merm ommm ommm oven ooem oamm announumn «saws
m or or or m m

see one ops one? owe 0mm

ozmm can, mmmm omme Pmmv ova, mnflflumsam xmumo
m or o? o, m m

can Ola owe can mom? omm

ommr omqa one. owa_ owe, oqm muonuum xwumo
m op or or m m

oPm omm Noe can omoP omo

mmsm qmem 0mm, oaom oven olmm mhmcmcmcmo muumoummemnmo
m o, or o, m m

com com arm ommww com]; see

omam «Fem omam cram ooam Omar enammosusm asucmmummm
24m: Bmmm wba >436 mzao >42 weHzozzoo

 

mmHQEmm mmmEOHQ DOOM >H£ucoe

no imo.v a. mocmoflwgcmnm

.ic .Mm .m.
.m.m manna

72

 

 

 

 

 

 

 

 

 

m.mm m.mm m.@ 9.0 - 1 I m.Pm mscfiam> msmuuom
m.om - a.FF o.mo m.F 0.9 a.e ~.eF mnfiounuma msmwe
m.mm m.e m.m o.mv m.m~ - m.o m.mP mwafiumawm xmumo
m.om u m.P P.P u h.ho h.w~ a muowuum xmumu
v.om 1 N.o I N.m m.o m.om n mflmcmomcmo mflumoummemamu
0.00? n o.m u n u : m.mm Esmwmoxusm Esflcmmummm
qaeoe >m He mo no mo oo mm mmwe
MZON mmHUmmm NBHZDSZOU
mmaxu .Amcofiumfl>mnnnm wwflommm How N.m manna mom. mmflomam >3

>uflcoeeoo o mnu ou .mwmrv cofiusnfluucoo mmmeofln mo momucmoumm .m.m manna

73

 

 

 

 

 

 

 

 

 

 

 

 

 

33 j -
Sparganium eurycarpum Scirpus validus
23 ‘ .
z:
a?
13 a _
o I I T I I I I I I I l I
3.0 I TVPhl latifolia .. Carex aquatilis
2.0 d -I
2f
39
13 d .
o I I I I I I I I I I I I
_ Ca rox stricta
3,0 .. Calamagrostis canadenms q .
23 . .
12
a?
13 - .
o I T I I I I I I T I I I
I J J I 8 o I I J J I 8 o I
Date Date

Figure 3.5. Monthly changes in tissue nitrogen
concentrations in shoots of six major plant
species (1982).

74

 

 

 

 

 

 

 

 

 

 

 

 

 

3
.. Sparganium eurycarpum . Scirpus validus
°‘ -( q
a .3 - d
59 _ l
1 - .
o
1 I I I I I I I I r I I
TYPhl "IMO,“ Carex aquatilis
4 .1 -1
. 4
a .3 q d
g - c1
1 ‘ .
.1 q .1
0
r T I I I I I T r T I I
Calamagrostis canadensis Carex 3MP“
A . r
d d .
n. '3 " "
a? . .
1.. r
I - .
- a
0
I I I I I I I 1 1 I r’ 1
I J J I 8 0 N I J J A S 0 N
Date Date

Figure 3.6. Monthly changes in tissue P concentrations
in shoots of six major plant species (1982).

75

Table 3.11). Tissue N and P concentrations declined
rapidly throughout the growing season, with mid summer
values less than half the spring maximum values. Several
species exhibited a small increase in tissue nutrient
concentrations in October. Seasonal trends in the nutrient
content of root and rhizome tissue were less variable than
those of shoot tissue (Table 3.12 and Table 3.13). In most
communities the highest nutrient concentrations in
belowground biomass occurred during the summer, with
minimum values measured during May or June.

By combining the tissue N and P concentration data
with the estimates of plant standing crop, the nutrient
mass of the various biomass components was calculated.
During the growing season, the shoot nutrient mass
reflected an increase in plant biomass and a decline in
nutrient concentrations. Even though the tissue nutrient
concentrations declined by over 50% (Figure 3.5 and 3.6),
biomass greatly increased (Figure 3.4), so the maximum
nutrient standing crop in shoots occurred near the peak in
plant biomass. The total N content in all 6 communities
peaked with shoot biomass (Figure 3.7), while maximum P
content (Figure 3.8) in three communities (Sparganium

eurycarpum, Scirpus— Sparganium, and Carex aquatilis)

 

 

occurred in July, prior to maximum biomass. Maximum shoot

2

nitrogen mass ranged from 4.8 g m- in Carex stricta to'.

2

7.9 g m' in Scirpus validus, while the shoot phosphorus

maxima ranged from 0.6 g m-2 in Carex stricta to 2.0 g

76

 

 

 

 

 

 

 

 

 

 

m mmP.o bmm.— msmuflom
m moo.o mmh.o mnmmw
or vmo.o mmP.F mflaflumsmm hm
me weo.o mmo.P mumuumou hm
m mop.o mam.F muofluum .U
or ome.o mvo.o mflumoummsmamu
me mmF.o mmm.P Esflcmmummm aaso
or ono.o omm.P msmuwom
o nvo.o omm.P mnmmw
mm on.o mmn.P wfiaflumswm hm
w mm~.o smP.N mumuumou hm
v whmm.o amv.P mquHum .U
m mmP.o woa.o mflumOumMEmHmu
me moo.o mmm.e Esflcmmummm mean
he www.o Poc.m msmuwum
m mmm.o NN.N mammw
m vmm.o who.m mflaflumomw hm
m mmm.o For.m mumuumOH hm
m hvm.o vmo.~ muofluum .0
PP omm.o mom.m mfiumoummemamo
FF mam.o P>.m Eswcmmummm mm:
c xm z w wwwommm mama

 

.cwumz Hmum3ucmm
cfi mwaomam ucmaa ucmcflEO© h on» no mmsmmflu
boonm c3 iuzonmz sum w an. samucoo z sasucoz .o..m wanna

77

 

Q'OC’IOI‘OO

mop.o
woo.o
vmv.o
opm.o

omo.o
mmo.o
woo.o
Poo.o
mwp.o
omo.o
mmo.o

voo.o
mvN.o
Pwe.o
PFN.0
mme.o
omp.o
mor.o

moP.P

mech.o

oww.o
«Nh.o
b¢¢.o
mmh.o
mhv.w

mmh.o
mow.o
mmm.o
th.o
mem.w
mmh.o
«00.9

wwr.w
mmm.o
mor.e
mom.o
Nom.P
¢o>.o
0mm.e

msmuaom
mnmNB

mnaflumsam .o

mumuumou .U

muUHHum .U

 

 

mwumoummemamu
Ezflcmmummm

msmufiom
mnmmm
mfiwaumawm hm

mumuumou .U

ou0wuum .o

 

 

 

mflumoummEmamo
Edwcmmummw

msmuflom
mnmmw
magnumsam .o

mumuumou .U

muofiuum .U

 

 

 

mflumoummemamu
Enacmmummm

 

.Ap.ucoov

Honouuo

.uamm

umsms<

.op.m manna

78

 

 

 

 

 

 

 

 

 

 

n moo.o erm.o msmuflom
m o—o.o mmF.o mcmmw
GP aoo.o mmr.o mnauumsam hm
mp meo.o va.o mumnumou PM
m hao.o mmr.o muofiuum .0
PF Fro.o vwo.o mflumoummemHmu
me mop.o mom.o Esflcmmummw >H56
or bvo.o mov.o msmuflom
w omo.o mh~.o osmmw.
mm ooo.o oom.o mflafiumsmm MW
w moo.o mv~.o mumuumou hm
v Nmo.o hmw.o muofluum .0
m mro.o vmr.o mHumOHmMEmHmo
me mom.o mm>.o Esflcmmummm mean
he nvo.o mmm.o msmufiom
w mmo.o omm.o mnmmw
m moo.o omm.o mfiaflumsmo hm
m Noo.o m>N.o mumuumou hm
m mmo.o oom.o muofiuum .U
FF omo.o mvm.o mflumoummsmamo
or «mo.o vrm.o Eoflcmmummm >6:
: Mm m w mmwommm mama

 

.nmumz kumzucmm
:fl mwflomam ucmHQ ucmcHEOU h mnu wo wwsmmfiu
uoonm c“ Aunmfiwz >u© w Cw. ucmucoo m xacucoz .Pp.m manna

 

79

ee0.0
090.0
000.0
000.0

v90.0
mF0.0
000.0
mpo.0
090.0
000.0
000.0

090.0
m00.0
000.0
«No.0
«No.0
0P0.0
0F0.0

000.0
000.0
vor.0
00P.0
00F.0
000.0
b0P.0

hwv.0
000.0
MNP.0
00F.0
0NP.0
000.0
Pmp.0

v0~.0
NOF.0
00?.0
Nmp.0
00P.0
¢FP.0
00?.0

msmuaom
.mnmxe

mgaflumswm .o

mumuumou .0

muofluum .0

 

 

mflumoummemamu
Esflcmmummm

msmuflom

m: B
mflaaumsmm .
mumuumou .0

I

muofluum .0

 

2]

 

U

 

mfiumoummemHMU
Esflcmmwmmm

msmufiow
mnmxe

mannumnmm .o

wumuumou .0

muofluum .0

 

 

 

mfiumonmmemHm0
Eswcmmummm

 

.A©.uc00.

umnouoo

.uamm

umsmsc

.Pw.m manme

80

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.12. Nitrogen concentrations (% by dry weight) in
root/rhizome tissue during 1983.

Date Community % N sf n
May Sparganium 0.941 0.062 11
Calamagrostis 1.122 0.691 8

g; stricta 1.193 0.107 9

g; aquatilis 1.094 0.100 7

Typha 1.022 0.039 7

Scirpus 1.202 0.054 9

June Sparganium 1.037 0.082 6
Calamagrostis 1.172 0.051 5

g; stricta 0.911 0.073 9

g; aquatilis 1.003 0.034 5

Typha 1.149 0.103 5

Scirpus 1.044 0.066 5

July Sparganium 1.081 0.129 9
Calamagrostis 1.028 0.107 6

g; stricta 1.261 0.156 6

g; aquatilis 1.501 0.476 7

Typha 1.013 0.103 5

Scirpus 1.061 0.049 5

August Sparganium 0.820 0.097 9
Calamagrostis 1.198 0.187 10

C. stricta 0.900 0.083 11

g; aquatilis 0.691 0.098 9

Typha 1.023 0.094 6

Scirpus 0.802 0.091 7

September Sparganium 0.700 0.066 10
- Calamagrostis 0.959 0.076 10
g; stricta 0.078 0.054 10

g; aquatiIis 0.836 0.051 10

ypha 0.768 0.055 9

Scirpus 0.882 0.121 8

 

81

 

 

Table 3.13. Phosphorus concentrations (% by dry weight)
in root/rhizome tissue during 1983.

Date Communityy % P sf n
May Sparganium 0.119 0.017 11
Calamagrostis 0.124 0.012 8

g; stricta 0.116 0.015 9

g; aquatilis 0.121 0.022 7

Typha 0.16 0.036 7

Scirpus 0.105 0.007 9

June Sparganium 0.148 0.042 6
Calamagrostis 0.099 0.012 5

g; stricta 0.075 0.010 9

g; aquatilis ' 0.092 0.014 5

Typha 0.136 0.035 5

Scirpus 0.099 0.018 5

July Sparganium 0.165 0.025 9
Calamagrostis 0.114 0.014 6

g; stricta 0.1017 0.015 6

Q; aquatilis 0.1278 0.007 7

'Typha 0.130 0.040 5

Scirpus 0.095 0.010 5

August Sparganium 0.128 0.022 9
Calamagrostis 0.128 0.013 10

g; stricta 0.084 0.0135 11

g; aquatilis 0.124 0.017 9

Typha 0.281 0.070 6

Scirpus 0.100 0.008 11

Sept. Sparganium 0.116 0.023 10
'Calamagrostis 0.120 0.015 10
g;_stricta 0.100 0.011 10

g; aquatilis 0.093 0.015 10

Typha 0.068 0.011 9

Scirpus 0.092 0.013 9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N (9 - "1'21

 

82

Sparganium eurycarpum

 

N (9 - m")

 

Typha Iatllolia

 

I f 7

Calamagrostis canadensis

N (9 - m'z)

 

 

Figure 3.7.

Date

 

Scirpus validus

 

 

Carex aquatilis

l

 

l L _l

 

 

 

Ca rex stric ta
.1
r I I I I I
I J J A 8 II I
Date

Monthly changes in the N mass in the shoot
biomass of the six communities sampled

(1982).

83

.1 Sparganium eurycarpum '1 $607008 validus

 

 

   

 

 

Typha latifolia Carex aquatilis

 

 

   

 

 

Calamagrostis canadensis Carex stricta

 

 

 

 

 

Date Date

Figure 3.8. Monthly changes in the P mass in the shoot

biomass of the six communities sampled
(1982).

84

m"2 in Sparganium eurycarpum communities. The emergent
plant communities contributed 52% of N, and 50% of the P
distributed to shoot biomass (Table 3.14). On an areal
basis these communities covered only 26% of the marsh, as

compared to 36% for the meadow communities (Table 3.4).

Litter biomass and nutrient content

During 1982, litter biomass declined throughout the
spring and early summer and then increased to an autumnal
maximum at or after the peak in shoot biomass (Figure 3.9).
Even though the minimum litter biomass ranged from 24% to
54% of the maximum values, due to the large variability of

this biomass component, the decline was not significant in

 

Carex stricta, Scirpus and Calamagrostis stands (Table
3.15). The smallest litter biomass was found in the
emergent Scirpus and Sparganium stands.

The %N and %P content of the litter generally declined
throughout the growing season (Figure 3.10 and Figure 3.11,
Tables 3.16 and 3.17), and increased during the
overwintering period (November to May). A large decline in
the mean tissue concentration of both nutrients occurred
with the influx of new litter to the existing litter pool
at the time of plant senescence.

In most stands, the storage of nutrients in litter
(Figures 3.12 and 3.13) declined during the growing season
(Figure 3.9), and increased again in the fall following

plant senesence.

 

 

 

 

 

 

om 0e om mm A0909

MH m mflumou0memamm HH N msmufiom

a mp mfiaflumswm .o my me mzmxe
5 mm mm muofluum xmumo vm 0N asficmmummw
8

a m z w zoomms pm: a w z w bummuwem

 

.mmaxu xuaczssoo nonme xflm Gnu OD maomusnwuuum
mmmE ucmfluusc Doozm umsmsd 00 mmmucwoumm

.v—.m mHQMB

86

 

 

 

 

 

 

 

 

 

 

 

 

 

loo .1 Sparganium eurycarpum J Scirpus validus
-I -I
A .00 -( -(
o .4 d
.E 400 - ..
OI d 4
I; 200 -( .1
m 4
a d
5 I00 '4 d
o I I T I I I I I I I I I
000 -< Typha latifolia -( Carex aquatilis
d
A
.1
1
I I I I I I T
”a J Calamagrostis canadensis j Carex stric ta
«37‘ m « -
E q
.00 " d
g 4 cl
3 m '1 d
g .1 ‘ .
um- .
A cut
0 I I fl 1 I I I I I I I I
I J .I I 8 II I I J .I I 8 II I
Date Date

Figure 3.9. Monthly changes in litter biomass of the six
communities sampled (1982).

87

 

 

 

 

ace mmm map map mm? Pom msoflam> msmuflom

 

 

 

 

 

 

 

mmoP mmo owe eve omm mma mHHoMAumH mnmse
«am Pom oma mma amp omm mflaflumsam xmumu
mam cam mme Pom mmm mmm muofluum xmumo

 

«em cam mrm mam .arm mam mfimcmnmcmo mnumoummemamo

 

 

 

8mm Pom mmF em, mmm mam samumowusm sancmmumam

 

 

 

BOO Hmmm 0:0 NASH mZDH >42

 

Nwficseso0

 

.Aowumoflocfl Doc mwumo Hawucmsvwm

cmwzuwb mocmoHMHcmflm >Hco .ucmoflmacmflm

Doc mum momma cwcflaumocsv mmaasmm mmmaofln
umUbgH sasbcoe mam, Mo imo. v 6. wucmofluflcmflm

.mF.m manme

88

 

 

 

 

1.25q -
1-0 - .. Scirpus validus
g 15 q ..
Sparganium eurycarpum ..
.5] J
1 J
.25 .1 .1
-I -(
T I I I I I I I T I I T
- Typha latifolia a Carex aquatilis
Lud .

96N
P‘-
J

"I
ll
LLLL

 

 

 

 

 

 

 

 

15‘ .
J J

I I I I I I I T I I I I

125‘ . .
Calamagrosus canadensis 7 Carex stricta

-( -4
L04 3
z . J
8i J5. 2
I .J
.5. -
15‘ I
A d

I I I I I I I I I I I I

I J J A 8 o I I J J I 8 o I

Data Data

Figure 3.10. Monthly changes in tissue nitrogen
concentrations in the litter of six major
plant species (1982).

r8 9

 

 

 

 

 

 

 

 

 

 

 

 

 

3 1 . .
] Sparganium eurycarpum _[ Scupus vahdus
4 -‘ -I
T
3 4
a d
g d
1 ‘
J
a
o I T I I I I I I I I I I
Typha latifolia Carex aquatilis
.4 .. J
.3 '1 .1
g .
.2 d .1
d .1
,1 - J
.1 d
o I I I I I I I I I I I I
Calamagrostis canadensis Carex stricta
4J 1
cl d
n- .3 - d
a? J ..
.2 «J d
.1 -J
‘1 4
cl d
0 I I
I I I I I I I I I I
I J J I 8 0 I I J J I S 0 I
Date Date

Figure 3.11. Monthly changes in tissue P concentrations
in the litter of six major plant species
(1982).

 

 

 

 

 

 

 

 

 

 

 

0v m00.0 bmo.P cwDMOmcs
0 000.0 Nmm.0 msmuaom

up m00.0 000.0 mnmmw

Fe 0>0.o m0¢.0 mflawumsmm MI

0 000.0 NNm.0 mumuumou hm

m m00.0 N00.0 muoflnum .0

PP >0P.0 bm0.0 mflumoummsmHMU

Fe Fmo.o omp._ asfiammummm sass
cm 000.0 00~.P omuHOmGD

0 mmo.0 0~h.o msmuwom

0 0m9.0 000.0 650mm

mp va0.0 000.0 mflaflumsmm hm

m mmo.o m>0.0 mumuumou hm

m Nm0.0 0m0.o muofluum .0

PF 000.0 0m>.0 wflumoummEmHMU
am vac.o omm.o ssflcmmummm mass
0N m00.0 00m.r cmuuomco
me evo.o ~00.0 msmuaom

v m00.o 0mn.0 mammw

we F00.0 00m.0 mflaeumsmm MW

0 00F.0 N00.0 mumuumou .

0 m09.0 >00.F macauum .0

we 000.0 Ph0.0 mflumoummemHmU
em mao.o aro.r ssflcmmmmmw mm:
c Mm z w mwfiommm mama

 

.cmumz Hmumzucmm
CH mmaowmm ucmHQ ucmcfleov h mzu mo mmmEOHn
umuufia cfl Aunmflmz >06 m a“. 0:60:00 2 sanncoz .oP.m magma

91

 

00
PF
0P
0P

0?
0m
0m

ov
op
mm
0—

Ne

N
010000000

[\

N00.0
000.0
000.0
«FF.0
«em.0
000.0
000.0
m>0.0

000.0
«00.0
0m0.0
«0F.0
0—0.0
000.0
vmp.0
000.0

000.0
000.0
0MP.0
000.0
m00.0
000.0
009.0
0m0.0

00?.—
000.0
00m.0
«00.0
000.0
000.0
00v.0
000.0

00m.r
000.0
0mm.0
vmh.0
000.0
000.0
009.?
P00.F

N00.P
qpv.0
000.0
000.0
000.0
000.0
000.0
000.?

cwDHOmCD
msmnfium

m:
mwawumswm .
mmmmmmmm .

muoflnum .0

.131

U

 

mfiumoummemHMU
Eswcmmummw

 

omuHOmcs
msmuflom

an IE
mwaflumamm .
mmmmmmmm .

muoauum .0

I

 

U

U

 

mwumoummEmHMU
Esficmmnmmm

 

UmuHOmcs
msmuwom

m: 1%
mfiaflumsmm
mumuumou .

muofluum .0

:1

 

II;

0

 

mwumou0MEmHm0
Esflcmmwmmm

 

.Ac.uc000

HmQODUO

.uamm

umsmsd

.0P.m manna

 

 

 

 

 

 

 

 

 

 

 

 

00 000.0 000.0 00000m20

m 00P.0 00P.0 msmuflom

he 000.0 000.0 mnmxe

PP 000.0 N00.0 mfiaflummmm .0

m Nm0.0 000.0 mumuumou .0

m mpo.0 N00.0 muofluum .0

00 mPo.0 000.0 m00moummsm0m0

00 000.0 0m—.0 Esflcmmummm >090
0m P00.0 000.0 cwuHOmcs

0 >m0.0 «99.0 msmuflom

0 Pm0.0 m00.0 mnmmm

me 000.0 mm0.0 mflawumamm NW

m meo.0 m00.0 mumuumou hm

m 0m0.0 m00.0 wuofluum .0

00 000.0 mm0.0 mflumoummsmamo

mm moo.o m~0.o ss0cmmummm 0:56
mm 000.0 0~F.0 cmuHOmGD

NP mm0.0 m0—.0 mamuaom

v 0—0.0 NPP.0 mammw

00 000.0 000.0 mflaaumsmm hm

m mmo.0 ~00.0 mumuumou kw

0 m00.0 00—.0 muofluum .0

PP 090.0 N00.0 mfiumoummemHMU

em P00.0 hm~.0 Enacm0ummm xmz
c Mm m w mwflommw wumo

 

.Lmumz Hmumzucmm
00 mmwomam pcmHQ 0cmcflaoo 0 0:0 00 mmmeofln
060000 :0 A0:m0m3 >00 0 :0. 0:00:00 : 00:0:02 .ap.m m00me

 

00
PF
00
00

0—

ON
00

00
00
mm
00

NF

N
NQKDKDPKDCDM

[\

0000.0
0000.0
0000.0
N000.0
0000.0
0N00.0

0000.0
0000.0

0000.0
0000.0
0000.0
N000.0
0000.0
0000.0
0000.0
0000.0

0000.0
0000.0

000.0
0000.0
vao.0
0000.0
0000.0

000.0

0000.0

000.0
0000.0
0000.0
0000.0
0000.0

0000.0
0000.0

0000.0
0000.0
0000.0
0000.0
0000.0
0000.0

NP—.0
0000.0

000—.0
0000.0
000P.0
0000.0
0000.0
0000.0
0000.0
NO0P.0

cwuuomca
msmufiom
mnmNB
mflaflumswm .0
mumuumou .0
muofluum .0

 

 

 

 

mwumoummEmHMU

 

Esflcmmummm

cwuuomca
msmufium
m:mxe
m0000msam .o
mumuumou .0
muofluum .0

 

 

 

 

mwumowmemHMU
Esficmmummm

 

cmDuOmCD
msmuflom
mmmxm
m0000msam .o
mumuumou .0
muofluum .0

 

 

 

 

mflumoumdsmHMU
.Eswcmmummm

 

.AU.uc000

umnouoo

umnemummm

umsmsd

.0P.m manme

94

 

 

 

 

 

 

 

 

I . Sparganium eurycarpum .. Scirpus validus
J -
f o -I d
E Cl -1
' 4 d J
2 J ‘ \J/
12
20 -
d J
o I T I I I T I I T I I T
- Typha latifolia 0 Carex aquatilis
B. 0
a] all
f a d cl
E d d
6 4 .. J
5' J J
2- -
- d
o I I I I I I I I I I I I
l .. Calamagrostis canadensis . Carex stricta
d d

N (9 - m")

 

 

 

 

2- -
d -
II I I I I I I I I I I I I
I J J I 3 o N H J J I 3 0 I
Date Date

Figure 3.12. Monthly changes in the N mass in the litter
biomass of the six communities sampled

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ a
13 - . " Scirpus validus
Spargamum eurycarpum
.1
97" A _
E I .
m .5 '1 q
n. -I 0
0 J
.1 d
o I T I I I I I I I I IV I I I
0 Typha latifolia fl Carex aquatilis
1 .0 -' q
- .1
a? d
E 0
a, .5 J -
I . I
-( d
0 I I I T P I 7 l T I I r I I
Calamagrostis canadensis Carex stricta
1.0-1 J
A
Q - 0
s q -
:3 3J .
0 0
0 I I I I I I I I I I I *r I I
I J J A 8 o I I J J A 8 0 I
Date Date

Figure 3.13. Monthly changes in the P mass in the litter
biomass of the six communities sampled
(1982).

96

Image Analysis:

 

Analysis of the historical photographs of the marsh
basin indicated that considerable changes occurred in the
areal extent of the the emergent and meadow vegetation
zones (Figure 3.14 and 3.15). Wet meadow vegetation
covered 50% (59 ha) of the total marsh during the lowest
waterlevel in 1965, but decreased to 20% coverage (24 ha)
in 1978 after several years of high water (Figure 3.1).
Trends for the emergent vegetation opposed those of the wet
meadow. The emergent vegetation covered 8% of the marsh in
1965 but increased to over 40% coverage in 1978 when the
wet meadow occupied the least area. The combined coverage
of these zones has remained fairly constant at 55% - 61% of
the marsh except during the highest water level in 1975
when their coverage was 41% of the basin. The largest
areal change occurred in the wet meadow and emergent zones,
whereas the shrub zone changed the least (14 ha). The
shrub zone steadily decreased in area from low water in
1965 (34 ha) to a 1983 coverage of 20 ha. Open
water/submergent vegetation ranged from 13% of the marsh in

1965 to 35% in 1975.

DISCUSSION

Estimated shoot production in the emergent and wet

97

HSHVW dO 9”

0..

ON;

001

“001,

00..

 

 

520..

.000PI0000
.cmnmz uwumzucmm 00 meON coflumummm> momma

0:0 00 mmmuw>oo ms» :0 mmmcmco Dmumeflumm

.v..m musmflm

 

 

m<m>
omm. mum? 00:: new:
P r P _, _Inv
I:
IIIIIII ‘
.1 .....
...... .. Tow
...... o .....
:6 .8
J o:
02005239522, 4 .....
ezmommzm T
26:52 his :1: ,, on

00010 O .....

96

HSHVW :IO

98
IVU

 

 

           

modal
9.53
pzmomuzmamiupssl
‘ maazmnu
269%: 53.00
pzmummsmm
nzfiasnu

 

 

     
       
 

5v..- - _
\\\iu. :
IIHIIIIII. -. 5 .
2W WWMMW=;=
3: 0L1— I
t..—

........._
________

 

    

4

__..

I P...
.0.

.A

  

 

. 75::

l.
'
.

    
  

\\

Im¢<5 mm...<>>._.z.mn_
ZO_P<Pm0m>

 

 

.000PI000P
umumsucmm 00 meON coflumuw0w> momma 0:0 00 coflusnfiuumflo may :0 mmmcmLU

.nwumz

.00.m 005000

anion?!-

mxjs
szocusnamambss'
20¢sz

:85: ngau

. A OZSADD

000 _.
Imm<5 mmh<>>hzma
ZO_._.<._.m0m>

 

   

W
O

1"
I

1
Al
a“?
u ""1

.4.‘ '-."

     
 

 

  

       
 
 

VEGETATION
PE NTWATER MARSH
1 983

—
.

 

 

\\

“"5“.“

 
   
    
     

 
 
  

   

. u 1
q. n"
- ‘
nun: A
a-Ir- .
Illl.‘ I-HF‘
.. .
-----.
all-Illll‘jilb
Il-l-I-I'P‘lh

  

lj

mm

lmnmmumm'

.J’

lI-iy

I”Rik '—

“,In:i:l!‘!

w m!!..\ ..

W, H'”

“numb

g...

z

m.

o

5

3—3
r-2 “3
03‘ m
m E
2830”
5:532?
gmiggj
DWBUI§

   

 

‘ “mamas . i

 

 

 

 

I5:
I"

  
   
  

 

 

“5'3"!"

E,
a .

1|

  

09"
”0 v '-

m

'. I".

  

1

I
.
O

0
l
t

VEGETATION
PE NTWATER MARSH
1975

 

\\\\

    
  
  
  

'
I
I
‘
u
a
t

165.!!!

‘5
';
I
‘.
.U
I

 
 
    
 
   

my.“ ‘

—; 'u . _

' "Juli". ‘-_

:===;h1‘i

-.=-; I -.

==fiW'
l' '

unnr‘p
I

g1 {£9.19 r.
a“ ,. W.

a~fl

o
o I\

f

‘90'

I .

: I .go . . u
.0
’l
O
.I,
I .
Q

P'llt‘ .

fim}‘\\

“0"
L‘.‘fl;.”d|m“

   
   

.. —_
' ‘IIIIIIII'
‘.I...

'

Ilwf' 'fi-f‘F-J'”
mWW¢E

   

Héae= 5E

“ m

(3

E

’5

F' .,

M§
=2a~$ 2
2:53:33
amiggjc
UMEUIEI

  
    
    
    
 

 

 

(Cont'd.).

Figure 3.15.

100
meadow plant communities fall into the general ranges
reported for temperate marshes (Keefe, 1972; Penfound,
1956; Westlake, 1982). The maximum aboveground biomass
estimates measured in Pentwater Marsh are near the median
values reported in the literature for similar communities
in temperate North America (Table 3.18). Much of the
between site variations in production estimates are
probably related to a number of factors including climate,
soils, and hydrologic conditions.

Though the maximum shoot biomass may considerably
underestimate net annual shoot production for some species
(Bernard anb Gorham, 1978; Westlake, 1982), the measurement
can be an reliable estimate of net aboveground production
for others (Davis and van der Valk, 1978). The life
history strategy of the particular species will determine
the potential for underestimating net production.
Nevertheless, maximum shoot biomass provides at least a
"lower limit" of net production in temperate climates.

There are little comparative data available on root
biomass and production (Table 3.19) and the variation in
methods measuring this biomass component make comparisons
between studies more tenuous. In this study, root and
rhizomes could not be separated by species from cores
obtained in mixed species stands. Some of the monthly
variation in the data probably represents an integration of
the differential root growth of the species present.

The tissue nutrient concentration and total nutrient

‘lOl

 

muowuum

 

 

 

 

 

 

 

moms .ma>umn Nov? omq-vme xmumo
mfiaflumswm
memF .muwssom a Emnuoo ovm www-mho xwumo
owme .mwuum: ormpuhmm mwmcwvmcmo
some .umflmmwe mmm m¢5(vmo maumoummemamo
cams .mauumm mom-hes
0mm? .mammcfiuuom own IIIIIII
wear .ma>mo a me> umn cm> mom-mv~ mspuflm>
who, .smflmccaa Pam mom-mpm mamuflom
mam, .mcumwum a xwummon «was
ommr .mfluumm omme-momr
mam, .oooo a cooo one «haemauma
hem, .smfimocaq mmm mop-h~o mamxe
enas.>wdmccfla omm
wear .mfl>mo a xam> umo cm> «mos-opn
msmr .wxso cm> ass
«was ..Hm um umflmmwe mom
ommr .mfluum: mnvnpnm Esmumoxusm
PmmP .umxmm a couasm use arm-mqm snacmmummm
wusumumuflq umumzucwm wmwe
mousOm N:E.Em mmmeofim EDEmez muHGDEEOU

I

.mmSHm> wuoumumufla on swam:
uwum3ucmm qu mwsHm> cofluospoua mmmEOHn muficseeoo mo COmflquEOU .mp.m magma

102

mosmHm mamNB w
monsuaso UmNfiHfiuumw x can .m .2 Ce ESEflxme **
mowuam >Humm *

 

 

 

 

 

mum? .mxso cm> omvp-smn mangam>
sums ..Hm um smamccgq oovr-m~m comm mamuflom
mam, .mxso cm> moms—-Fma
mam, .mcummum can xmumaon use, maaoufiuma
some .goflufla somms warm mnmxa
who, .mcummum new xmummon . ssmmop
«mm, .cofiuap soon,
who? .mxsn cm> mvmp-mom
.nnmr .pumcuwm pom oumcuwm mmm Esmmwownsm
.hhms ..Hm um swamped; ommr-ocm omnm esgcmmummm
Ammcmu. AcmmE. mm>9
momDOm mmoe<mmqu mmeazezmm weHzozzoo

 

.A :6 Boy modam> musumumufla on :mums umumzucmm
nOm monamw cofiuospoud mmmEOHn xuflcseeoo mo :Omwumaeoo .mp.m wanme

103

mass in shoot tissue reported in the literature (Table
3.20) are similar to the findings of this study.
Variations in nutrient content can be due to strong
seasonal patterns, site, and or/genetic variations (Boyd,
1970, 1971).

During the growing season, an estimated total of 7,220
and 960 kg P was translocated to aboveground shoot tissue
in Pentwater Marsh. The uptake and release of N and P in
these plant communities is depicted in Figure 3.16.

The temporary storage of nutrients in shoot tissue
brings nutrients from the sediments to the marsh surface,
but, depending upon the community type and the local
environmental conditions a substantial portion of these
nutrients may be returned to the sediments as nutrient
enriched humus. During the 21 months following senesence,
anestimated 30% - 89% of these nutrients are released from
the litter biomass (Table 3.21). Some nutrient releases
are likely to continue over longer time spans, but since
litter is buried annually by litter production from
subsequent years, it gradually becomes incorporated into
environments unfavorable to decay process. Mineralized
nutrients from buried litter are also less likely to
exchange with the water column.

Translocation of nutrients to the belowground tissues
at the time of shoot senesence is a conservative mechanism
which can prevent nutrient export from marshes (Shaver and

Melillo, 1984). This study illustrates that nutrient

104

 

 

 

 

 

 

 

>Hco mmmE ucwfluusc uoonm w
moomwm mammfi a
mmusuaso pmmflafluuwm x pom .m .2 Ca Eseflxme w
Esficnmummmumnmwe +
mcflumm >Humm *
ohms .ugmao:< am.P m.m m.P o.m mannam>
m.¢ «.mm msmuflom
News .ucmsum cam cemmz a¢.P Po.mP _.P m.»
Pam? .csom F.P-~.o o.~F-o.m
qmms .coauaa wm.q o.Pm
mum, .mxso cm> am.m m.qm
who. .fixucmum m.m Pm mfiHOMHumH
ohm, .xmumaon ~.m m.Fm m.m «.mm mnmxe
wear .xoumaofix +..m ¢.o_
swap .coflufis we.m m.m~
mam? .mxso cm> m.¢ m.vF
who. .oumcumm ocm pumcumm {P.m «.mp wo.m F.m Esmumowusm
who? ..Hm um smflmocgq m.¢ k. p.m h.~m enacmmwmmm
m z m z wmmfi
mousow muoumuwufiq umum3ucmm auflCDEEOU

 

umumzucmm HOw mmmEOHQ

.A :5 so. mosam> wuaumuwufla ou smums
o>HH cw mmme m com 2 Hmuou mo GOmflummeoo

.om.m manmfi

105

Sparganium eurycarpum

LE ACHNO

4.51

 

    
    
   
 

7.96

I swam moans

LIYTEI

1.07
June

 

  

 

 

, 1)

noon-mama: mom-sf UPTAKE gm - m
27.2 0.30 2.68
(ME AN) RELEASE MlcnoaIAL MINERALIZA‘HON

UPT AK E

 

 

MOCROBML
UPTAKE

1.09

 

   
 
 
  

”(001 «can: 7
1.75

  

LIYYER

0.66

A'RIL

 

  
 
 

 

 

1

noonnmzou News: UPTAKE
009 Q55
2.54 T o“
_ (“5“") RELEASE ”figvog‘e‘l- menu‘s n

Scirpus validus
a. 12

anu “n“
4.98 233
m N
_ 1 .
aoonnmzou noun UPTAKE gm - m
24.58 2.00 1.46
(MEAN) HELIIII MRALIZA'DN

 

LEACHINO

 

    

$01007 llamas:

8.10

 
     
  

 

 

 

 

 

LEACM

1.24

 

  
 
  

2.01

l IIOOY IOOIASI

   
 

  
 

U111.

0.77

APR‘ 1

LIYVII

0.49

SEP‘I’

 

  

        

 

 

’ ‘ m - ni
noonnmzout nouns UPTAKE Q
3.67 0-28 0.23
(MEAN) anus: muemuzmou

Figure 3.16. Nitrogen and phosphorus cycling in
six emergent and meadow communities.

 

106

Carex aquatilis

“I!!!
6.18 N
SEPT

LEACflINO

* 3.87

LOYYEI
4.06
APfll

 

7.93

  

[ 511001 moans:

   

 

  

    
  

 

 

 

 

 

 

 

 

 

 

 

1 ’ 2
noormmzout menus» UPTAKE 9m ' m
30.1 2'12 2-79
‘ (MEAN) “an“ “$271.2.” MINERALIZATvou
LEACHING
311001 noun: 7 0'47
1.05
LITTER LITTER Luvs.
1 0.58 0.48 0.55 P
APHL SEPT , JUNE .
‘ - gm ' m2
noovmmzouc moms: UPTAKE
3' 59 0° 10 £0307IIAL
NEW) RELEASE MINERALIZATION UPTAKE
Carex stric ta
LEACHINO

 

3.42

ShOO! IIOMASS I
4.86
LITTER Lufttfl
1 1.44 4.12
APRI.
_ 11

11001101111011: noun: 7 UPTAKE
11.32 2.68 0-73
(MEAN) RELEASE ”$371.? MINERALIZATION

 

 

 

 

 

 

 

0.19

        
 
 
 

 
   

LEACHING
011001 mount 7
0.57

 
 

   

 

 

LITTER ‘ LITTII urn.
1 0.38 0.35 P
APRL 1 SEPY JUNE
1 gm , "32

 

 

 

noonnmzmu menus: UPTAKE
#—
1.14 0.06 0.09
. (MEAN) 7 RELEASE “5:33:21. MINERALIZATION

Figure 3.16. (Cont'd.).

107

Typha latifolia

LE ACT-0N0

5.30
LITTER
1.46 ‘
1 APfll
‘ o
, noonumzo-u manna!» UPTAKE gm m
31.4 1.51 0.90
(MEAN) RELEASE “$271.2!” MINERALIZAHON

 

    

6.75

 
    
 

     
     
  

| 51100! nouns

LITTEI

2.97

SE PT

LOYYEI

2.07
JUNE

 

     

 

 

 

 

 

    
   
   

    
     
 

LEACHING
SHOOT noun: 0 0.71
1.14
. A

LITTER

0.43

APE“.

  

LITTER

0.32

SEPT

 

 

 

 

 

 

 

    
    
   
 

      
   
  

 

   
   
    

   
  

1 ‘ gm . "32
noonnmzout nouns UPTAKE
430 ‘— 0.1 1 0.10
(MEAN) RELEASE mnenmzmou
mm? Calamagrostis canadensis
UP
- 0.21
SHOOT IIOHASO
L 5.19
LITTER “11““ “‘1‘. ‘
1 5.40 4.16 4.05 N
APML SEPT m
_ 1 -2

 

 

 

noonamzou moans UPTAKE 1 gm - m
28 7 1.24 0.1 1
(MEAN) AELEAu MINERALIZATION
LEACHING

011001 noun: 022
0.76
_ LUVYER

0.44
SEPT

 

 

    
 
   

   

 

       
   
 

APRI.

unu
0.27
JUNE
‘ gm - "12
AOOTmmzo-AE IIOIAII UPTAKE
3.02 0.10 0.17
(ME AN) nun u MINERAIZATION

Figure 3.16. (Cont'd.).

 

 

 

108

 

 

 

 

 

m.mm 0.66 webmoummEmamm «.4? «.mr mamuflum

o.mm >.mv magnumswm .o m.mF o.om mamxe
q..m «.mm muofluum x0060 o.m~ ~.mP esgcmmummm
m m z 4 306602 003 m m z w bammumem

 

.Ammficsum coflumwflamumCHE

meMUUUMH co comma mwumefiummv >mom©

mo msucos Fm umumm mcflcflmemu Amm>u xuflcseeoo
>2 UmEESm. mmme ucmfiuusc Eoeflxme wo unmouwm

.Fm.m wanme

109
transfers associated with root storage, turnover, and
production can be as large or larger than the flux of
nutrients to above ground structures. Belowground tissues
are thus likely to be very important to the accumulation of
organic matter in the marsh basin. In terms of controlling
the retention or export of nutrients, root production and
decay probably does not directly affect this process since
there direct transfers of nutrients in root biomass to the
water column do not occur.

The plant associations found in Pentwater Marsh
correspond to those found regionally in the Great
Lakes-Upper St. Lawrence basin (Auclair et al., 1973;
Harris et al., 1978; Geis, 1979; Jaworski et al., 1979;
Tessier, 1981). Riverine marshes similar to Pentwater
Marsh occur along Lake Michigan and the results of this
study should be applicable to them. The response of the
aquatic vegetation to water level fluctuations in a marsh
similar to Pentwater marsh was studied by Jaworski et a1.
(1979) using similar techniques and their findings are in
agreement with ours (Table 3.22). The most noteable
difference between these studies is a decline in the
emergent plant coverage at high water (1973) in the Betsie
River Marsh while this zone appeared to increase at high
water levels (1975) in Pentwater Marsh. Since the
Pentwater Marsh data was obtained from images taken two
years after the record high water (Figure 3.1), the

difference may be explained by emergent plant colonization

110

.mfimwamcm m0 newumm

on» mcfiuso omuusooo apacseeoo nounm may mo unmemmoH0>mc modem wummmmomc
ma >Hommumo was» as Gena pwmon>m© mo conSHUCw 0:9 .vcma ©0Q0H0>0©
pom mummHOM mopsaocfi nmumz Hm>flu mflmuwm may aw mcoN naucm mnBN

.mhmp ..Hm um fixmuozmn scum mamaP

 

 

h 1 mm m 1 vm mnsunm
wm1 om mr1 6m 300002 um:
NF+ m m 1 om ucwmumam
cm+ NP . mm+ mr . ucmmumensm \umumz
.mmcmso w. Ammmum>oo w. Ammcmzo w. Ammmnm>oo w.
mum, mom. mhms momP moss :oflumummm>
swam: um>flm mflmumm swam: umum3ucom

F

 

.zmumz um>flm mflmumm
0:0 0:0 Lmumz umum3ucom cm mmmcmco coflumummm> m0 conflumaEOU .mm.m manma

111

subsequent to an initial decline.

Many wetlands contain a mosaic of plant communities
and these communities can be expected to cycle nutrients in
characteristic patterns (Table 3.23). In the shrub zone,
nutrient storage in aboveground tissue is large and there
are two important aboveground biomass components: woody
stems which store nutrients for relatively long periods
(many years), and deciduous leaves that accumulate and
release nutrients annually. Other vegetation zones lack
the long term storage component of the shrub zone.
Overall, production and nutrient uptake appears to be
greatest in emergent plant biomass. Litter biomass and
storage of nutrients in litter is greater in the
non—flooded zones of the marsh where decomposition rates
are slower.

In the shrub and wet meadow zones, the soil/soil water
complex are the primary sites of nutrient uptake and
release. In flooded stands (emergent and submergent) the
surface water as well as the soil/soil water complex are
sites for nutrient uptake and release. Nutrients removed
from the surface water are temporarily stored in plant
biomass and become unavailable to downstream communities.
Nutrients released to the surface water through leaching
and mineralization may be available to other producers in
the marsh, or they may be flushed from the system. Thus,
the dominance of flooded communities (emergent and

submergent) or non-flooded communities (meadow and shrub)

112

 

00003 0000 \0000

090000

00003 0000\0000

 

300 “00003 0000090 009000 00 0m000 “00003 0000000 0000006050

00003

00003 0000\0000 050000 0000050 “00003
300 “00003 0000050 H0scc0 £0 “00000 0000\0000 0:0000Em

00003 050000 00003
:00: 0000\0000 009:00 00 “00000 0000\0000 300005 003

050000 H0scC0

smHn 00003 0:0 050000 >0oo3 00003

100000005 0000\0000 000::0000 :0 “00000 0000\0000 0:000
mMBBHA 2H mezmHmbaz 090030 1wazmHMHDz mch

mammoem ommémqmm mo me<m mo momDom ZOHB¢Bm0m>

ZH mwdmOBm BzmHmBDZ

 

.0000000m0> 00006 00 000000000000030

mc0ao>o 0:000052

.mm.m manmfi

113

will control, at least in part, the retention and storage
of nutrients in the marsh ecosystem as a whole.

I have estimated how changes in the areal extent of
the emergent and meadow zones of the marsh could
potentially alter patterns of nutrient uptake and storage.
These calculations are based on the weighted mean N and P
content of the two zones and on the estimated areal extent
of the zones at different lake stages (Table 3.24, Table
3.25, and Figure 3.15). The major assumptions of this
analysis are: 1) as water levels change, the species
composition and productivity in the two zones studied is
similar to the 1983 measurements, and 2) the nutrient
content of these species has remained constant with
changing lake stage.

Total storage was at a minimum during high water
(1975) when open water and submergent plant communities
dominated the marsh. Changes in the area of the emergent
and the wet meadow zones resulted in estimated changes of
1,420 kg N and 240 kg P accumulation in shoot tissue
between maximum (1978) and the minimum (1975) storage
(Figures 3.17, 3.18). Storage changes in the litter
component were 1,300 kg N and 120 kg P between a 1965
maximum and 1975. Total storage in litter biomass was
predicted to be greatest during low water, while total
uptake by live tissue was greatest in periods of
intermediate water levels when the emergent plant zone was

most extensive. The largest storage of N (7,220 kg) and P

114

:0 0005000 0:0 00 um :020 000000» :00: 0000000 000 00 :00020000000 0:0 00000000 0:00:00 0:0000sz

.0a»0 000:35500 0:0

0

 

 

 

 

 

 

 

 

s.m we owe.c A=005 000cm00av
30000: 003
c.m_ mm oom.o A:005 000:»00zv
0:0m005m
3.5 ow 00 000.0 00.0 000.0 mm.o om m.omo o~<.~ 0000000050000
«.00 00 m0 moo.o m_.o 000.0 00.0 o_ 0.000 600.0 000000000 1m.
0.0 03 0 000.0 m~.o nno.o -.~ 00 ~.oom o~w.m 0000000 x0000
zoo<mz 003
~.<~ ~n m0 m_o.o mm.o mmc.o 00.0 o— «.mmm oao.n 0=m000m
5.3— mm 0 «No.0 o~.o o-.o q~.~ m0 o.c~o~ 000.5 mnmab
0.0. 00 o 00o.o o~.o 0o_.o 00.0 o: m.oon om_.0 e=0=mmtmmw
hzmommzm
m z = N0 0 N R0 2 N c um 000500m 0mmh
uoosm 000::5500
.mmmP .00:m:¢ 0:0030 00000:=5500 300005
0005000 000:0 :0 A0:\ox. m 0:0 z 00 :00050000000 0:5 .vm.m 00005

003 0:0 0:050050 00

115

000000 000 00 N0 0000 0000000 0003 0000000 000 00 000000000000 0:0 00000000 0:00:00 00000002

I"!

.0000 000002500 0:0 :0 0005000

0

 

 

 

 

 

 

 

0.0 00 000.0 0:005 000000030
30000: 003
0.0 00 000.0 0:005 000000030
00000020
0.0 00 0 000.0 00.0 000.0 00.0 00 0.000 000.0 00000.0mea0m0
0.0 N0 00 000.0 00.0 000.0 00.0 00 0.000 00~.0 000000000 .0
0.0 00 00 000.0 00.0 000.0 00.0 00 0.00N0 000.0 0000000 x0000
zoo<mz km:
0.0 00 00 000.0 00.0 000.0 00.0 00 0.000 000.0 0000000
0.0 an 00 «no.0 00.0 m~0.0 00.0 00 0.000 000.0 0:000
0.0 00 00 000.0 00.0 000.0 00.0 00 0.000 000.0 5:00000000
0.2000020
0 z : 00 0 N 00 z N : “0 0005000 0000
000000 0000:5500
.m000 .00005< 0:0000 00000::EEOU 300006 003 _
0:0 0:000020 00 000E000 000000 :0 00:00x0 0 0:0 z 00 :00000000000 0:? .m~.m 00009

 

 

 

 

 

LIVE
5,000.9 AWET MEADOW
o EMERGENT
- TOTAL
4000-4
2 3.000—
U,
x
2.000-
L0009
0 I 1
1965 1970 19'75 1990
3000-1 LITTER
2
2.0004
U!
x
1965 1970 1975 1900
YEAR

Figure 3.17. Estimated changes in the August nitrogen storage
in shoot and litter biomass, 1965-1983.

117

LIVE

A WET MEADOW
O EMERGENT
ITOWAL

700—

  
   

600-4

500‘

0.. 400-4

300‘

200-4

100‘

 

 

I I I
1965 1970 1975 1980

LITTER
300a

200-J

1OO--i

 

 

l
1965 1970 1975 1980
YEAR

Figure 3.18. Estimated changes in the August phosphorus storage
in shoot and litter biomass, 1965-1983.

118
(960 kg) in the two zones occurred in 1983, during a period
of above average water levels, and was 33% greater than the
1975 minimum.

This study as well as those of Jaworski et al. (1978),
Geis (1979), van der Valk (1980), and Harris et al.

(1979) illustrate the resiliency of these ecosystems to
perturbation. In Pentwater Marsh, three years after the
minimum coverage of the emergent and wet meadow zones,
coverage of these had nearly recovered to pre—disturbance
conditions (Figure 3.15). This resilience can be
attributed to life history strategies which closely adapt
wetland species to disturbance (van der Valk and Davis,
1978; van der Valk, 1981; Verhoven et al., 1982).

The amount of nutrients cycled through the flooded,
emergent communities varied directly with water levels.
During the highest water levels, more N and P cycled
through the emergent community (Figure 3.17, 3.18) while
this trend was reversed during lower water levels when more
of these nutrients were stored in the wet meadow
communities. This partitioning is critical in determining
the effect of this marsh on water quality and long term
nutrient storage. The marsh may act as a flow through
system during high water with only seasonal changes in the
inputs and outputs of N and P. High N and P uptake by
plants in the spring would result in the marsh acting as a
temporary sink during this season. During the fall, when

leaching and senescence occur, the marsh may be exporting

nutrients.

The wet meadow community is much more likely to serve
as a long term sink for N and P because of slower
decomposition rates and the release of mineralized
nutrients to the soil rather than to surface water.
Seasonal inundation of the wet meadow zone often occurs
during the spring, a period of active nutrient uptake.

This pattern increases the liklihood of long term storage
in the wet meadow soil. Litter biomass may also accumulate
nutrients from flood waters through adsorption and
microbial immobilization at this time.

Sedimentation processes would be greatest at higher
water levels when reduced water velocities increase
retention time. Increased sedimentation might not lead to
long term nutrient storage in the emergent zone since plant
production processes "pump" nutrients from the sediments to
shoot biomass which, upon death, rapidly decompose.

These studies have quantified seasOnal N and P uptake,
storage, and release patterns in emergent and wet meadow
vegetation. The emergent communities appear to cycle these
nutrients at greater rates than the meadow communities.

The dynamic marsh hydrology, controlled in part by Lake
Michigan water levels, influences plant zonation and likely
alter patterns of nutrient processing by the marsh

ecosystem.

IV. NUTRIENT MASS BALANCE STUDIES

 

INTRODUCTION

Marshes and other wetlands are thought function as
important regulators of nutrient cycling, having beneficial
influences on surface and/or ground water chemistry.
Several ecological processes that might control water
quality in marshes include nutrient storage in plant
biomass and sediments, microbial and chemical nutrient
transformations, gaseous losses of nitrogen, and
sedimentation processes. The effects of freshwater marshes
on water quality has been investigated at only a few sites.
In general these studies have not quantified the hydrologic
budget required for mass balance calculations (e.g. Lee et
al., 1975; Peeverly, 1981; and Klopatek, 1975, 1978).

Water quality in several managed, enriched, Wisconsin
marshes was investigated by Lee et al. (1975) and Klopatek
(1975). For N, there appeared to be a pattern of summer
storage with an export during the dormant period.
Similarly, P concentrations dropped to very low levels in
the outflow during summer base flow, but increased during
the fall and/or spring months. Fetter et al. (1978)
studied the water quality of another Wisconsin marsh
receiving treated municipal effluent, stormwater effluent,
and industrial effluent. Again, since no hydrologic and

nutrient input-output analyses were conducted, their

120

121

conclusions on nutrient retention were based on a
comparison of input and output concentrations and must be
considerded tentative. Their data did indicate a 13%
reduction in total P and 51% nitrate-N reduction. A
nutrient mass balance calculated for a managed and
partially cultivated wetland system in western New York
demonstrated that of the large nutrient load (from
cultivated histosols upstream) only small amounts of N and
P were retained. The wetland acted as a source for K, Ca,
and C (Peeverly, 1981).

Based on the studies of Wisconsin marshes and other
studies (van der Valk, 1979), marshes appear to be at least
seasonal sinks for N and P and are probable sinks on an
annual basis for both. These conclusions must be
substantiated with reliable mass balance data from riverine
marshes. This mass balance data must be developed from
measurements of hydrologic flux as well as water quality.

Pentwater Marsh appeared to be a suitable site to
develope nutrient mass balance data. The marsh is formed
at the junction of the North and South Branches of the
Pentwater River, two fourth order streams, as they enter
Pentwater Lake (Figure 1.1). In addition to these river
inputs, the marsh basin receives runoff from two first
order streams (Lambricks Creek and Watsons Creek), the
runoff from a small (< .5 ha) man-made pond and two known
springs which surface at the edge of the marsh basin.

There is additional groundwater seepage into the marsh, but

122

this seepage is likely to be small since the area of
watershed draining directly into the marsh is small (7.2
ha).

Surface water exits the marsh through a short bridge
span associated with an earthen road bed that separates the
marsh from Pentwater Lake. Seepage through this roadbed is
not likely due to the lack of any hydrologic head between
the water bodies and the consolidated nature of the
roadbed. The marsh soils are organic peat and alluvial
sediments underlain by a fine grained marl. Seepage

through these sediments is also assumed to be small.

OBJECTIVE
The objectives of this study were to use hydrologic,
nitrogen and phosphorus mass balance data collected in

Pentwater Marsh to:

1. determine the annual flux of N and P through the marsh.

2. determine the patterns of nutrient export and retention
by the marsh basin.

3. evaluate the potential role of emergent and meadow

plant nutrient dynamics in controlling water quality
and nutrient retention in the marsh.

METHODS
Hydrologic budget
Hydrologic gaging stations were installed on the North

and South Branches of the Pentwater River following

123

standard procedures (Carter and Davidian, 1968; Buchanan
and Sommers, 1968). The stations were equipped with a
strip chart recorder (Stevens, Type A-71) allowing
continuous stage measurements. On each of these rivers,
discharge was measured over a range of stages to develop
stage—discharge relationship (rating curve). Following the
methods of Buchanan and Sommers (1969) discharge was
calculated using a velocity area method. The midsection
method of computing cross sectional area was used, with
velocity measurements taken at 0.6 depth using a vertical
axis flowmeter (Pygmy or Price type). The section width
was 0.76 m (selected so the discharge per section would
typically be less than 8% of the total discharge). Using
the continuous stage records from the gaging stations and
the developed rating curve, hourly computations of
discharge were summed by day.

lStage-discharge relationships could not be used to
calculate discharge from the marsh basin because the water
level at the marsh outlet is controlled primarily by the
Lake Michigan water level rather than stream discharge. At
the outlet, discharge measurements were periodically made
with permanently installed current meters (General
Oceanics). These meters were modified with propeller stops
to prevent interference from seiche induced current
reversals. Four meters were installed in two pairs at
1800 to each other to measure both outflow and seiche

induced current reversals. The meter sets were positioned

V24
at 1/3 and 2/3 of the channel width and at 0.6 the mean
depth. Lake Michigan seiche influences were also measured
with a battery operated timing device connected to a flow
sensitive switch. These procedures did not result in
reliable discharge measurements (see results) and discharge
from the marsh was estimated as discussed below.

The hydrologic inputs from two smaller streams,
Lambricks Creek and Watsons Creek were estimated from
weekly measurements of discharge. Direct precipitation
inputs to the marsh basin were calculated from
climatological data taken at a nearby (6 km) weather
station (Hart, Mi) (NCAA, 1983) and from measurements taken
at a station established 15 km north of the marsh.
Groundwater inputs were estimated from yield/area
relationships calculated for the gaged watersheds.
Hydrologic outputs were assumed to equal to the sum of
water inputs. Evapotranspiration was not measured and
estimations are not included in the water budget because

reliable data for temperate freshwater marshes are lacking.

Chemical Sampling

Water samples were obtained at least weekly (x = 4.2
days) from the North Branch, South Branch, and outlet
sampling sites. Watsons Creek, Lambricks Creek and the
spring were sampled weekly (Figure 4.1). Water samples
were taken in 250 ml polyethelene acid washed bottles from

the mid-channels at approximately 0.6 depth. The bottles

125

1

    
     
   
    
  
 
  
    

\.
1

US 31

1”.

Pentwater Lake North Branch
l‘\‘

\f‘

  

Lone Bndo
no.

‘u/
BUS 31

 

 

. water sampling statio

o recorder station 8

© recorder 81 sampling static

 

, 7 I .
\ 1"
Lambricks Creek .

Figure 4.1. Locations of stream gaging stations and
water sampling sites.

126
were rinsed several times in stream water before sampling.
Samples were immediately stored on ice until return to the
laboratory where they were frozen and stored at -250 C
until chemical analysis could be completed. At the marsh
outlet, care was taken to always obtain water samples
during periods of outflow. Periodically at this station,
inflow was also sampled.

Additional sampling was conducted at the outlet to
determine if, during seiche cycles, the water entering the
marsh from Pentwater Lake is chemically different from the
river water flowing out of the marsh. An automatic water
sampler (ISCO) was programed to collect water samples at 15
minutes intervals over a 7 hour period. Correlating the
stage data with flow observations and the water chemistry
data allowed analysis of water quality during seiche
induced current reversals.

To check the precision and accuracy of the sampling and
analytical procedures, one set of quality control samples
was collected according to U.S. E.P.A. (1979)
recommendations. Quality control samples were taken at the
North Branch, South Branch or outlet sites on each sampling
date and included 1) duplicate samples, 2) spiked samples,

and 3) blanks (distilled H20).

Chemical analysis
Water samples were analyzed for total phosphorus,

total reactive phosphorus, total Kjeldahl nitrogen, ammonia

127

nitrogen, nitrate-nitrite nitrogen, nitrite— nitrogen and
chloride using a Technicon Auto-Analyzer system, following
U.S. Environmental Protection Agency (USEPA, 1979)
techniques (Table 4.1). Total nitrogen and organic
nitrogen were calculated from measured parameters as
follows:

Total-N = TKN + NO —N + NO -N

3 2

Organic-N = TKN - NH4-N

Budget Calculations

 

Input-output budgets for the marsh were calculated as
follows. The year was divided into intervals for which
water quality data had been collected with the collection
date defined as the mid—point for each interval. Water
inputs for each component of the water budget were
calculated for these intervals. Nutrient inputs for each
water budget component and for each interval were
calculated as the product of the hydrologic input and the
measured nutrient concentration (LaBaugh and Winter, 1984;
Scheider et al., 1979). Water output from the marsh was
estimated as the sum of all hydrologic inputs. Chemical
outputs were calculated as the product of the estimated
water output and the measured nutrient concentrations in
the outflowing water. Total nutrient inputs during each
interval were calculated by summing the component inputs.
For each season, and for the year, the mean nutrient inputs

were tested against the mean output using a t-test for

 

opflcm>ofluumm pmumeouo< N.m~m Ho

oumcoco poumeouom

+ :Ouumwmnc xooam ~.me z-H:mnH0flx
map Cum + Zumoz
cofluoocou Eofleomo coumeouo< m.mmm +21moz
map on pwumeouo< m.mmm anoz
oumcosa boomeouoé P.omm Zuvmz
8
2 ucommwu mamcflm m1o>wuomou
1 .oflom oflnuoomm ooumeouoé N.mmm amuoe

ucommou mamcflm
.pflom ofinuoomm
omumeousm + acuummmflc xooam «.mom m1Hmuoe

 

.msmp 44mm ma.
wm<zzom .mmm oomemz mmemz<mcm

 

.moaoemm kum3
mo wcoflumcfiEuouoo ucofluuoc MOM pom: mflmxamcm >u0umuonmq .P.v manme

129

paired observations (Steele and Torrie, 1980).

RESULTS

Hydrologic budget

 

The largest hydrologic inputs to the marsh were the
North and South Branches of the Pentwater River. The mean

flow for these rivers was 2.278 m3 sec"1 (sx = 0.151)

and 3.256 m3 sec"1 (sx = 0.0741) representing 40% and
56% respectively, of the estimated total water inputs to
the marsh (Table 4.2, Figure 4.2). The measured
instantaneous discharges of these rivers were weakly
correlated to each other (r2 = 0.40) but significant at p
<0.01. The total discharges for each sampling interval
(3—8 days) were more strongly correlated (r2 = 0.70) to
each other. Estimated discharges for Watsons Creek and

Lambricks Creek were x = 15.3 x 10"3 m3 sec“1 (sx =

0.4 x 10'3) and x = 50.8 x 10"3 m3 sec'1 (sx = 2.2

x 10-3), respectively. The discharge of these smaller
streams remained relatively constant throughout the year.

The discharge of the North and South Branches were
correlated with the larger precipitation events, although a
lag time of 1-3 days between storm events and maximum daily
discharges existed. During 1983, precipitation totaled
99.1 cm, 11.9 cm above normal. Monthly precipitation

averaged below normal from January through April and

fluctuated thereafter. The period January through March

 

130

m.¢ m.m I w.

 

s.mom~ Ammp.oPV .me.omPV someoo
m.mmo~ mm..oP ms.om. mesmzH a<eoe
8.0 eqe.o mm.m cmmmmcs
m.m Pmp.o me.o xmmuo mcomumz
m.¢m ooe.o oo.F xmmuo mxoMHQqu
o.mmFP mmm.m em.Fs nocmum :uuoz
c.9m4P vm..m P~.moP nocmum cusom
«.mmmm ooo.oF F~.mk. q<eoe mmozam
P.o o.o o.o Hamusun
m.o amp.o sm.. chum
4.0 sm..o s~.F gases one<eHmHommm
m
.moF x x. Amos 9 me.
mpfluoHcU Houum nwumz mouoom

 

.Amwmp .Fm Monsooooummmr .P >umocmh. nmumz
noum3ucom How oocmHmn mmmE wpfluoHco 0cm pompon noun: one .m.v manna

131

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PRECIPITATION - 15
141
12-1
‘ 10: 1-10
0 .1
v- 8-
.“ . H
E 6-
. -5
4d
2: ML
0% r l AAjI T 0
J F M A M J J A s d A E J
DATE
MEAN DAILY DISCHARGE
al N
6 - North Branch
4-
22W
'5 °‘
v- 4-
x .
.E 2‘ W
- SouthBranch
0
6d
4:
- Outlet
2 I l I I F l 1 l 1 1 T 1
JFMAMJJASOND'J
DATE

Figure 4.2. Precipitation (Hart, MI) and streamflow

during 1983 sampling periods.

132

averaged 2.70 C above normal. As a consequence of the
light precipitation and warm temperatures, snow
accumulations in the watersheds were small, reducing early
spring runoff. The greatest monthly precipitation occurred
in September (18.3 cm) and the greatest daily precipitation
was 11.7 cm (July 28) (Figure 4.3). At these times, some
of the largest discharge events were measured.

The flowmeters installed at the marsh outlet did not
provide reliable velocity measurements. During short (1-3
day) intervals when meter operation was checked daily,
discharge estimates generally fell within -10% to +20% of
the estimated inputs. The meters were prone to clogging
with aquatic vegetation and fishline. Since they could not
be cleaned daily, reliable and continuous discharge records
were not obtained. Even when the meters had been operating
for 5-7 days without apparent clogging, discharge estimates
were typically -15% to +30% of the total estimated water
inputs.

The timing device measuring current reversals did
provide reliable estimates of current reversals. The
measurements indicate that reversals can occur for up to
40% of the time during certain 4-10 day intervals (Table
4.3). Flow reversals were longer and more frequent during
periods of low flow and/or high Lake Michigan water levels.
From July 15, 1983 to March 15, 1984 inflow into the marsh
basin occurred an average of 27.5% of the time. Velocity

measurements made with a Price Current Meter over seiche

133

Table 4.3. Percentage of time outflow from Pentwater
Marsh was measured at the marsh outlet
(July 1, 1983 - March 16, 1984).

 

 

DATES # Days % OUTFLOW
7. 1- 7. 7 6 61
7. 7- 7. 8 1 69
7. 8— 7.15 7 63
7.15- 7.28 13 70
7.28- 7.29 1 100
7.29- 8. 4 6 (m)
8. 4- 8.20 16 58
8.20- 8.26 6 87
8.26- 8.27 1 82
8.27- 9. 1 4 76
9. 1- 9. 6 5 84
9. 6- 9. 7 1 64
9. 7- 9. 8 1 58
9. 8- 9.25 17 91
9.25- 9.29 4 98
9.29-10.16 17 78
10.16-10.20 4 88
10.20-10.27 7 86
10.27-11. 3 7 79
11. 3-11. 4 1 88
11. 4-11. 6 2 86
11. 6-11.12 6 86
11.12-11.17 5 85
11.17-11.26 8 84
11.26-12. 4 8 83
12. 4-12.1O 6 84
12.10-12.17 7 86
12.17-12.24 7 95
12.24- 1. 4 1O 97
1. 4- 1.25 21 99
1.25— 2. 4 9 93
2. 4- 2.18 14 99
2.18- 2.22 4 99
2.22- 3.16 23 76

MEAN 82.5%

 

134

cycles (Figure 4.3) indicated that during periods of
inflow, velocity was frequently below the threshold of the
permanently installed current meters, and this may
partially explain their poor performance. Because of these
problems, the most reliable estimate of water discharge
from the marsh was believed to be the total water inputs.
Errors in the water budget (Table 4.2) were estimated
from replicate measurements of discharge (x = 5.02% on the
North and South Branches), from calculations of rainfall
variations between three nearby stations (x = 10.01% for
wet fall) and from variations in water yield/unit area (x =
24.95% for the remaining components). The overall error
associated with this water budget (component errors
weighted by the component discharges) is less than 6.0%.
The relatively high accuracy of these estimates is due to

the reliable gaging of the North and South Branches.

Water chemistry

The annual mean nutrient concentrations for all
sampling stations are summarized in Table 4.4. Most
nutrient constituents measured lacked seasonal fluctuations
in concentration. The exception to this was nitrate-N (and
total-N, as a result of the N03 contribution to this
parameter) which had maximum concentrations during the
early winter and minimum concentrations during the summer
(Figure 4.4). The nutrient concentrations also lacked

strong correlations to instantaneous discharge

135

.uoauoo chum: noumzucmm
um Ammo? noneouoom mev oao>o ozoflom m mcfiuop >uwoon> noum3 :fl mmocmcv .m.v ouomflm

‘ 0E:

008 mqum Omum mwnm ooum me; On; my;

 

(see/we) MgooleA

 

136

 

02\m:\wxv

 

 

000.0 000.0 020.0 000.0 000.0 000.0 000.2 00.0 2000 0000200
000.0 200.0 000.0 000.0v 000.0 000.0 002.0 002.0

00.0 20.0 0.0.0 200.0 200.0 00.0 000.0 000.0 2000 0002
0002.0 000.00 200.0 200.0 000.0 000.0 000.0 200.0

00.02 000.0 000.0 200.0 0_0.0 000.0 020.0 000.2 2000 000000
000.0 200.00 000.0 000.00 000.0 000.0 000.0 000.0

00.0 000.0 020.0 200.0 02.0 00.0 02.0 00.0 2200 000000
000.0 000.0 000.0 200.0 200.0 200.0 000.0 000.0

00.02 000.0 020.0 000.0 200.0 000.0 000.0 000.0 2200 20000 0000003
000.0 200.00 000.0 000.0 000.0 000.0 000.0 000.0

02.02 000.0 000.0 02.0 000.0 00.0 00.0 020.2 2200 20000 020000000
_00.0 200.00 000.0 200.00 000.0 000.0 0000.0 000.0

00.02 000.0 000.0 000.0 200.0 000.0 000.0 022.0 2000 200000 20002
000.0 200.00 200.0 200.0v 000.0 000.0 000.0 000.0

00.02 000.0 000.0 020.0 000.0 000.0 000.0 000.2 2000 200000 20000

002 0120000 2-002 21002 2-022 2-0000000 2100000 0 000000
.Anouuo

cumocmum mom came. mmmr .Pm Hooeoooou mwmr .0 >umocmh .cmumz Houmzucom

wo mEmouum Doouoo pom uoo:0 0:0 wo AA\OEV mcoflumuucoocoo ucofluuoz .v.v oHQMB

137
NITRATE NITROGEN

North Branch

W

 

 

 

 

 

 

 

 

 

 

 

0
e 2- South Branch
0 d
E 1‘
V d
0
29 Outlet
1-
.1
0 I I I F T T T I fl
J F M J J A s o N D J
DATE
. TOTAL NITROGEN
NORTH BRANCH
2.04
L04
0
4.0- SOUTH BRANCH
3.04 U
15:10:
E
0
2.0- OUTLET
1.0-
.1
c
I l I I I
.1 F 111 .1 .1 A I 0 1'1 0 I

Figure 4.4.

Nitrate-N and Total-N concentrations in the
North Branch, South Branch, and marsh outlet

138

measurements. Total reactive phosphorus and total-P
concentrations were 10 — 15 times their mean concentrations
during some of the largest discharge events (Figure 4.3 and
Figure 4.5), but over the entire study period, the
concentrations of these constituents were not significantly
correlated to discharge. Some concentration - discharge
regression lines were significant (P < 0.05) but the r2
value of these relationships was low (< 0.15) and the
regression line slopes were close to zero.

Water samples were collected at the outlet bridge on
three dates to test for concentration differences during
flow reversals. A total of 20 seiche cycles were studied.
There was no statistical difference between inflowing water
and outflowing water except for nitrate-N on one of the
sampling dates (Figure 4.6). A 4% decline in nitrate-N
concentrations was observed during the inflow periods on 13
June, 1983. On 15 sampling dates, inflowing water was
collected along with outflowing water and there was no
evidence (p <0.05) that the two flow periods represented
different water masses. Thus, the potential for seiche

induced nutrient inputs were ignored.

Mass Balance

During the 3-7 day sampling intervals, the marsh
appeared to be variable with regards to nutrient export or
retention (Figure 4.7). To evaluate nutrient retention or

export on a seasonal basis, the year was divided into 4

139

 

 

 

 

    
 

 

 

 

 

 

 

 

TOTAL
002 REACTIVE PHOSPHORUS
.. North Branch
0.01-
0‘ I
E; 0.02- South Branch
‘g 00fl
0 .
0.01 : Outlet
0-
I I I I I T I I V I I ‘1
J F M A M J J A s o N 0
DATE
TOTAL PHOSPHORUS
004: North Branch
0.02 :
111a-
: 0-04: South Branch
3’ 0.02. M
0‘ _ A
0-043W 0mm
0.02.
o I I A I I I I I I I I m
J F M M J J A s o N o J

DATE

Figure 4.5. Total reactive-P and total-P concentrations in
the North Branch, South Branch, and marsh
outlet (1983).

140

.0000 .0000
«PIMP .uoauso Lmumz Houmzucom wcu um Uwusmmoe wcofiumuucoucoo zlwumuufiz .m.v whomflh

. . 0:5
00.00 00.00 . 0060
21 p1 P - 1P1 P n .P P _ h p P. P P .1 p n p p

b

— Datum"
Pp PPP m¢.o

 

1lomgu

 

25:5 26.2.5

I/Bw

rlmmgu

 

21m._.<m._._z I000

141

ZIUficmmuo pom 06000050

 

ZOFthwC

wh<o
0 0

PCOnxw

mDmOIawOIa 12.—.0...

5

 

 

 

      

O
Q

 

efiueuo 95

O
N

     

 

 

 

 

wh<oa .2 a
_I w a My W “ M. P (p‘ N — tar 0Q
100
IO?
20:2whum I
18%
L > > 57 7 > lo qua.
< _ 2 < < Z 1 m
Iowa
Eonim How
Iow
ZMOOEFE w...<m._._2 I

.mam>uouc0 mc0aosmm mmmp mcfiuoo
0210000002 .0100000 no coflucouwu pcm unoaxm

 

.0.0 003000

 

 

mh<o
a o z o w M a _. .2 < .2 .0 _.
Fl _ _ _ _ _ _ _ _ 0 .
100
100
20:20:00 I
> Ion an
F D 5h > 53> P > \IO 0
<5<€<Z<§<1 £150 1 m
1006
I O
10..
:00wa I
100

mo_m01=._0

wh<o
..

 

20:2whwm

thaxw

ZMOOICZ 0_2<0¢O

 

 

efiueuo 95

    
 

142

periods, corresponding roughly to the seasonal activity of
the aquatic vegetation of the marsh. These periods were
defined as:

1. overwintering (January 1 - April 15).

2. rapid growth and nutrient uptake (April 16 - July 15).

3. cessation of growth and nutrient uptake (July 16 -
September 30).

4. senescent period and onset of dormancy (October 1 —

December 31).
For each of these periods, nutrient flux was calculated by
summing flux data for each sampling interval of the period.
On a seasonal or annual basis, for many of the parameters
examined, there were significant alterations of water
chemistry and the flux of nutrients through the marsh.
Annually, the dominant inorganic-N forms (NH4-N and
NO3—N) were retained in the marsh by 11.9% and 6.8%
respectively (Figure 4.8 and Table 4.5). An annual export
of organic N occurred (26.1%). On an annual basis, the
estimated 6.2% export of total N was not significant and
the total nitrogen budget appeared to be in balance.
Phosphorus was retained in the marsh on an annual basis.
The export of total phosphorus was 21% less than the
total-P inputs and total reactive P export was reduced 9.4%
compared to inputs (Figure 4.9 and Table 4.6).

A chloride input-output budget was constructed to

verify the hydrologic budget (Table 4.2). Since chloride
remains soluble in freshwater systems and biological

utilization is insignificant relative to its abundance, the

 

 

143

.5000: uwum3ucwm
mo ummcsn cmoouu0c Ammmev Hanson cam HMGOmmmm

50> .< :w om >> 50> 2 am am
2950000 0

$$

 

 

.0.0 000000

I00

ION

To

To.

ON

50.020
zmoomtz 0252 2085.2 <_zo2.2< . 00
295050
N
kmOaxm 0s

 

Ion

 

__.[_

Zw-UOECZ 0.2(0m0 ZwOOth 1_<._.O._.

39NVHO 96

39NVHO$6

144

 

 

0.01 0.0. 0.0 0.0m1 ..~.1 2~1e 90. 200002002

0.0.1 0.0 0.0 0.001 0.01 2oHeoaomm 0

0.01 0.0. 0.0 0.001 0.0.1 200. 02. 200200002

0.. 0.00. 0.0 0.... 0.00m 000000

0.. 0.00. 0.0. 0..0 0.00m 002020 00000

..0 0.0 . 0.0 0.0 00000::

..0 0.. . 0.0 0.. 20000 0000002

..0 0.. . 0.0 0.0 20000 020000200

0.0 0.00 . 0.00 0..0 200000 20002

0.. 0.00. . ..00 0.00. 200000 20:00

0.. 0.00. 0.00 0.000

..0 ..0 ..0 0.0

..0 0.0 0.0 0..

..0 0.0 0.0 0.. 04009 2002000000020
.00. x 02.

21~oz 21002 21022 210020020 2100000 000000

 

..mmm. ..m
00250000 1 mmm. .. >umocmwv £0002 umumzuc0m “Ow 00:0002 000E c0mouu0c 0:8 .m.v 0Hnme

.5000: umum3ucmm 00 0000039 w©0000£u
cam 0000590050 Ammmpv Hmsccm cam 00:0000m .m.v 005000

30> 2 0m 0m >> 50> .< 06 am 3
10'

29020000 H 8

TON

SBNVHO 96

 

a...

4 0000100010 .2000 0

 

ZOFthwm I

I
‘BSNVHO as

is I

i. .I Op
500%.. 0 ..

wDEOImwOIQ w>_._.0<mm wo_mO._IO
m._.<om>1_0_2

 

146

 

 

0..0 00.. 00.. .010 00. 200000000

0.0. 0.00 0.00 .0. 200000000
00. 0.0. 000. .00. 200000000
000. 0000 0000 000000
000. 0000 00.0 000020 00000
0. 00 00 0000000
N m m xwth mCOmum3
0 00 00 00000 000000000
000 000. 000. 000000 00002
000 0.00 00.0 000000 00000
0.0. 0000 0000 00000 000200
0 00 00 0000000
N. 0 om :00m
0. 00 00 00000 2000000000000

.00. .00. .00.

0 w>0uommm 0 0000000 0 00000 000000

 

mwm—

.0 >Hmscmn.

.Amwmr

..m uwnemomo
0000: 000030000 000 0000000 0005 0000000000 was .0.0 00900

147

chloride inputs to an ecosystem should equal the ecosystem
export of chloride. Thus, a chloride budget provides a
"tracer" for estimating errors in water and nutrient budget
calculations. A balanced chloride budget suggests accurate
hydrologic measurements and nutrient mass balances.

The chloride budget was calculated by the same methods
used for the budget calculations of the other nutrients.
As with most other nutrients examined, chloride
concentrations were relatively constant throughout the year
(Figure 4.10). There was a weak, negative correlation
between discharge and stream water chloride concentrations.
As with other nutrients, the greatest contribution of this
constituent was attributable to the North and South
Branches of the Pentwater River. A significant (p (0.01)
annual export of chloride was measured (4.9%) (Figure 4.8
and Table 4.2) but this export was less than the 6% error

expected due to the accuracy of the water budget.

DISCUSSION

Water yields per unit of watershed area of watershed
were variable (Table 4.7). This variation may be
attributed to several factors including 1) variations in
soil prOperties of the watersheds; 2) errors in the
interpolation of topographic divides from USGS 1:62,SOO

maps; and 3) discrepencies between topographic and

148

.Ammme. umauso
:mnme 6cm nucmum zusom .Locmum Luuoz mnu CH mcoflumuucmocoo mcfluoHnU .oP.w musmflh

 

mk<o
a. 0 Z Om < _. 5 .2< .2“. .a
__P _ _ _ F _ __rr 0
giro?
8.50 -8
O

 

I .II
I w

0.. 5

V

I\.

50:95 £30m

 

 

5965 £52
m0_mO._IO

149

 

 

om.o h.r ~.h cmmmmca
mmm.o o.P q.v xwmuu mxoflunema
vom.o v.9 m.m xmwuo mGOmumz
omm.o m.>v m.mmp socmum nunoz
vom.o m.m¢ m.~o~ nucmum nusom
ANIE may mmum

anfiz uwumz Hmuou mo w Exv mmu< Gammm mmmcflmuo
AN

 

.AmmmP .Pm “mnemomo-mmmF .P >Hmscmn. nmumz
umumzucwm Ou acmommcm mcmnmumum3 EOuw oamfl> uwum3 Ucm mmmum wmmcfimuo .h.v manme

150
hydrologic divides. The yields (except Lambricks Creek)

fall within the range of values for rivers in west-central
Michigan (Knutilla, 1967).

Discharge events were not always correlated similarly
to precipitation events (Figure 4.3). For instance, the
intense rainstorm of 27 July resulted in a large discharge
response, but the response was smaller than that which
occurred during late September following a less intense
storm. This indicates that additional factors besides
precipitation affect runoff patterns. These factors may
include the artificial retention of water in the Hart
Reservoir, antecedent soil moisture conditions and patterns
of evapotranspiration. The greater variability of flow on
the South Branch was in part the result of regulation of
discharge for the purpose of power generation at the Hart
reservoir.

The precipitation chemistry and dryfall inputs (Table
4.4) are generally close to other values for the Great
Lakes. Per centimeter of rainfall, the inputs of P
measured at Ludington (this study) are 60% less than those
measured at other Michigan stations. Chloride inputs at
Ludington were much below regional values (Richardson and
Merva, 1976). Total precipitation (dryfall plus wet fall)
inputs of N and P were within the range reported for the
Great Lakes region (Wetzel, 1983).

The export of chloride (4.9%) was about equal to the

estimated error of the water budget, but its statistical

151
significance indicates that the hydrologic inputs may be
underestimated, or alternatively, the hydrologic outputs
overestimated. Since evapotranspiration was not included
in the water budget, actual discharge from the marsh was
likely somewhat less than value used in these calculations,
and this might contribute to the chloride imbalance. The

4 9

137 x 10 kg Cl error represents 8.76 x 10

-1 2

L H20
(at 15.6 mg Cl L ), or 7600 L m of marsh surface.

This value is many times regional potential
evapotranspiration, indicating the calculated export of
chloride is probably attributable to measurement errors.

Nutrient concentrations were not strongly controlled
or correlated to discharge and the calculations of nutrient
loading based on flow-concentration regression equations
would be similar to the calculations presented which were
based on discharge and unweighted concentration data.

The variable function of the marsh with regards to
nutrient retention may be related to temporary storage and
the variable time of travel of water masses moving through
the marsh. Due to the relatively large volume of water in
the marsh basin, the turnover time of water was much longer
than the sampling intervals. The mean retention time,
assuming total circulation, during 1983 was estimated to be
60 days. During periods of high and low flow, retention
times vary between less than 15 days and greater than.81t

days respectively. Thus, the chemical outputs measured for

one period may reflect, in part, the inputs of a sampling

152
period several weeks prior.

Dissolved diurnal 02 measurements indicate that the
marsh contributes a considerable oxygen demand on the water
flowing through it (Figure 4.11a). Even at mid-day, 02
concentrations at the outlet are 30% below those of the
incoming water. The reduction in oxygen is likely due to
decomposing plant matter, as measurements of the 02
content of water flowing across vegetated meanders show a
considerable reduction (Figure 4.11b). The measurements
indicate that significant water quality changes occur as
water flows through flooded stands of vegetation.

These studies have focused on hydrologic transport of
nutrients and have not included gaseous N fluxes. Nitrogen
fixation and denitrifiction processes are likely to be of
significance in wetland systems (Hemmond, 1983; Keeney,
1973; Reddy and Patrick, 1977). For instance, in the shrub
zone of the marsh, symbiotic N-fixation with glnug may
contribute to the N inputs (Wetzel, 1983). Non-symbiotic N
—fixation also occurs in th rhizosphere of wetland plants
(Biesboer, 1984). Significant N losses would be expected
where NO3 can diffuse into anaerobic environments
(Keeney, 1973).

In summary, with regards to hydrologic nutrient
inputs, the overall function of this marsh appears to be
the removal of modest amounts of total and inorganic
phosphorus from surface water inputs. Inorganic N forms

appear to be transformed into organic forms within the

onunmu. onssowso o,

oussorvco oxvorN (mg n

 

 

 

TIME Ihour.‘

 

 

Figure 4.11. Diurnal 0 changes (6 September, 1983) (A),
and change in 0 concentrations across
flooded marsh communities (B).

154

system. This organic N is then exported from the marsh.
Much of the nutrient retention and transformations occur
during the growing season suggesting biotic control of the

processes.

V. SUMMARY

Numerous mechanisms for nutrient retention and
transformations exist in marsh ecosystems. Both N and P
may be stored in plant biomass, accummulate in litter
biomass, 01 become immobilized through decay processes.
Nitrogen (NBA) and phosphorus may be adsorbed to
sediments, and microbial denitrification can result in
gaseous N losses.

The N and P mass balance data collected at Pentwater
Marsh demonstrate that ecosystem processes in this marsh
alters surface water chemistry and nutrient export.
Compared to the input-output budgets, nutrient assimilation
by shoots represent a small percentage (2%) of the-total N
and a larger (15%) percentage of total P inputs. Plant
uptake-storage-release processes could not be expected to
remove more than these amounts. Further, on an annual
basis, decomposition processes result in the mineralization
of a considerable portion of the nutrient annual uptake,
reducing the potential effect of plant nutrient uptake on
water chemistry.

In the communities studied, substrate analysis (Table
5.1) indicates that, to a depth of 20 cm, the substrate

2 and 23-40 kg p m‘z. With the

contains 380-750 kg N m—
exception of Typha latifolia communities, the soil N and P
concentrations are greatest in the meadow zone. The soil

nutrient reserves represent the largest nutrient pool in

155

156

 

 

 

 

 

 

 

 

moo.o mo.o P.o m.m No.o
mmo.o mm.P o.m v.mm hp.o mflmcmcmcmo maumOHmmEmamu
moo.o no.0 N.o N.N Po.o
mor. no.~ v.0 m.mm mp.o mflawumawm xmumo
voo.o mo.o P.o m.P Po.o
mNP.o vm.~ N.m m.mn Np.o muofiuum xmumo
oFo.o mm.o 9.0 b.h No.o
opp.o vo.m m.m P.mm NP.o mflHOMflumH MLQNB
hoo.o Fw.o v.0 m.m mo.o
mko.o m~.F m.o o.mm kP.o msoflam> msgufiom
moo.c 09.0 P.v >.m no.0
moo.o om.o m.m ~.Fm mm.o esmumomnam Enacmmummm
monogamona cwmouufl: :ofluficmfl xufimcwc waxy
Hmuou w Hmuou w mm :0 mmoH xasn . ccmum

 

.nmumz
umpm3ucmm cfl moaxu coflumuomo> xflm on» wo monumfiumuomumso Hwom .—.m manna

157

all communities studied. The nutrient storage in August
aboveground plant biomass (live shoots and litter) was a
small percentage (<2.5% N and < 6% P) of the soil nutrient
pools. "A larger portion of the N and P pools were
associated with the belowground root and rhizome tissues
(up to 8% N and 32% P). Only a very small proportion of N
and P (< 1%) were in inorganic forms in the soil
interstitial water. Because the plant P requirements are a
larger percentage of the soil pool than N requirements,
phosphorus turnover is probably somewhat greater than N
turnover rates.

The annual nutrient requirements of the vegetation and
the small inorganic soil nutrient pools suggest that
mineralization processes, soil nutrient inputs, and/or
plant conservation of nutrients (translocation) are
important in sustaining the high rates of nutrient
assimilation observed. If annual production was dependent
upon a static inorganic, soil water nutrient pool only
about 50% of the annual plant P requirements would be met.

The conservation of nutrients by translocation prior
to plant senesence may be an important mechanism reducing
annual requirements for soil derived nutrients.
Translocation was not measured in this study, but a maximum
value can be estimated by the decline in the nutrient
content of shoots between maximum biomass and shoot
senesence. For the species studied in Pentwater marsh,

translocation may conserve an average of 27% and 24% of the

158

maximum N and P shoot mass (Table 3.10 and 3.11). Shaver
and Melillo (1984) measured translocation from 3 wetland
perennials (leaf tissue only) and found 24 - 46% of the
maximum N and 2 — 63% of the maximum P were recovered from
senescent leaves prior to death. Klopatek (1978) estimated

12 - 13% of the N and P content of Scirpus fluviatilis was

 

retained at the end of the growing season. Hopkinson and

Schubauer (1984) found Spartina alterniflora to recover

 

over 54% of its aboveground nitrogen requirements. Thus,
internal plant nutrient cycles appear to be conservative of
a significant percentage of the annual nutrient
requirements.

While mineralization of soil N and P likely plays a
key role in the replenishment of the inorganic soil
nutrient pool, other mechanisms of replenishment are
possible. For instance, nutrient movements into the
rhizosphere might occur as a result ground water seepage
into the soil, as precipitation inputs (in nonflooded
areas), or by transpiration driven water and nutrient
movements into the substrate from the surface water. The
plant communities studied do not remove nutrients directly
from the surface water, but rather from the soil
interstitial water. Plant uptake could facilitate nutrient
removal from surface water by depleting the inorganic
nutrients of the soil water and increasing the rates of
diffusion into the substrate.

The high organic matter content of the meadow and

159

Typha soils versus Scirpus and Spargnium soils suggest
major differences in the processing of carbon and nutrients
between these two zones. These differences may be

attributed to the following:

1) Greater microbial activity in the emergent zone litter
increasing carbon and nutrient mineralization. This
could result from the nutrients supplied by
streamflow.

2) The greater abundance of invertebrates in litterbags
placed in emergent zones, increasing litter loss by
consumption and communition. The higher invertebrate
populations may be linked to increased litter quality
which occurs with greater microbial activity.

3) The hydrologic export of fine particulates from
emergent communities, reducing organic matter inputs
to the to the substrate.

4) Greater inorganic inputs to the emergent zone
substrate "diluting" the organic input to the soils

Organic matter production and litter accumulation,
while being directly responsible for nutrient transfers
between storage pools may control other aspects of nutrient
cycling. For instance associated with litter decomposition
are periods of nutrient immobilization that may alter
nutrient availability to other organisms. Organic matter
in anaerobic environments may serve as an energy substrate
for denitrifying bacteria promoting gaseous nitrogen
losses. The nitrate source for these bacteria may be
mineralized N from the litter itself, or NO3 which has
seeped or diffused into the sediments from the surface. In

the soil, NH4 may be oxidized to NO3 in the rhizosphere

of plants which transport 02 to their belowground organs.

160

This NO may be subsequently denitrified.

3
Though this data suggests that emergent vegetation
could potentially cycle only a small percentage of the
annual N and P inputs to the marsh, both inorganic N and P
were reduced about 160 kg/ha and 14 kg/ha respectively.
These amounts are 25% and 59% the annual N and P shoot
requirements respectively of the emergent and meadow

vegetation. Inorganic N (NH and N03), total—P and

4
molybdate reactive-P were retained in the marsh primarily
during the growing season. The reductions are a large
percentage of the annual emergent plant N and P
requirements, suggesting that the production and decay
processes may control, at least in part, the flux of N and
P through riverine marshes. An annual export of organic
nitrogen was found to be statistically similar to the
inorganic nitrogen removal. Thus, with regards to the
hydrologic flux of nitrogen, the input-output budget
appears to balance, and the marsh functions by transforming
inorganic N to organic forms which may be exported from the
system.

Wetlands have been thought to function to varying
degrees as "nutrient traps" which improve surface water
quality through nutrient removal. This study supports this
belief and quantifies some of the nutrient removal
mechanisms in marshes in relation to hydrologic nutrient
flux through the system. Additional study is required to

further identify the relationships between marsh nutrient

161

cycles and nutrient flux. In particular, the control
wetland hydrology exerts on the physical, chemical and

biochemical environment must be understood and quatified.

LIST OF REFERENCES

LI ST OF REFERENCES

Aber, J. D. and J. M. Melillo. 1982. Nitrogen
immobilization in decaying hardwood leaf litter as
function of initial nitrogen and lignin content.
Can. J. Bot. 60: 2263-2269.

Auclair, A. N. D., A. Bouchard, and J. Pajaczkowski.

1976. Plant standing crop and productivity in a
Scirpus-Eguisetum wetland. Ecology 57: 941-952.

Auclair, A. N. D. 1982. Seasonal dynamics of nutients
in a Carex meadow. Can. J. Bot. 60: 1671-1678.

Auclair, A. N. D. 1979. Factors affecting tissue
nutrient concentrations in a Scirpus-Equisetum
wetland. Ecology 60: 337-348.

 

Auclair, A. N. D., A. Bouchard, and J. Pajczkowski.
1973. Plant composition and species relations on
the Huntingdon Marsh, Quebec. Can. J. Bot. 51:
1231-1247.

Berg, B. and G. Ekbohm. 1983. Nitrogen immobilization
in decompsing needle litter at variable
carbon:nitrogen ratios. Ecology 64: 63-67.

Bernard, J. M. and F. A. Bernard. 1977. Winter standing
crop and nutrients contents in five central New York
wetlands. Torreya 104: 57-59.

Bernard, J. M. and F. P. Bernard. 1973. Winter biomass
in Typha glauca Godr. and Sparganium eurycarpum
Engelm. Torreya 100: 125-131.

Bernard, J. M. and E. Gorham. 1978. Life history
aspects of primary production in sedge wetlands.
IN: R. E. Good, D. F. Whigham, and R. L. Simpson
(eds.). Freshwater wetlands: Ecological processes
and management potential. Academic Press, New York.
378 pp.

Biesboer, D. D. 1984. Nitrogen fixation associated with

natural and culitivated stands of Typha latifolia L.
(Typhaceae). Amer. J. Bot. 71: 505-511.

162

163

Boyd, C. E. 1971. The limnological role of aquatic
macrophytes and their relationship to reservoir
management. In: C. E. Hall (ed). Reservoir
fisheries and Limnology. American Fisheries Society
Special Publication No. 8, pp 153-166.

Boyd, C.E. 1970. Losses of mineral nutrients during
decomposition of Typha latifolia. Arch. Hydrobiol.
66: 511-517.

Brinson, M. M., A. E. Lugo and S. Brown. 1981. Primary
productivy, decomposition and consumer activity in
freshwater wetlands. Ann. Rev. Ecol. Syst. 12:
123-1610

Buchanan, T. J. and W.P. Sommers. 1969. Discharge
measurements at gaging stations (Chapter A8). IN:
Applications of hydraulics (Book 3). U.S.
Geological Survey, Arlington, VA. 65 pp.

Buchanan, T. J. and W.P. Sommers. 1968. Stage
measurements at gaging stations (Chapter A7). IN:
Applications of hydraulics (Book 3). U.S.
Geological Survey, Arlington, VA. 28 pp.

Buresh, R. J., and W. H. Patrick, Jr. Nitrate reduction
to ammonium in anaerobic soil. Soil Sci. Soc. Am.
J. 42: 913-918.

Carter, R. W. and J. Davidian. 1968. General procedure
for gaging streams (Chapter A6). IN: Applications
of hydraulics (Book 3). U.S. Geological Survey,
Arlington, VA. 13 pp.

Chamie, J. P. M. and C. J. Richardson. 1978. -
Decomposition in northern wetlands. IN; R. E. Good
D. F. Whigham, and R. L. Simpson (eds.).
Freshwater wetlands: Ecological processes and
management potential. Academic Press, New York.378

PP-

Coulson, J. C. and J. Butterfield. 1978. An
investigation of the biotic factors determining the
rates of plant decomposition on blanket bog. J. of
Ecol. 60: 631-650.

Davis, C. B. and A. G. van der Valk. 1978. Litter
decomposition in prairie glacial marshes. 1E5 R.
B. Good, D. F. Whigham, and R. L. Simpson (eds.).
Freshwater wetlands: Ecological processes and
management potential. Academic Press, New York. 378

PP-

164

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

de la Cruz, A. A. and B. C. Gabriel. 1974. Caloric,
elemental, and nutritive changes in Juncus
rcmerianus leaves. Ecology 55: 882-886.

 

Enwezor, W. 1976. The mineralization of nitrogen and
phosphorus in organic materials of varying C:N and
C:P ratios. Plant and Soil 44: 237-240.

Fetter, C. W., W. E. Sloey, and F. L. Spangler. 1978.
Use of a natural marsh for wastewater polishing. J.
Water Pollution Contr. Fed. 50: 290-307.

Fitter, A. J. and R. K. M. Hay. 1981. Environmental
physiology of plants. Academic Press, NY. 355 pp.

Frasco, B. A. and R. E. Good. 1982. Decomposition
dynamics of Spartina alterniflora and Spartina
tens in a New Jersey salt marsh. Amer. J. Bot.
69: 402-406.

Fulton, G. W. and W. T. Barker. 1981. Aboveground
biomass of selected wetlands on the Missouri Coteau.
Proc. N.D. Acad. Sci. 35:6.

Geis, J. W. The Great Lakes: demands, problems and
opportunities. pp. 529-542. IN: Forty fourth
North American Wildlife Conference.

Godshalk, G. L. and R. G. Wetzel. 1978. Decomposition
in the littoral zone of lakes. IN: R. E. Good, D.
F. Whigham, and R. L. Simpson (§d§.). Freshwater
wetlands: Ecological processes and management
potentals. Academic Press, NY. 378 pp.

Good, R. E. and N. F. Good. 1975. Vegetation and
production of the Woodbury Creek-Hessian Run
freshwater tidal marshes. Bartonia 43: 38-45.

Gorham, E. and M. G. Sommers. 1973. Seasonal changes in

the standing crop of two montane sedges. Can. J.
Bot. 51: 1097-1108.

Gosselink, J. G. and R. E. Turner. 1978. The role of
hydrology in freshwater wetland ecosystems. INI R.
E. Good, D. F. Whigham, and R. L. Simpson (eds.).
Freshwater wetlands: Ecological processes and
management potential. Academic Press, New York. 378

PP-

165

Harris, H. J., T. R. Bosley, and F. D. Roznik. 1977._
Green Bay's costal weetlands- a picture of dynamic
change. pp. 337-358. IN: C. B. DeWitt and E.

Scloway. Wetlands ecology, values and impacts.
University of Wisconsin, Madison, WI.

Harris, H. J., G. Fewless, M. Milligan, and W. Johnson.
1980. Recovery processes and habitat quality in a
freshwater coastal marsh following a natural
disturbance. IN: Proceedings of the Midwest
conference on wEt1ands values and management, St.
Paul, MN, June 17-19, 1981. MN Water Planning
Board, St. Paul.

Hemmond, H. F. 1983. The nitrogen budget of Thoreau's
bog. Ecology 64: 99-109.

Hoffnagle, J. R. 1980. Estimates of vascular plant
primary production in a West Coast salt
marsh-estuary ecosystem. Norhtwest Sci. 54: 68-79.

Hough, J. L. 1958. Geology of the Great Lakes. Univ.
of Illinois Press, Urbana, IL.

Hopkinson, C. S. and J. P. Schubauer. 1984. Static and
dynamic aspects of nitrogen cycling in the salt

gzrsg6graminoid Spartina alterniflora. Ecology 65:

Jaworski, E., C. N. Raphael, P. J. Mansfield, and B. B.
Williamson. 1979. Impact of Great Lakes water
level fluctuations on coastal wetlands. Inst. Water

Research, Michigan State University, E. Lansing, MI.
351 pp.

Jervis, R. A. 1969. Primary production in the
freshwater marsh ecosystem of Troy Meadows, New
Jersey. Bull. Tor. Bot. Club. 96: 209-231.

Keefe, C. 1972. Marsh production: A summary of the
literature. Contr. Mar. Sci. 16: 163-181.

Keeney, D. H. 1973. The nitrogen cycle in
sediment-water systems. J. Environ. Qual. 2:
15-29.

Klopatek, J. M. 1978. Nutrient dynamics of freshwater
riverine marshes and the role of emergent
macrophytes. IN; R. E. Good, D. F. Whigham, and R.
L. Simpson (eds). Freshwater wetlands: Ecological

processes and management potential. Academic Press.
195-216.

166

Klopatek, J. M. 1975. The role of emergent macrOphytes
in mineral cycling in a freshwater marsh. pp.
367-393. IN; F. G. Howell, J. B. Gentry and H. M.
Smith (eds.). Mineral cycling in southeastern
ecosystems.

Klopatek, J. M. and F. W. Stearns. 1978. Primary
productivity of emergent macrophytes in a Wisconsin

freshwater marsh ecosystem. Amer. Mid. Nat. 100:
320-332.

Knutilla, R. L. 1967. Flow characteristics of Michigan
streams. Michigan Dept. of Conservation, Geol. Survey
Section, Lansing, MI, 337pp.

LaBaugh, J.W. and T.C. Winter. 1984 The impact of uncertainties
in hydrologic measurements on phosphorus budgets and
empirical models of two Colorado reservoirs. Limnol.
Oceangr. 29: 322-339.

Lee, G. P., E. Bentley, and R. Amundson. 1975. Effects
of marshes on water quality. IN; A. D. Hasler
(ed.). Coupling of land and water systems.
Springer-Verlag, New York.

Likens, G. E. 1975. Nutrient flux and cycling in
freshwater systems. pp. 314-348. IN; F. G.
Howell, J. B. Gentry and H. M. Smith (eds.).
Mineral cycling in southeastern ecosystems. U.S.
ERDA, Washington, D.C.

Lindsley, D., T. Schuck, and F. Stearns. 1977. Primary
productivity and mineral regimes in a northern
Wisconsin marsh. IN; C. B. DeWitt and E. Soloway
(eds). Wetlands ecology, values and impacts.
University of Wisconsin, Madison, WI. 38-52.

Mason, C. F. and R. J. Bryant. 1972. Production and
nutrient content of Phragmites communis Trin. and
Typha latifolia L. J. Ecology 63: 71-95.

 

McRoy, C. P. and R. J. Barsdate. 1970. Phosphate
absorption in eel grass. Limnol. Oceanogr. 15:
6-130

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.

Mellilo, J.M., R.J. Naiman, J.D. Aber, and A.E. Linkins. 1984.
Dynamics of plant litter decay in northern streams. Bull.
Mar. Sci. (in press).

167

Nelson, D. and L. Sommers. 1973. Determination of total
N in plant materials. Agronomy J. 65: 109-112

Nichols, D. S. and D. R. Keeney. 1973. Nitrogen and
phosphorus release from decaying water milfoil.
Hydrobiologia 42: 509-525.

Nixon, S. W. 1980. Between coastal marshes and coastal
waters - A review of twenty years of specuation and
research on the role of salt marshes in estuarine
productivity and water chemistry. IN; P. Hamilton
and K. MacDonald (eds.). Estuarine and wetland
processes with emphasis on modelling. Plenum Publ.
Corp., N.Y.

NCAA. 1983. Climatological Data annual summary,
Michigan (v. 98). Nat. Climatic Data Center,
Ashville, NC. 26 pp.

NCAA/NOS. 1984. Great Lakes water levels, N/OMS124.
U.S. Dept. Commerce, Rockville, MD.

Odum, W. E. and M. A. Heywood. 1978. Decomposition of
intertidal freshwater marsh plants. IN; R. B. Good
D. F. Whigham, and R. L. Simpson (eds.). Freshwater
wetlands: Ecological processes and management
potential. Academic Press, New York. 378 pp.

Olson, J. S. 1963. Energy storage and the balance of
producers and decomposers in ecological systems.
Ecology 44: 322-331.

Penfound, W. T. 1956. Priamry production of vascular
aquatic plants. Limnol. Oceanogr. 1: 92-101.

Peverly, J. H. 1982. Stream transport of nutrients
through a wetland. J. Environ. Quality 11: 38-43.

Pomeroy, L.R. and R.G. Wiegert. 1981. The ecology of a
salt marsh. Springer-Verlag, NY. 271 pp.

'Prentki, R. T., T. D. Gustafson, and M. S. Adams. 1978.
Nutrient movements in lakeshore marshes. IN; R. E.
Good, D. F. Whigham, and R. L. Simpson (eds.).
Freshwater wetlands: Ecological processes and
management potential. Academic Press, New York. 378

PP-

Puriveth, P. 1980. Decomposition of emergent
macrophytes in a Wisconsin marsh. Hydrobiol. 72:
231-242.

168

Reddy, K. R. and W. H. PATRICK. 1977. Effects of
placement and concentration of applied NH4+-N on

nitrogen loss from flooded soil. Soil Science 123:
142-148.

Richardson, C. J., D. L. Tilton, J. A. Kadlec, J. P. M.
Chamie and W. A. Wentz. 1978. Nutrient dynamics of
a northern wetland ecosystem. IN; R. E. Good, D.
F. Whigham, and R. L. Simpson (eds.). Freshwater
wetlands: Ecological processes and management
potential. Academic Press, New York. 378 pp.

Richardson, C. J. and G. E. Merva. 1976. The chemical
composition of atmospheric precipitation from
selected stations. Water, Air, and Soil Pol. 6:
385-393.

Scheider, W. A., J. J. Moss and P. J. Dillon. 1979.
Measurement and uses of hydraulic and nutrient
budjets. IN: Lake Restoration. Proceedings of a
national conference, August 22-24, 1978,
Minneapolis, MN. U.S. EPA 440/5-79-001, Washington,
D.C. 254 pp.

Sculthorpe, C. D. 1967. The biology of aquatic vascular
plants. Arnold, London. 610 pp.

Seeling, W. H. and R. M. Sorensen. 1977. Hydraulics of
Great Lakes inlets. (CERRE-CS, Report # TP 77-8).
U.S. Dept. of the Army, Fort Belvoir, VA. 69 pp.

Shaver, G. R. and J. M. Mellilo. 1984. Nutrient
budgets of marsh plants: Efficiency concepts and
relation to availability. Ecology 65: 1491-1510.

Simpson, R. L., D. F. Whigham, and R. Walker. 1978.
Seasonal patterns of nutrient movement in a
freshwater tidal marsh. IN; R. B. Good, D. F.
Whigham, and R. L. Simpson (eds.). Freshwater
wetlands: Ecological processes and management
potential. Academic Press, New York. 378 pp.

Smirnoff, N. and R. M. M. Crawford. 1982. Variation
in the structure and response to flooding of root
aerenchma in some wetland plants. Bot. Gaz. 51:
237-248.

Steel, R. G. D., and J. H. Torrie. 1980. McGraw-Hill,
NY. 633 pp.

169

Tchobanoglous, G. and G. L. Culp. 1980. Wetland systems
for wastewater treatment: An engineering assessment.
IN: S. C. Reed and R. K. Bastian (eds.),
Aquaculture systems for wastewater treatment: An
engineering assessment. 430/9-80-007, MOD-68, U.S.
EPA, Washigton, D.C.

Tessier, C. A., A. Aubin, and D. Chenevier. 1984. Les
facteurs elevation et distance dans la structure et
la productivite d'une communaute riveraine de
macrOphyte. Can. J. Bot. 62: 2260-2266.

Tessier, C., A. Marie, and A. Aubin. 1981. Etude de la
vegetation des zones riveraines de l'archipel des
Centiles du fleuve Saint-Laurent, Quebec. Can J.
Bot. 59: 1526-1536.

Ulrich, K. E. 1984. The role of emergent aquatic
vegetation in the nutrient budgets of managed
marshes. Ph. D. Dissertation, Michigan State
University, E. Lansing, MI. 113 pp.

Ulrich, K. E. and T. M. Burton. 1985. The effects of
nitrate, phOSphate and pottassium fertilization on
growth and nutrient uptake patterns of Phragmites
australis (Cav.) Trin. ex. Steudel. Aquatic Botany
21: 53-62.

 

U.S. EPA. 1979. Handbook for analytical quality control
in water and wastewater laboratories.

U.S. EPA. 1979. Methods of chemical analysis of water
and wastes. EPA-600/4-79-020, U.S. EPA, Cincinnati, OH

. Valiela, I., and J. M. Teal. 1978. Nutrient limitation in
salt marsh vegetation. pp. 547-563. IN; Ecology of
halophytes. Reimhold, R. J., and W. H. Queen
(eds.). Academic Press, NY. ‘

van der Valk, A. G. 1981. Succession in wetlands: a
Gleasonian approach. Ecology 62: 688-696.

van der Valk, A. G., and C. B. DAVIS. 1980. The impact
of a natural drawdown on the growth of four emergent
species in a prairie glacial marsh. Aquatic Bot. 9:
301-322.

van der Valk, A. G. and C. B. Davis. 1979. A
reconstruction of the recent vegetational history
of a prairie marsh, Eagle lake, Iowa, from its
seed bank. Aquatic Bot. 6: 29-51.

170

van der Valk, A. G., C. B. Davis, J. L. Barker and C.
E. Beer. 1979. Natural freshwater marshes as
nitrogen and phosphorus traps for land run967. A
vegetative key to the genera of submersed and
floating aquatic vascular plants of Michigan.
Michigan Botanist 6:35-50.

Wallsten, M. 1980. Effects of the growth of Elodea
canadensis Michx. in a shallow lake (lake
Tamnaren, Sweden.). Developments in Hydrobiology
3: 139-146.

Westlake, D. F. 1982. The primary production of
water plants. _IN : J. J. Symoens, S. S. Hopper
and P. Compere, (eds.). Studies on aquatic
vascular plants. Royal Botanical Society of
Belgium, Brussels. 424 pp.

Westlake, D. F. 1963. Compraison of plant
productivity. Biol. Rev. 38: 385-425.

Wetzel, R.G. 1975. Limnology. W.B. Sanders Co.,
Philadelphia, PA.

 

rum-(man STATE UNIV. LIBRARIES
1NWW111111111111")W”MWWWW“
31293107075958