METHANE AND CARBON DIOXIDE FLUXES IN CREATED RIPARIAN WETLANDS IN THE MIDWESTERN USA: EFFECTS OF HYDROLOGIC PULSES, EMERGENT VEGETATION AND HYDRIC SOILS
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
The Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Anne E. Altor, B.S.
* * * * *
The Ohio State University
2007
Dissertation Committee
Dr. William J. Mitsch, Advisor
Dr. Virginie L. Bouchard Approved by
Dr. Warren A. Dick
Dr. Rattan Lal Advisor Environmental Science Graduate Program
ABSTRACT
Wetlands are important ecosystems involved in the global carbon cycle as producers
and consumers of the greenhouse gases methane and carbon dioxide. The global
warming potential of methane – 23 times greater that that of carbon dioxide over a 20
year time horizon – warrants examination of the dynamics controlling its emission from
temperate zone wetlands created and restored for habitat replacement and water quality
improvement. Research on carbon dynamics in created and restored ecosystems can
enable greater understanding of management practices to promote carbon sequestration in
these ecosystems. In the research conducted for this dissertation, ecosystem and
mesocosm-scale investigations were carried out in experimental riparian wetlands of the
Midwestern USA, with hydrology, vegetation and soils as independent variables.
Methane and carbon dioxide fluxes were the dependent variables of interest in each of the
studies. In a one-year field study, flood-pulse hydrology typical of floodplains in the
Midwestern USA was simulated in two one-hectare riparian marshes. Methane fluxes
were measured from February-December using non-steady-state chambers located in
marsh zones with and without emergent vegetation in which soils were intermittently
exposed and inundated, and in permanently inundated wetland areas. Annual methane
fluxes from intermittently flooded zones were 30% of fluxes from permanently inundated
-2 -1 wetland areas, which emitted ~42 g CH4-C m yr . Average growing season rates of
ii methane flux from intermittently flooded zones with and without macrophytes did not
-2 -1 differ significantly (~3.5 mg CH4-C m h ), but both were significantly less than those
-2 -1 from permanently inundated areas (~8 mg CH4-C m h ).
In an extension of the first, one-year study, methane and carbon dioxide flux rates
were measured in the morning, afternoon and nighttime in the same experimental
marshes over a second year during which hydrologic inflow was maintained at a
relatively constant rate. Under these ‘steady-flow’ conditions, approximately the same
total volume of inflow was delivered as during the flood-pulse year. Gas flux rates were
measured in the same three wetland zones: continuously inundated areas; edge zones with
emergent macrophytes; and edge zones without emergent macrophytes. Methane fluxes
between the two years were not significantly different in edge zones with and without
emergent vegetation, but were twice as high in continuously inundated zones during the
steady-flow year. There was no apparent relationship between emergent vegetation and
methane flux, as mean flux rates were not significantly different in either year in edge
zones where emergent vegetation was removed, compared to edge zones containing
emergent vegetation. Continuously inundated wetland zones emitted methane from
summer through fall, while in edge zones methane fluxes were only substantial in spring
and summer. Neither daytime rates of carbon dioxide uptake or nighttime rates of
respiration were significantly different between the years for any wetland zone. When
CO2 flux rates (daytime uptake plus nighttime respiration) were normalized for solar
radiation and day length, solar efficiency was found to be comparable between the steady
flow and pulsed years. Methane fluxes were more strongly affected than carbon dioxide fluxes by the differences in hydrology, but only in the deeper areas of the wetlands.
iii T he hydrology and physiochemical properties of soils forming the foundation for
created and restored wetlands determine what processes are likely to occur in these
systems. In a companion, replicated study, effects of intermittent vs. continuous
inundation, and hydric vs. non-hydric soils on fluxes of methane and carbon dioxide were
investigated using 20 wetland mesocosms. The hydrologic treatments represented
contrasting wetland restoration scenarios, and the soil treatments represented newly
created and established wetlands. Hydric soils and continuously inundated treatments
exhibited the greatest methane flux, while intermittently inundated conditions reduced
methane fluxes significantly from hydric soils. Methane fluxes were not affected
significantly by hydrologic treatment in mesocosms containing non-hydric soils. No
relationship was observed between emergent vegetation and methane flux, and carbon
dioxide and methane fluxes were not directly correlated. However, the highest rates of
both CO2 uptake and CH4 flux were observed in treatments with steady-flow hydrology.
As part of soil organic matter determination, the effect of combustion time was examined and a significant difference in soil organic matter content was found from one to three hours of combustion in both hydric and non-hydric soils ashed at 550°C. Microbially available organic carbon content was significantly greater in hydric soils than non-hydric soils, despite similar organic matter contents in the contrasting soil types. Methane fluxes from created wetland mesocosms fell within the ranges reported for comparable, natural wetlands.
iv
DEDICATION
“Yesterday is already a dream and tomorrow is only a vision, but today well-lived makes every yesterday a dream of happiness, and every tomorrow a vision of hope.”
Sanskrit proverb
“What I love is near at hand, Always, in earth and air”
Theodore Roethke
v
ACKNOWLEDGMENTS
There are many individuals whose support and encouragement during my graduate work was invaluable.
My advisor and dissertation committee made formative contributions to my research
and academic development. I appreciate the many opportunities made available to me by
my advisor, Dr. William Mitsch. Working at the ORWRP was never dull, and I am
thankful for the independence I was given to pursue my research. Working under Bill
provided me the opportunity to encounter Swamp Thing first hand, and on a more serious
note, to interact with many interesting researchers in the fields of wetland ecology and
ecological engineering. Bill’s keen insight and editing were very valuable during the
writing process. Discussions with Dr. Virginie Bouchard gave a jump-start to my research methods and helped me substantially as I was beginning to work on interpreting field data. I appreciate the time Virginie took to read and discuss my work. Dr. Warren
Dick provided a thorough reading and critique of my research plan, and helped me to identify avenues of soil research that would complement the gas flux studies. Thanks to
Warren I was made aware of the Soil Science minor option and opportunities available therein. Discussions with Dr. Rattan Lal about the mesocosm study were helpful, and his rigorous emphasis on soil physical properties helped me to gain a solid understanding of this important foundation for ecological processes.
vi Thank you to my parents, Nancy and Chuck Cladel, who have always been there for
me and encouraged me to pursue work that I find fulfilling, and who have participated all
along the way in the life and educational journey I have been on. Mom, I often think of
your help in building the mesocosm chambers, when we were sawing up PVC and plastic
boxes all day, and how you insisted we not take a break until the work was done. I’ve
come back to that example often in my mind, and sometimes in practice! Our many
discussions of science and ideas have been influential and inspirational to me. Pop, I
appreciate your genuine interest in and support for what I’m doing, and the example you
provide of dedication to your work and diverse interests. I also am thankful to my
Grandma McMurry and Grandma Cladel for being important and supportive people in
my life, each a positive example in her own way.
One of the very best parts of my graduate experience has been the friendships and
creative collaborations made with fellow lab mates and researchers. Maria E. Hernandez has been a cherished friend, mentor and collaborator, and our friendship and intellectual discussions have been an integral part of my life at and away from work. I’ll always be your friendly Buscapleitos! Chris Anderson provided friendship, knowledge and many valuable research discussions throughout, and I also appreciate the good times and ideas shared with friends and lab mates Dan Fink, Amanda Nahlik, Debra Gamble, Jennie
Morgan, Blanca Bernal, Cassie Tuttle, and Chen Huang. Lab and office friends whose hard work and help I have appreciated include Kyle Chambers, Amber Hanna, Sherr Vue,
Natalie Pinheiro, Jan Thompson, Brittany Cleveland, Monica Noon, Ryan Younge and
Jeremiah Miller. A special thanks goes to Kyle Chambers who spent hours as a volunteer and then lab tech helping me in the field amidst hungry chiggers and horseflies, beautiful
vii days and muck. Without his help, I could not have sampled as extensively, and would
have missed some great conversations! Dr. Li Zhang deserves a special thank you. If
there was a problem, Li was there to help, immediately. If there was ever a need to be
filled, Li made sure it was taken care of. Chris Holloman’s help with statistical methods
was timely and much appreciated. Finally, I appreciate design and construction ideas and
help provided by Mark Warren, as well as his encouragement to pursue my education.
Funding for my research and education was provided by several agencies including
U.S. Department of Agriculture NRI CSREES Award 2003-35102-13518, an Ohio
Agricultural Research and Development Center of The Ohio State University Payne
Grant, a Rhonda and Paul Sipp Wetland Research Award, the Environmental Science
Graduate Program, and the Schiermeier Olentangy River Wetland Research Park.
viii
VITA
January 19, 1970 ...... …. Born - Anchorage, Alaska USA
2003...... B.S. Environmental and Forest Biology, State University of New York College of Environmental Science and Forestry, Syracuse, NY. Summa Cum Laude
2003-2007………………...... Graduate Research and Teaching Assistant, The Ohio State University Environmental Science Graduate Program, School of Environment and Natural Resources
PUBLICATIONS
Peer-reviewed journal articles
Altor, A. E., and W. J. Mitsch, 2006. Methane flux from created riparian marshes: relationship to intermittent vs. continuous inundation and emergent macrophytes. Ecological Engineering 28: 224-234.
Mitsch W.J. L. Zhang, C. Anderson, A.E. Altor, and M.E. Hernandez, 2005. Creating riverine wetlands: Ecological succession, nutrient retention, and pulsing effects. Ecological Engineering 25: 510-527.
Other publications, including those in Olentangy River Wetland Research Park Annual Reports
Altor, A. E., and W. J. Mitsch, 2006. Methane and carbon dioxide fluxes in wetland mesocosms: Relationships to hydrology and soils. Pp. 199-211 in Mitsch, W. J., L. Zhang, C. L. Tuttle, and K. Jones (Editors), Olentangy River Wetland Research Park at the Ohio State University 2005 Annual Report.
ix Altor, A. E., and W. J. Mitsch, 2005. Methane flux from created marshes: Effects of intermittent vs. continuous inundation and emergent macrophytes. Pp. 103-108 in Mitsch, W. J., L. Zhang, and A. E. Altor (Editors), Olentangy River Wetland Research Park at the Ohio State University 2004 Annual Report.
Mitsch, W. J., C. J. Anderson, M. E. Hernandez, A. E. Altor, and L. Zhang, 2005. Net primary productivity of macrophyte communities in the experimental marshes after eleven growing seasons. Pp. 57-60 in Mitsch, W. J., L. Zhang, and A. E. Altor (Editors), Olentangy River Wetland Research Park at the Ohio State University 2004 Annual Report.
Altor, A. E., and W. J. Mitsch, 2004. Characterization of the water quality and biota of a stormwater wetland one year after its creation. Pp. 199-205 in Mitsch, W. J., L. Zhang, and C. L. Tuttle (Editors), Olentangy River Wetland Research Park at the Ohio State University 2003 Annual Report.
Mitsch, W. J., C. J. Anderson, M. E. Hernandez, A. E. Altor, and L. Zhang, 2004. Net primary productivity of macrophyte communities after ten growing seasons in experimental marshes. Pp. 75-78 in Mitsch, W. J., L. Zhang, and C. L. Tuttle (Editors), Olentangy River Wetland Research Park at the Ohio State University 2003 Annual Report.
Hernandez, M. E., A. E. Altor, and W. J. Mitsch, 2004. Below-ground biomass and nitrogen accumulation by four dominant wetland plant species in the experimental wetlands. Pp. 99-104 in Mitsch, W. J., L. Zhang, and C. L. Tuttle (Editors), Olentangy River Wetland Research Park at the Ohio State University 2003 Annual Report.
Halpern, A. D., C. A. Boesse and A. E. Altor, 2001. Bad seeds: an introduction to invasive plants. Clearwaters, New York Water Environment Association, Inc. Spring issue.
FIELD OF STUDY
Major field: Environmental Science Specialization in wetland ecology and biogeochemistry Minor: Soil Science
x
TABLE OF CONTENTS
Page
Abstract……………………………………………………………………………………ii
Dedication…………………………………………………………………………………v
Acknowledgments………………………………………………………………………..vi
Vita……………………………………………………………………………………….ix
List of Figures…………………………………………………………………………...xiv
List of Tables………………………………………………………………………… xviii
Chapters:
1. INTRODUCTION ...... 1 1.1 Wetlands and the global carbon cycle ...... 1 1.2 Research goals and objectives ...... 2 1.3 Wetland hydrology and carbon cycling ...... 3 1.4 Wetland vegetation and methane fluxes ...... 5 1.5 Hydric soils and carbon dynamics ...... 6 1.6 Literature cited...... 7
2. METHANE FLUX FROM CREATED RIPARIAN MARSHES: RELATIONSHIP TO INTERMITTENT VS. CONTINUOUS INUNDATION AND EMERGENT MACROPHYTES ...... 12 2.1 Abstract...... 12 2.2 Introduction ...... 13 2.3 Materials and methods ...... 15 2.3.1 Study site ...... 15 2.3.2 Hydrology...... 18 2.3.3 Gas sampling ...... 18 2.3.4 Analysis...... 20 2.4 Results ...... 22
xi 2.4.1 Hydrology and vegetation...... 22 2.4.2 Methane flux ...... 23 2.5 Discussion ...... 25 2.5.1 Hydrology and methane fluxes ...... 25 2.5.2 Emergent macrophytes and methane flux...... 27 2.5.3 Comparison with other wetland methane studies ...... 28 2.6 Conclusions...... 29 2.7 Acknowledgements...... 30 2.8 Literature cited ...... 30
3. PULSING HYDROLOGY, AND METHANE AND CARBON DIOXIDE FLUXES IN CREATED MARSHES: A 2-YEAR ECOSYSTEM STUDY...43 3.1 Abstract...... 43 3.2 Introduction...... 44 3.3 Methods...... 49 3.3.1 Study site...... 49 3.3.2 Hydrology ...... 50 3.3.3 Gas sampling ...... 50 3.3.4 Analysis of data ...... 55 3.4 Results...... 56 3.4.1 Hydrology ...... 56 3.4.2 Vegetation ...... 56 3.4.3 Methane fluxes ...... 57 3.4.4 Carbon dioxide fluxes and solar radiation ...... 58 3.5 Discussion...... 59 3.6 Conclusions...... 64 3.7 Acknowledgements...... 65 3.8 Literature cited...... 66
4. METHANE EMISSIONS AND CARBON DIOXIDE FLUXES FROM WETLAND MESOCOSMS: EFFECTS OF HYDROLOGIC REGIME AND HYDRIC SOILS ...... 81 4.1 Abstract...... 81 4.2 Introduction...... 82 4.3 Methods...... 86 4.3.1 Site description...... 86 4.3.2 Study design ...... 86 4.3.3 Hydrology ...... 88 4.3.4 Gas sampling...... 89 4.3.5 Soil sampling ...... 91 4.3.6 Data analysis ...... 94 4.4 Results ...... 95 4.4.1 Hydrology and methane flux...... 95 4.4.2 Soil physical properties and methane flux...... 97
xii 4.4.3 Importance of vegetation and carbon dioxide fluxes in relation to methane flux...... 99 4.5 Discussion ...... 101 4.5.1 Methane fluxes in relation to hydrology and hydric vs. non-hydric soils ...... 101 4.5.2 Importance of dynamic hydrology in relation to methane flux ...103 4.5.3 Soil physiochemical properties in relation to methane flux...... 103 4.5.4 Relationship between vegetation, carbon dioxide and methane flux ...... 105 4.5.5 Methodological issues for soil analyses...... 106 4.6 Conclusions...... 108 4.7 Acknowledgements ...... 109 4.8 Literature cited ...... 110
5. CONCLUSIONS...... 126 5.1 Hydrology and methane fluxes in experimental wetlands and mesocosms ...... 126 5.2 Hydrology and carbon dioxide fluxes in experimental wetlands and mesocosms ...... 128 5.3 Vegetation and diurnal methane fluxes in experimental marshes and mesocosms ...... 129 5.4 Methodological issues for wetland studies ...... 130 5.4.1 Non-steady-state chamber design ...... 130 5.4.2 Soil organic matter determination...... 131 5.4.3 Hydrologic pulsing experiments...... 131
Bibliography ...... 132
Appendices Appendix A: CH4 and CO2 flux data, 2004 ...... 145 Appendix B ...... 150 Appendix C ...... 162
xiii
LIST OF FIGURES
Figure Page
1.1 Pathways of methane and carbon dioxide uptake and release from wetlands. CH4 can be released from emergent vegetation by passive diffusion or by convective flow. CO2 is taken up by plants, and O2 diffuses to the root zone through aerenchyma, where it may be used by methanotrophs to oxidize methane to CO2. Plant roots release organic exudates that serve as substrates for methanogenesis, and methane not oxidized in the unsaturated sediment zone is released to the atmosphere by diffusion, or by bubble ebullition when the partial pressure of methane in porewater is greater than its partial pressure in the sediments……… 11
2.1 The Schiermeier Olentangy River Wetland Research Park on the campus of the Ohio State University, Columbus, USA. Research was conducted in the kidney- shaped “experimental wetlands 1 and 2”…...... 35
2.2 Locations of gas-sampling chambers in the two experimental wetlands. Straight lines represent permanent boardwalks…………………………………………...36
2.3 Inflow to each experimental wetland in 2004. Each bar represents the average pumping rate for a 1-week period………………………………………………..37
2.4 A. Average (with standard error) soil temperature between 5-10 cm depths for each wetland zone on each sampling date; B. Percentage of intermittently flooded plots that were inundated on each sampling date; and C. Average (with standard error) CH4-C flux rate for each wetland zone on each sampling date…………...38
2.5 Seasonal and annual methane flux for each wetland zone. Different letters indicate a significant difference in methane emissions between seasons or wetland zones. Bars represent standard error. Sample sizes are as follows: intermittently flooded with emergent macrophytes—growing season n=185, non-growing season n=87; intermittently flooded without emergent macrophytes —growing season n=115, non-growing season n=57; permanently inundated—growing season n=111, non-growing season n=38……………………………………….39
2.6 Correlation between methane flux and the level of the water table above (positive numbers) and below (negative numbers) the soil surface for A: permanently
xiv inundated wetland areas, B: intermittently flooded wetland zones containing emergent macrophytes, and C: intermittently flooded wetland zones in which emergent macrophytes were removed...... 40
2.7 Average methane flux rates when soil in pulsed wetland zones was inundated vs. exposed during sampling (A), and overall rates for all wetland zones (B) during the 2004 growing season. Different letters within a panel indicate a significant difference in average methane flux rates between conditions. Bars indicate standard error. The number of samples is as follows: intermittently flooded with emergent macrophytes —inundated n=114, exposed n=71; intermittently flooded without emergent macrophytes—inundated n=70, exposed n=45; permanently inundated n=111 ...... 41
2.8 Diurnal patterns in methane flux for each wetland zone. Different letters indicate a significant difference in methane emissions between time of day or wetland zone. Bars represent standard error. Gas samples were taken within the indicated time intervals. The number of samples is as follows: intermittently flooded with emergent macrophytes —7-11 a.m. n=59, 1-5 p.m. n=61, 8 p.m.-12 a.m. n=65; intermittently flooded without emergent macrophytes—7-11 a.m. n=36, 1-5 p.m. n=37, 8 p.m.-12 a.m. n=42; permanently-inundated—7-11 a.m. n=35, 1-5 p.m. n=38, 8 p.m.-12 a.m. n=38...... 42
3.1 Wilma H. Schiermeier Olentangy River Wetland Research Park (ORWRP). Research described in this study was conducted in ‘experimental wetland 1’ and ‘experimental wetland 2.’ ...... 69
3.2 Hydrology over the two-year study period illustrating flood pulses in 2004 and steady-flow conditions in 2005...... 70
3.3 Schematic of the two experimental wetlands at the ORWRP at The Ohio State University, Columbus, and the three wetland zones in which gas sampling was carried out. “Edge” zones contained chambers with and without emergent macrophytes; “Marsh” zones in the inflow exhibited dynamics comparable to edge zone chambers with emergent macrophytes; “Marsh” zones in the outflow were comparable to chambers in continuously-inundated zones. “CI” refers to continuously inundated zones in which floating chambers were deployed. Straight lines represent permanent boardwalks located throughout the wetlands.70
3.4 Non-steady-state chamber design, illustrating A: HDPE chamber base and PVC frame left in wetlands for duration of experiment, and B: placement of plastic bag over chamber before gas sampling ...... 71
3.5 Carbon dioxide flux rates measured within non-steady-state chambers according to sampling time. Each point represents a flux rates measured after approximately a half hour (Y-axis) and an hour (X axis) of closing an individual chamber. This
xv relationship was used to normalize 2004 data before they were compared to 2005 data...... 71
3.6 Percent cover of dominant plant species in each wetland zone during the pulsing year (2004) and the steady-flow year (2005)...... 72
3.7 Mean methane flux rates within each wetland zone during the flood pulsed year (2004). Different letters represent a significant difference (p<0.05) between wetland zones or seasons. Bars represent standard error...... 73
3.8 Mean methane flux rates within each wetland zone during the steady flow year (2005). Different letters represent a significant difference (p<0.05) between wetland zones or seasons. Bars represent standard error...... 74
3.9 Relationship between soil temperature (°C) and methane flux rates (mg CH4-C m-2 h-1) for all wetland zones during the flood pulsed and steady flow years...... 75
3.10 Relationship between soil temperature (°C) and nighttime carbon dioxide flux -2 -1 rates (mg CO2-C m h ) for all wetland zones during the flood pulsed and steady flow years...... 75
3.11 Relationship between water depth (cm above or below soil surface) and nighttime -2 -1 carbon dioxide flux rates (mg CO2-C m h ) for all wetland zones during the flood pulsed and steady flow years...... 76
3.12 Mean growing- and non-growing season carbon dioxide uptake for the pulsing and steady-flow years. Different letters represent a significant difference (p<0.05) between wetland zones or years. Bars represent standard error. Flux rates are shown as negative numbers to represent removal of carbon dioxide from the chamber environment...... 77
4.1 Experimental mesocosm design, illustrating a) dimensions, French drains, stand pipes for adjusting water level, and outflow opening; and b) gas sampling chamber design (illustration 4.1b. by Jung-Chen Huang)...... 118
4.2 Mean methane emissions over the entire study period for each treatment. Number of samples is as follows: non-hydric continuously inundated: 214; non-hydric pulsed: 171; hydric continuously inundated: 176; hydric pulsed: 180. Bars represent standard error……………...... 119
4.3 Mean daily water level, and mean diurnal methane flux on each sampling date for each treatment: a: Hydric pulsed; b: Hydric continuously inundated; c: Non- hydric pulsed; d: Non-hydric continuously inundated...... 120
xvi 4.4 Percent soil organic matter (SOM) according to number of hours combusted at 550°C. Different letters indicate a significant difference in SOM between combustion times. Bars represent standard error...... 121
4.5 Cold-water extractable organic matter (CWEOM) and hot-water extractable organic matter (HWEOM) for hydric and non-hydric soils, measured one year after mesocosms were inundated. Different letters indicate a significant (p≤0.05) difference among treatments. Bars represent standard error ...... 121
4.6 Mean rates of daytime CO2-C uptake and nighttime efflux for each treatment over the entire gas-sampling period (April – September 2005). Different letters indicate a significant difference between treatments for either uptake or efflux 120
xvii
LIST OF TABLES
Table Page
3.1 General Linear Model for methane flux from each wetland zone, for the pulsed and steady flow years...... 79
3.2 General Linear Model for nighttime CO2 flux from edge zones with emergent macrophytes and continuously inundated wetland zones, for the pulsed and steady flow years...... 79
-2 -1 3.3 Mean, minimum and maximum methane fluxes (mg CH4-C m h ) from each wetland zone in each season in the pulsing (2004) and steady flow (2005) years 80
3.4 Mean, minimum and maximum soil temperature (°C) from each wetland zone in each season in the pulsing (2004) and steady flow (2005) years...... 81
4.1 Univariate general linear model for methane flux from mesocosms, indicating significant terms and interactions ...... 122
4.2 Physical and chemical characteristics of hydric and non-hydric soils...... 123
4.3 Significant results from regressions between soil temperature and methane flux from mesocosm treatments, and analysis of variance results for each regression123
4.4 Plant species identified in each treatment (wetland indicator status in parentheses‡) during the gas sampling period. An asterisk indicates the species was planted in all mesocosms; capitalized, bold X indicates the species was dominant in a given treatment...... 124-125
4.5 Univariate general linear model for daytime carbon dioxide uptake in mesocosms indicating significant terms and interactions …………………………………..125
4.6 Univariate general linear model for nighttime carbon dioxide efflux from mesocosms, indicating significant terms and interactions……………………..126
xviii -2 -1 A.1 Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2004...... 148
-2 -1 A.2 Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2004...... 149
-2 -1 A.3 Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the growing season, 2004...... 150
-2 -1 A.4 Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2004…....150
-2 -1 A.5 Methane fluxes (mg CH4-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the growing season, 2004. Fluxes in red on August 10 were not used in the analysis...... 151
-2 -1 A.6 Methane fluxes (mg CH4-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the non-growing season, 2004… .152
-2 -1 A.7 Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time ...... 153
-2 -1 A.8 Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time… ...... 154
-2 -1 A.9 Carbon dioxide fluxes (mg CO2-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time ...... 155
-2 -1 A.10 Carbon dioxide fluxes (mg CO2-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time ...... 155
-2 -1 A.11 Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time… ...... 156
-2 -1 A.12 Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the non-growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time ...... 157
xix -2 -1 B.1 Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2005...... 159
-2 -1 B.2 Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2005...... 160
-2 -1 B.3 Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the growing season, 2005 ...... 161
-2 -1 B.4 Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2005...... 161
-2 -1 B.5 Methane fluxes (mg CH4-C m h ) from continuously inundated zone chambers on each sampling date during the growing season, 2005...... 162
-2 -1 B.6 Methane fluxes (mg CH4-C m h ) from continuously inundated zone chambers on each sampling date during the non-growing season, 2005 ...... 163
-2 -1 B.7 Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2005...... 164
-2 -1 B.8 Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2005...... 165
-2 -1 B.9 Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers without emergent macrophytes on each sampling date during the growing season, 2005...... 166
-2 -1 B.10 Carbon dioxide fluxes (mg CO2-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2005...... 166
-2 -1 B.11 Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the growing season, 2005...... 167
-2 -1 B.12 Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the non-growing season, 2005...... 168
xx -2 -1 C.1 Methane fluxes (mg CH4-C m h ) from mesocosms with non-hydric soils and steady-flow hydrology on each sampling date (2005)...... 170
-2 -1 C.2 Methane fluxes (mg CH4-C m h ) from mesocosms with hydric soils and steady-flow hydrology on each sampling date (2005)...... 171
-2 -1 C.3 Methane fluxes (mg CH4-C m h ) from mesocosms with non-hydric soils and flood pulse hydrology on each sampling date (2005)...... 172
-2 -1 C.4 Methane fluxes (mg CH4-C m h ) from mesocosms with hydric soils and flood pulse hydrology on each sampling date (2005) ...... 173
-2 -1 C.5 Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with non-hydric soils and steady-flow hydrology on each sampling date (2005) ...... 174
-2 -1 C.6 Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with hydric soils and steady-flow hydrology on each sampling date (2005)...... 175
-2 -1 C.7 Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with non-hydric soils and flood pulse hydrology on each sampling date (2005) ...... 176
-2 -1 C.8 Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with hydric soils and flood pulse hydrology on each sampling date (2005) …………………………..177
xxi
CHAPTER 1
INTRODUCTION
1.1 Wetlands and the global carbon cycle
Carbon is key to global climate, particularly when it takes the forms of carbon
dioxide (CO2) and methane (CH4). Rising atmospheric concentrations of these gases, beyond the range of natural variation, correlate closely with the onset of industrialization in the West, and increasing average global temperature over the last 200 years (IPCC,
2001). The concentration of methane (CH4), which has a global warming potential 23
times higher than that of carbon dioxide over a 100-year time horizon, has more than
doubled since the preindustrial age (Megonigal et al., 2004). Wetlands play an important
role in the global cycling of carbon, acting as both sources and sinks for this element.
The balance between carbon storage in and CH4 emission from wetlands is dependent
on numerous factors including climate (temperature), type of hydrologic regime (e.g.,
dynamic and fluctuating vs. static and ponded), nutrient status, type of vegetation present,
- +3 -2 and concentration of alternate electron acceptors including NO3 , Fe and SO4
(D’Angelo and Reddy, 1999; Whalen, 2005). Pathways of methane release from wetlands
1 include passive diffusion through soil or vascular vegetation, convective flow by
pressurized ventilation through plants with extensive arenchymatous systems, and
ebullition from the sediment surface (Walter et al., 1996) (Figure 1.1). Anoxic conditions
in wetland soils inhibit decomposition and foster accumulation of refractory organic
matter – a source of long-term carbon storage (Moore and Turunen, 2004). Of the
approximately 1500 – 2000 Gt (gigatonnes = 1015 g) C stored in terrestrial soils, it is
estimated that up to 25% is stored in wetlands (Schlesinger, 1997; Roulet, 2000; Mitra et
al., 2005). At the same time, oxygen-poor environments that are enriched in organic
matter are ideal habitats for methanogenic microorganisms, and wetlands are estimated to
contribute between 15 - 40% of annual global methane emissions to the atmosphere
(Bridgham et al., 2006).
1.2 Research goals and objectives
The goal of this dissertation is to examine spatial and temporal patterns of carbon
biogeochemistry in freshwater wetlands under both steady flow and flood pulsing
hydrologic conditions. The specific objectives of this research were as follows:
1) Examine the effects of a flood-pulsed hydrologic regime on spatial and temporal
patterns of methane emissions, and investigate the impact of emergent vegetation
on methane fluxes in created, riparian marshes in the Midwestern U.S. (Chapter
2).
2) Evaluate relationships between flood pulse vs. steady flow hydrology and
methane/carbon dioxide fluxes, and examine the relationship of emergent
2 vegetation to methane fluxes under the contrasting hydrologic regimes (Chapter
3).
3) Assess variation in methane and carbon dioxide fluxes in replicated mesocosm
wetlands with hydric and non-hydric soils, under flood-pulsed and continuously
inundated conditions, and relate these dynamics to soil physiochemical properties
(Chapter 4).
All research was conducted at the Wilma H. Schiermeier Olentangy River Wetland
Research Park (ORWRP) of The Ohio State University, Columbus, OH. Two 1 ha
experimental wetlands created in 1994 were used for a two-year ecosystem scale studies
2 of CH4 and CO2 fluxes, and twenty 0.9 m wetland mesocosms were used for a replicated two-year study of carbon dynamics under conditions of contrasting hydrology and soils.
1.3 Wetland hydrology and carbon cycling
The recognition of damage caused by extensive wetland loss and destruction in the
United States and elsewhere has led to implementation of laws to protect wetlands, and
mitigation requirements to compensate for their continued loss. Achieving functionality
in created and restored wetlands that compares to natural ecosystems is a goal that has
stimulated research and debate (Mitsch et al., 1998; Zedler and Callaway, 1999; Brooks
et al., 2005). Hydrology is central to wetland function, and is arguably the most
important physical parameter to ‘get right’ in wetland restoration and creation projects
(McKee and Faulkner, 2000; Mitsch and Jorgensen, 2004; Montalto and Steenhuis,
2004).
3 Many natural wetlands are characterized by dynamic hydrology, receiving inflow in
pulses that correspond to precipitation or melting of snow and ice. A paradigm for the
restoration of riparian wetlands is to reestablish flood pulses into these ecosystems (Galat
et al., 1998; Middleton, 1999). Junk et al. (1989) developed the flood pulse concept and
describe the ways in which the ‘moving littoral,’ formed when rivers and streams spill
into their floodplains, enhances nutrient cycling and movement of suspended solids, soil
formation and spatial/temporal heterogeneity. Spink et al (1998) discuss the complexity
of flooding regimes in relation to nutrient dynamics and plant growth, emphasizing that the time interval, duration and magnitude of flood pulses as well as site-specific factors
all influence nutrient cycling. Odum et al. (1995) maintain that, on a continuum of flood
pulses, there is an optimum magnitude for maximizing ecosystem productivity while
avoiding major disturbance to plants and soils.
While developed for river-floodplain systems, the general concept of hydrologic
pulsing is relevant to other types of wetlands as well. Many natural wetlands experience
dynamic hydrologic conditions that vary with the seasons, including fluctuations in water
depth above and below the soil, expanding and contracting wetland area, or alternating cycles of wetting and drying (Gilvear and Bradley, 2000). Restoring flood pulses to wetland ecosystems can take a number of forms, such as reconnecting floodplains with the river system in which they are located by notching levees to allow reentry of floodwater; situating wetlands in the landscape to capture periodic runoff events; placing simple weirs at the outflow for manipulation of water levels; establishing wetland elevation so that flood stage events are captured and stagnant pools are avoided; or a combination of these approaches.
4 Relationships between hydrologic pulses and methane emissions are not well
documented in the literature, nor are reports of methane fluxes from created or restored
wetlands (Schipper and Reddy, 1994; Tanner et al., 1997; Tuittila et al., 2000). Assuming
that soils and biotic communities will develop in response to the hydrology of the system
they are a part of, the balance between CO2 uptake and CH4 emissions will be regulated
in large part by the quality and delivery dynamics of water received by a wetland.
Maximizing primary productivity while minimizing methane emissions from created and
restored wetlands is a desirable functional outcome that warrants practical investigation
under experimental conditions (Whiting and Chanton, 2001). Chapters 2 and 3
investigate relationships between hydrologic conditions (continuously vs. intermittently
inundated), presence/absence of emergent vegetation, spatial (edge vs. deeper water
zones) and temporal (diurnal and seasonal) variation on CH4 and CO2 fluxes.
1.4 Wetland vegetation and methane fluxes
Rates of aquatic and aboveground primary productivity in wetlands can be among the
highest of all ecosystems (Mitsch and Gosselink, 2000). Relationships among primary
productivity, plant species and methane fluxes in wetlands with a constant positive water
table have been described by numerous researchers (Chanton and Dacey, 1991; Whiting
and Chanton, 1993; Kim et al., 1998; Hirota et al., 2004). Numerous emergent plant
species convey methane from sediments to the atmosphere by pressure and temperature
driven convection (bulk flow), that balances the flow of oxygen downwards to the rhizosphere (Brix et al., 1992). Emergent species that do not exhibit ‘active ventilation’ of methane provide a conduit for sediment gases to diffuse through plant tissue,
5 bypassing the oxygenated sediment-water interface and enabling diffusion rates that are
approximately four orders of magnitude greater than diffusive gas movement through the
water column (Kutzbach et al., 2004; Lal and Shukla, 2004). Vegetation provides the
autochthonous organic matter needed for microbial activity in the form of root exudates,
and above and belowground detritus.
Emergent wetland vegetation can also inhibit emission of methane to the atmosphere.
An important adaptation of emergent wetland plants for survival in anoxic soils is
aerenchymatous vascular tissue, which delivers oxygen to the root zone enabling
continued respiration. Oxygen is necessary for aerobic methanotrophs to consume
methane, and thus oxygen released through aerenchyma enables methanotrophy to occur
in the saturated root zone of emergent plants. While methane dynamics have been studied
extensively in natural wetlands, methane emissions from created and restored wetlands,
or wetlands with a seasonally fluctuating water table, have not been frequently reported.
Chapters 2 and 3 investigate relationships between emergent vegetation and methane
fluxes in created wetlands.
1.5 Hydric soils and carbon dynamics
An important measure of wetland maturation is the development of hydric soils,
including redoximorphic features and organic matter content. Hydric soils are defined as
soils “that formed under conditions of saturation, flooding, or ponding long enough
during the growing season to develop anaerobic conditions in the upper part”(Mausbach
and Parker, 2001). As electron acceptors are sequentially reduced under anaerobic
6 conditions, thermodynamics become favorable for methanogenesis, and CO2 can be
reduced by H2 or acetate can be broken down to CH4 and CO2 (Conrad et al., 1999).
As a complement to the ecosystem-scale study on hydrology in relation to CH4 and
CO2 fluxes, a replicated mesocosm experiment examined the relationship between gas
fluxes, hydrology and stage of soil development (Chapter 4). Hydric soils removed from
an onsite created oxbow wetland were used as the substrate in ten mesocosm tubs, while
non-hydric soils from within the same floodplain were used in another ten mesocosms.
All mesocosms were planted with the same three species of emergent macrophyte, and
CH4 and CO2 fluxes were monitored intensively using non-steady-state chambers over one growing season after establishing hydrology and vegetation in the mesocosoms.
1.6. Literature Cited
Bridgham, S. C., J. P.. Megonigal, J. K. Keller, N. B. Bliss, and C. Trettin, 2006. The carbon balance of North American wetlands. Wetlands 26: 889-916.
Brix, H., B. K. Sorrell and P. T. Orr, 1992. Internal pressurization and convective gas flow in some emergent freshwater macrophytes. Limnology and Oceanography 37: 1420-1433.
Brooks, R. P., D. H. Wardrop, C. A. Cole, and D. A. Campbell, 2005. Are we purveyors of wetland homogeneity? A model of degradation and restoration to improve wetland mitigation performance. Ecological Engineering 24: 331-340.
Chanton, J. P. and J. W. H. Dacey, 1991. Effects of vegetation on methane flux, reservoirs, and carbon isotopic composition. Pp 29-63 In: T. D. Sharkey, E. A. Holland, and H. A. Mooney (Editors), Trace Gas Emissions by Plants. Academic Press, Inc. San Diego.
Conrad, R., A. Bollmann, H. Yao and R. Roy, 1999. Effect of water management on soil microbial communities and atmospheric trace gases. Microbial Biosystems: New Frontiers. Proceedings of the 8th International Symposium on Microbial Ecology. C. R. Bell, M. Brylinsky, and P. Johnson-Green, Editors. Atlantic Canada Society for Microbial Ecology, Halifax, Canada.
7 D’Angelo, E. M., and K. R. Reddy, 1999. Regulators of heterotrophic microbial potentials in wetland soils. Soil Biology and Biochemistry 31: 815-830.
Galat, D. L., L. H. Fredrickson, D. D. Humburg, K. J. Bataille, J. R. Bodie, J. Dohrenwend, G. T. Gelwicks, J. E. Havel, D. L. Helmers, J. B. Hooker, J. R. Jones, M. F. Knowlton, J. Kubisiak, J. Mazourek, A. C. McColpin, R. B. Renken, and R. D. Semlitsch, 1998. Flooding to restore connectivity of regulated, large-river wetlands: natural and controlled flooding as complementary processes along the lower Missouri River. Bioscience 48: 721-733.
Gilvear, D. J., and C. Bradley, 2000. Hydrological monitoring and surveillance for wetland conservation and management; a UK perspective. Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere 25: 571—588.
Hirota, M., Y. Tang, Q. Hu, S. Hirata, T. Kato, W. Mo, G. Cao, and S. Mariko, 2004. Methane emissions from different vegetation zones in a Qinghai-Tibetal Plateau wetland. Soil Biology & Biochemistry 36: 737-748.
IPCC, 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Houghton, J. T., Ding, Y., Griggs, D. J., Noguer, M., van der Linden, P. J., Dai, S., Maskell, K., Johnson, C. A. (Eds.), Cambridge University Press, Cambridge, UK and New York, NY, 881 pp.
Joabsson, A., and T. R. Christensen, 2001. Wetlands and methane emission. Pp. 1429- 1432 in: R. Lal, (Editor), Encyclopedia of Soil Science. Marcel Dekker, Inc. New York, USA.
Junk, W. J., P. B. Bayley and R. E. Sparks, 1989. The flood pulse comcept in river- floodplain systems. P. 110-127. In D. P. Dodge [ed.] Proceedings of the International Large River Symposium. Special Issue of the Canadian Journal pf Fisheries and Aquatic Sciences 106: 11-107.
Kim, J., S. B. Verma, and D. P. Billesbach, 1998. Seasonal variation in methane emission from a temperate Phragmites-dominated marsh: effect of growth stage and plant-mediated transport. Global Change Biology 5: 433-440.
Lal, R., and M. K. Shukla, 2004. Principles of Soil Physics. Marcel Dekker, Inc., New York, NY. 716 pp.
Megonigal, J.P., M.E. Hines, and P.T. Visscher, 2004. Anaerobic Metabolism: Linkages to Trace Gases and Aerobic Processes. Pages 317-424 in Schlesinger, W.H. (Editor). Biogeochemistry. Elsevier-Pergamon, Oxford, UK.
8 Kutzbach, L., D. Wagner, and E-M. Pfeiffer, 2004. Effect of microrelief and vegetation on methane emission from wet polygonal tundra, Lena Delta, Northern Siberia. Biogeochemistry 69: 341-362.
Mausbach, M. J., and W. B. Parker, 2001. Background and History of the Concept of Hydric Soils. Chapter 2 in: J. L. Richardson and M. J. Vepraskas (Editors), Wetland Soils: Genesis, Hydrology, Landscapes, and Classification. CRC Press, Boca Raton, FL.
McKee, K. L., and P. L. Faulkner, 2000. Restoration of biogeochemical function in mangrove forests. Restoration Ecology 8: 247-259.
Middleton, B., 1999. Wetland Restoration, Flood Pulsing, and Disturbance Dynamics. John Wiley and Sons, New York.
Mitra, S., R. Wassmann, and P. L. K. Vlek, 2005. An appraisal of global wetland area and its organic carbon stock. Current Science 88: 25-35.
Mitsch, W. J., X. Wu, R. W. Nairn, P. E. Weihe, M. Wang, R. Deal and C. E. Boucher, 1998. Creating and restoring wetlands: a whole-ecosystem experiment in self-design. Bioscience 48: 1019-1030.
Mitsch, W. J., and J. G. Gosselink, 2000. Wetlands, 3rd Edition. John Wiley & Sons, Inc., New York.
Mitsch, W. J., and S. E. Jorgensen, 2004. Ecological Engineering and Ecosystem Restoration. John Wiley & Sons, Inc., Hoboken, NJ. 411 pp.
Montalto, F. A., and T. S. Steenhuis, 2004. The link between hydrology and restoration of tidal marshes in the New York/New Jersey estuary. Wetlands 24: 414-425.
Moore, T. R., and J. Turunen, 2004. Carbon accumulation and storage in mineral subsoil beneath peat. Soil Science Society of America Journal 68: 690-696.
Odum, W. E., E. P Odum and H. T. Odum, 1995. Nature’s pulsing paradigm. Estuaries 18: 547-555.
Roulet, N. T., 2000. Peatlands, carbon storage, greenhouse gases, and the Kyoto Protocol: Prospects and significance for Canada. Wetlands 20: 605-615.
Schipper, L. A., Reddy, K. R., 1994. Methane production and emission from four reclaimed and pristine wetlands of southeastern United States. Soil Science Society of America Journal 58: 1270-1275.
9 Schlesinger, W. H., 1997. Biogeochemistry, An Analysis of Global Change. 2nd Ed. Academic Press, San Diego, CA. 588 pp.
Spink, A., R. E. Sparks, M. van Oorschot, and J. T. A. Verhoeven, 1998. Nutrient dynamics of large river floodplains. Regulated Rivers: Research and Management 14: 203-216.
Tanner, C. C., Adams, D. D., Downes, M. T., 1997. Methane emissions from constructed wetlands treating agricultural wastewaters. Journal of Environmental Quality 26: 1056-1062.
Tuittila, E. S., Komulainen, V. M., Vasander, H., Nykänen, H., Martikainen, P. J., Laine, K., 2000. Methane dynamics of a restored cut-away peatland. Global Change Biology 6: 569-581.
Walter, B. P., M. Heimann, R. D. Shannon and J. R. White, 1996. A process-based model to derive methane emissions from natural wetlands. Geophysical Research Letters 23: 3731-3734.
Whalen, S. C., 2005. Biogeochemistry of methane exchange between natural wetlands and the atmosphere. Environmental Engineering Science 22: 73-94.
Whiting, G. J. and J. P. Chanton, 1993. Primary production control of methane emission from wetlands. Nature 364: 794-795.
Whiting, G. J. and J. P. Chanton, 2001. Greenhouse carbon balance of wetlands: methane emission versus carbon sequestration. Tellus 53B: 521-528.
Zedler, J. B., and J. C. Callaway, 1999. Tracking wetland restoration: Do mitigation sites follow desired trajectories? Restoration Ecology 7: 69-73.
10
Figure 1.1. Pathways of methane and carbon dioxide uptake and release from wetlands. CH4 can be released from emergent vegetation by passive diffusion or by convective flow. CO2 is taken up by plants, and O2 diffuses to the root zone through aerenchyma, where it may be used by methanotrophs to oxidize methane to CO2. Plant roots release organic exudates that serve as substrates for methanogenesis, and methane not oxidized in the unsaturated sediment zone is released to the atmosphere by diffusion, or by ebullition when the partial pressure of methane in porewater is greater than its partial pressure in the sediments. (Figure from Joabsson and Christensen, 2001).
11
CHAPTER 2
METHANE FLUX FROM CREATED RIPARIAN MARSHES: RELATIONSHIP TO
INTERMITTENT VS. CONTINUOUS INUNDATION AND EMERGENT
MACROPHYTES
2.1 Abstract
Methane’s importance as a greenhouse gas warrants examination of the dynamics
controlling its emission from temperate zone wetlands created and restored for habitat
replacement and water quality improvement. In this one-year field study, hydrology
typical of floodplains in the Midwestern USA was simulated in two experimental riparian
marshes. Methane fluxes were measured from February-December using non-steady-
state chambers located in marsh zones with and without emergent vegetation in which soils were intermittently exposed and inundated, and in permanently inundated wetland areas. Annual methane fluxes from intermittently flooded zones were 30% of fluxes
-2 -1 from permanently inundated wetland areas, which emitted ~42 g CH4-C m yr .
Average growing season rates of methane flux from intermittently flooded zones with
-2 -1 and without macrophytes did not differ significantly (~3.5 mg CH4-C m h ), but both
-2 were significantly less than those from permanently inundated areas (~8 mg CH4-C m
12 h-1). We suggest that incorporation of seasonal floods followed by drier periods in
created riparian wetlands could minimize methane emission.
2.2 Introduction
Wetlands are the major natural source of the greenhouse gas methane (CH4) due to
high rates of methanogenesis enabled by the presence of anaerobic soils (Le Mer and
Roger, 2001). The majority of studies reporting on CH4 emissions from wetlands have
been conducted in natural ecosystems or heavily-managed rice paddies (Cao et al., 1998).
Few publications exist that document CH4 flux from created and restored wetlands
(Schipper and Reddy, 1994; Tanner et al., 1997; Tuittila et al., 2000). Because CH4 has a global warming potential 21 to 23 times greater than that of carbon dioxide over a 100- year time horizon (IPCC, 2001; Whalen, 2005), minimizing its emission from created and restored wetlands is a desirable goal.
Riparian areas (floodplains of rivers and streams) are subjected to seasonal flooding/drying cycles, when not obstructed by levees and other flood control structures.
Hydrologic dynamics in connected floodplains thus involve the lateral and vertical movement of water across land outside of river channels (Junk et al., 1989). It has been argued that the restoration of over 20,000 km2 of wetlands in former floodplains of the
Mississippi River Basin would help to alleviate hypoxia in the Gulf of Mexico by reducing nitrate-nitrogen pollution, reduce the intensity of extreme floods in the lower basin, and reestablish habitat and ecosystem functions that have been lost due to extensive regulation of rivers and streams, and drainage of wetlands (Mitsch et al., 2001;
2005a, Mitsch and Day, 2006). Given climatic perturbations projected to arise from
13 increasing atmospheric concentrations of greenhouse gases (IPCC, 2001), designing
created and restored wetlands to minimize CH4 emissions should be an important
objective (Whiting and Chanton, 2001).
A primary goal of riparian wetland restoration and creation is to re-inundate
disconnected floodplain areas, often by partial removal of river bank levees, or by
creating wetlands to intercept lateral runoff before it reaches the river. Experimental, whole-ecosystem research has been advocated as an essential means of assessing the effects of flood pulses on the structure and function of created and restored wetlands
(Bayley, 1995; Molles et al., 1998; Mitsch and Day, 2004). Primary productivity of riparian wetlands can be enhanced by seasonal flood pulses (Mitsch and Ewel, 1979;
Odum et al., 1995; Tockner and Stanford, 2002), in part because of the influx of nutrients that often accompanies a flood. However, methane emissions have been correlated with macrophyte productivity despite potential CH4 oxidation (methanotrophy) that can occur
in the rhizosphere because of release of oxygen from the roots of emergent macrophytes
(Whiting and Chanton, 1993; Bellisario et al., 1999; Van der Nat and Middelburg, 2000;
Jia et al., 2001; Hirota et al., 2004). Gas transport by macrophytes is enhanced via
pressurized ventilation that causes convection of gases from sediments to the atmosphere
in some vascular species (Brix et al., 1992) and diffusion through air-filled tissue
(aerenchyma) in plant roots and stems. In addition, plant root exudates and detritus
provide simple organic compounds that enhance establishment of sediment microbial
communities and can be used for methanogenesis under anaerobic conditions (Schütz et al., 1991). However, when the water table is below the soil surface, the positive effect of macrophytes on CH4 emission may be negated by CH4 oxidation (Waddington et al.,
14 1996). The overall relationship between macrophytes and CH4 flux remains poorly
understood in wetlands with a seasonally fluctuating water table.
The objectives of this study were to examine the effects of winter and spring floods
and periods of low flow, permanent inundation, and emergent macrophytes on methane
flux from multiple sites in two 1-ha, 10-year old created, riparian wetlands in the
Midwestern U.S. Hydrology in created and restored wetlands could then be managed to minimize methane production, if that management is compatible with other desired wetland functions. Hypotheses were that 1) greater methane emissions will occur in permanently inundated wetland zones than in intermittently flooded zones where the water table fluctuates above and below the soil surface; 2) more methane will be released from intermittently flooded zones containing emergent macrophytes than from adjacent areas in which macrophytes are removed; methane emissions will be greater during the day than at night from intermittently flooded areas with emergent macrophytes.
2.3 Materials and Methods
2.3.1 Study site
The experiment was conducted in two adjacent, 1-ha riparian marsh wetlands created
in 1994 at the Schiermeier Olentangy River Wetland Research Park (Figure 2.1) at The
Ohio State University, Columbus, Ohio (40°0′N, 83°1′E) USA. The two experimental
wetlands, designed with the same geomorphology and hydrology, are flow-through
systems that have received water that is pumped continuously from the adjacent, third-
order Olentangy River since 1994 (Mitsch et al., 1998; 2005b). Water level fluctuates in
the wetlands primarily according to the pumping rates, which have either been
15 proportional to fluctuations in river flow (1994-2002), or which have been chosen for
specific hydrologic experiments (2003-05). About 40% of each wetland consists of
deeper water areas that have been consistently inundated for ten years, except for
infrequent occasions when pumps fail or are shut off for maintenance or biomass harvest.
About 60% of each wetland consists of shallower areas that have experienced regular
cycles of inundation and soil exposure, and generally support growth of emergent macrophytes (Mitsch et al., 1998; Kang et al., 1998; Harter and Mitsch, 2003). During this experiment, the deepwater basins generally had water levels >+20 cm, while zones containing emergent vegetation varied from <-20 cm to >+20 cm. Previous studies conducted in these experimental wetlands include examinations of aquatic consumers
(Metzker and Mitsch, 1997), early algal growth patterns (Wu and Mitsch, 1998), hydrology (Koreny et al., 1999), water quality (Kang et al., 1998; Nairn and Mitsch,
2000; Spieles and Mitsch, 2000; Anderson et al., 2002), vegetation succession (Mitsch et al., 1998, 2005b; Selbo and Snow, 2004), sedimentation (Harter and Mitsch, 2003), soil
development (Anderson et al., 2005; Anderson and Mitsch, 2006), and N2O emissions
(Hernandez and Mitsch, 2006).
Sampling locations were chosen according to hydrology and basin
morphology/elevation, with the objective of encompassing the spatial heterogeneity of
the wetlands. Eight gas-sampling chambers were deployed in each of the two marshes,
consisting of four in intermittently inundated edge zones, two in “marsh” zones, and two
in deep-water areas (Figure 2.2). Chambers in the edge zones of the inflow and outflow
section of each wetland were positioned to capture the intermittent inundation and
exposure of soils associated with the flood-pulse. The potential distance into the
16 wetlands for placement of edge zone chambers was determined from correlations
between elevation, water level and pumping rate at 1-m2 intervals throughout each wetland. Chambers were placed close to permanent boardwalks to facilitate movement between sampling locations and to minimize disturbance to sediments during sampling.
Wooden pallets (non treated lumber) were placed beside chambers that could not be reached directly from the boardwalks. Edge zone chambers were located a minimum of
10 m from chambers being utilized in a separate study to investigate gaseous nitrogen fluxes (Hernandez and Mitsch, in press). A chamber was located in the upper and lower third of each marsh between the edge and deepwater zones, in areas that generally supported growth of emergent macrophytes in previous years (the “marsh zone”).
Floating chambers were deployed in randomly selected permanently inundated areas in the deepwater basins of each marsh. During the study period these deepwater basins were completely and continuously inundated, with the exception of a brief period in
August when pumps were turned off for an annual biomass survey. In order to examine the effect of emergent vegetation on methane flux, emergent macrophyte species were continually removed (approximately weekly) from half (n=4) of the chambers in the edge
zones, while submerged, floating and creeping plant species were left intact.
Macrophytes were gently pulled up from the roots, but rhizomes were left in place in
order not to cause excessive disturbance to the sediments.
During the period of this experiment, the lower marsh chamber in Wetland 1
intersected a flow path that resulted in permanently inundated conditions lacking emergent macrophytes. Also during 2004, the lower half of Wetland 2 was sparsely vegetated, resulting in permanently inundated, macrophyte-free conditions for the lower
17 marsh chamber. We included these chambers in the analysis of permanently inundated
areas, as their hydrology and vegetation were consistent with that of the deepwater
basins. The two upper marsh chambers each contained robust growth of emergent
macrophytes and experienced periods of inundation and drawdown, and therefore were
analyzed together with the four chambers in the edge zones that contained emergent
macrophytes.
2.3.2 Hydrology
Controlled flood pulses, first created during 2003 and continued through 2004, were
designed to simulate a hydroperiod typical of riparian wetlands receiving floodwaters in winter and spring, with a drier period during summer and fall (Mitsch et al., 2005b).
During the first week of each month from February through June, the wetlands received water at an average rate of 2.2 m3 min-1 (632 U.S. gallons min-1, the flood pulse), and
during the remainder of each month the flow rate averaged 0.50 m3 min-1 (128 gallons
min-1). Problems with the pumps in January resulted in the flood pulse being delivered in the middle of that month. From July through December (low-flow months), the flow rate averaged 0.74 m3 min-1 (194 U.S. gallons min-1, Figure 2.3).
2.3.3 Gas sampling
A non-steady-state chamber design (Klinger et al., 1994; Livingston and Hutchinson,
1995) was employed for gas sampling, with permanent chamber frames located in the
edge and marsh zones, and portable floating chambers deployed in deeper, permanently
inundated zones of each wetland. Six 0.27m2 circular HDPE bases (diameter = 59.5 cm)
18 were inserted permanently to 5-10 cm soil depth in the edge and marsh zones of each
wetland; permanent PVC frames to which plastic-coated wires were attached for holding
non-mercury thermometers were installed within the chamber bases. Floating chambers were constructed from PVC frames (50-cm tall and covering 0.27m2) encircled by foam
at the bottom for buoyancy, and fitted with 4-mil polyethylene bags to which
thermometers, 3 m Tygon vent tube (1.6 mm i.d.) and grey butyl sampling ports were
affixed. Fitted, 4-mil (0.1 mm) polyethylene bags with vent tubing and sampling ports
were pulled over the permanent chamber frames at the time of sampling, and sealed
around each chamber base with two 3-cm wide elastic straps. Bags were removed at the
end of each sampling period. Chambers in which emergent vegetation was removed
measured 50-cm tall, while chambers containing emergent vegetation measured 150-cm
tall. Chamber volume varied with the changing height of the water table.
Gas sampling was carried out during 2004 over one diurnal cycle in February, over
two diurnal cycles each month from March-October, and during the daytime only in
November and December. Sampling was conducted during flood events and periods of
low flow. Floating chambers were deployed beginning in July. Samples were taken in
the morning, afternoon and after dark on diurnal sampling dates. The starting point
(wetland and inflow/outflow) was chosen randomly on each date in order to avoid
systematic bias due to changing intensities of solar radiation and soil temperature over sampling periods, which required approximately three hours to complete. Three to five samples were taken from the headspace of each chamber over a period of up to one hour during each sampling period (morning, afternoon, nighttime). Samples were extracted through septa with a 20-ml syringe with a one-way stopcock, and injected into pre-
19 evacuated, 10-ml autosampler vials sealed with grey butyl septa. Septa were pre-boiled for 20 min to eliminate potential volatile outgassing. Vials were overfilled in order to minimize potential diffusion across the septa. Gas samples were stored at 4°C until analysis, and were analyzed within 7 days. Matheson gas standards were taken to the field and stored in the same manner as experimental samples to verify that no significant change in concentration occurred during storage. Chamber temperature was recorded when each sample was extracted. Water level and temperature, soil temperature at five and ten cm depths, and percent cover of vegetation by species were recorded for each plot during each sampling period. Percent cover of vegetation (total and that of individual species) was estimated visually. Regressions between chamber water levels and wetland staff gage readings enabled calculation of negative water table depths for sampling periods in which the soil surface was exposed in edge and marsh zone plots.
2.3.4 Analysis
Gas samples were analyzed on a Shimadzu GC-14A gas chromatograph equipped with a 40-position HT200H Autosampler, by flame ionization detection. A 1.8 m
Porapak Q column was used for sample separation, and helium (25 ml min-1) was the carrier gas (oven, injection and detector temperatures at 40°C, 40°C and 150°C respectively). Matheson methane standards, balanced with N2, were used to perform 4- point calibration curves; check-standards were injected at a rate of approximately one check standard per every 50 field samples. Results were returned in parts per million
-2 -1 (ppm), which were then converted to flux rates (mg CH4-C m h ), corrected for chamber volume and temperature (Healy et al., 1996). Regressions were performed on
20 each flux rate in Microsoft Excel™ to determine linearity of flux. CH4-C flux rates with correlations and R2 < 0.85 were considered to be zero when individual measurements
2 varied less than 1 ppm. If R < 0.85 and CH4 concentrations varied by more than 1 ppm over the sampling period, the flux rate was discarded. Therefore, only linear positive,
negative or zero flux rates were used in the analyses; 90% of linear flux rates had R2 >
0.90. Where removing a sample corrected a poor correlation to > 0.90, the sample was
eliminated from the calculation (Holland et al., 1999). This was reasonable given potential disturbances during attachment of the bag to the chamber frame at time zero, and to natural variability in flux rate, sometimes due to ebullition. In no case were fewer than three samples used for a concentration regression.
Flux rates were analyzed for the growing season (late April through September) and
non-growing season, according to plot type (intermittently flooded with or without
emergent macrophytes, permanently inundated), whether soil within plot types was
inundated or exposed during sampling, water depth, and for time of day (morning,
afternoon, nighttime). Data were not normally distributed. Log-transformation resulted in
normal distributions for growing season fluxes from permanently inundated chambers and chambers with emergent macrophytes, but not for chambers without emergent macrophytes. One-way analysis of variance (ANOVA) was used to compare growing- season fluxes from permanently inundated chambers vs. intermittently flooded chambers with emergent macrophytes; Mann-Whitney U tests were used to compare flux rates between all other combinations of chambers. Log-transformation produced normal distributions for intermittently flooded chambers with and without macrophytes when soil was inundated, but not when soil was exposed. Comparisons between chamber types
21 during inundated sampling periods was made with ANOVA, while Mann-Whitney U
tests were used on non-transformed data to compare inundated vs. exposed conditions.
For diurnal flux data, paired t-tests were performed on normally-distributed, log-
transformed flux rates for permanently inundated chambers; Wilcoxon Signed Ranks
tests were employed to examine diurnal trends within pulsed chambers. Analyses were
performed in SPSS 11.0 (SPSS, 2004), except for correlations between water table level
and methane flux, which were performed in Microsoft Excel™. Average and standard
errors of CH4-C fluxes for the growing season from each chamber type were multiplied
by 24 hours and the number of days (158) in the growing season to obtain an annual CH4 budget. Because only afternoon sampling was conducted in December, the methane flux rate from each chamber for each sampling day during the non-growing season was averaged, and the average of all fluxes was then multiplied by 24 h and the number of days (208) in the non-growing season. Standard errors were obtained in the same way.
In this way a bias in favor of diurnal sampling months was minimized.
2.4 Results
2.4.1 Hydrology and vegetation
Water levels in pulsed chambers varied from < –20 to +39 cm, and in permanently inundated chambers from +1 to +41 cm deep. During low-flow periods some of the chambers in intermittently flooded areas were inundated, and some had exposed soils
(Figure 2.4B), and certain marsh areas experienced more dramatic or frequent periods of drawdown due to small-scale variations in topography. The soil in pulsed zone chambers was inundated or saturated during 66% of sampling occasions over the year, and 61% of
22 sampling occasions during the growing season. Although the exaggerated flooding/low
flow regime took place during the winter and spring, the relatively low flow conditions
maintained during the growing season resulted in soils in the edge and upper marsh zones
being periodically exposed when plant transpiration and minor fluctuations in the
pumping rate facilitated lower water levels. Also, the pumps were shut off in early
August for the annual biomass harvest that takes place at the research site, causing an
exaggerated drawdown. An extra set of samples was taken during this drawdown
(August 10), but only edge and marsh zone samples were included in the analysis.
Plant species in permanently inundated areas included Potamogeton nodosus Poir., P. pectinatus L., Ceratophyllum demersum L., Elodea Canadensis Rich., and several algal species. Emergent macrophyte species in pulsed chambers included Schoenoplectus tabernaemontani K.C. Gmel., Leersia oryzoides L. Sw., Sparganium eurycarpum
Engelm., Typha latifolia L., T. angustafolia L., Sagittaria latifolia Willd., and Eleocharis palustris L. Pulsed chambers in which macrophytes were removed contained occasional
Lemna major Meyer, Ludwigia palustris L., and algal species.
2.4.2 Methane flux
Substantial methane fluxes were observed beginning in May, after the average soil temperature between 5-10 cm depths reached 15°C. The average soil temperatures
between 5-10 cm depth during the growing and non-growing seasons respectively were
22.38±0.14 and 11.13±0.46 ºC in intermittently flooded zones, and 23.80±0.44 and
14.64±0.92 ºC in permanently inundated areas. Although soil temperatures remained at
20°C through October, CH4 fluxes from intermittently flooded areas virtually ceased by 23 late September, while in permanently inundated areas methane continued to be emitted
through November, when soils were approximately 12.5°C (Figure 2.4A and C). While
methane fluxes from intermittently flooded wetland zones were negligible during the
non-growing season, fluxes from permanently inundated areas during the non-growing
season were comparable to growing season fluxes from intermittently flooded areas
(Figure 2.5). Soil temperatures, averaged over all sampling periods, were 20.69±0.34°C
in October and 12.80±0.11°C in November in permanently inundated areas -
approximately one degree higher than soil temperatures in pulsed zones during those
months. Soils were exposed in 38% and 29% of pulsed chambers with and without
emergent macrophytes, respectively, during October and November sampling periods
(Figure 2.4B). Flux rates for January were assumed to be zero, as rates for February and
December were not significantly different from zero for any wetland zone.
Average annual CH4-C flux from intermittently flooded zones was significantly less
(p < 0.001) than that from permanently inundated areas. Wetland zones containing
-2 -1 emergent macrophytes emitted 13.01±1.48 g CH4-C m yr (31% of the 42.00±17.11 g
-2 -1 CH4-C m yr emitted in permanently inundated areas, p=0.001), while areas without
-2 -1 macrophytes emitted 13.52±3.43 g CH4-C m yr (32% of that from permanently
inundated areas, p=0.001). Average growing-season rates of methane flux from
intermittently flooded plots with and without emergent vegetation, 3.40±0.37 mg CH4-C
-2 -1 -2 -1 m h and 3.54±0.89 mg CH4-C m h respectively, did not differ significantly (Figure
2.5). Methane fluxes were greatest when the water table was closest to the soil surface.
Fluxes decreased substantially when the water depth was greater than 25 - 30 cm, and in
24 intermittently flooded areas no methane efflux was observed when the water table was 20
cm below the surface (Figure 2.6).
During the growing season, intermittently flooded plots containing emergent
macrophytes emitted, on average, 70% less methane when soil was exposed compared to
-2 -1 when soil was inundated (1.32±0.26 vs. 4.70±0.54 mg CH4-C m h ), while those without emergent vegetation emitted 50% less methane when soil was exposed compared
-2 -1 to when inundated (2.10±0.92 vs. 3.93±1.10 mg CH4-C m h , Figure 2.7A). The
difference in methane emission during inundated and exposed conditions for pulsed plots
with macrophytes was significant (p=0.000) while in pulsed plots without emergent
macrophytes the difference was less substantial (p = 0.082). During periods of
inundation, methane emissions in pulsed areas with and without macrophytes were
significantly lower than emissions in permanently inundated areas (p=0.015 and p<0.001
respectively, Figure 2.7B).
Diurnal patterns in CH4 emission during the growing season were found only for permanently inundated areas, where afternoon flux rates were significantly higher than morning rates (p=0.004) and substantially higher than nighttime rates (p=0.06, Figure
2.8). The warmest soil temperatures occurred in the afternoon, averaging 3°C warmer
than morning, and 1.2°C warmer than nighttime temperatures.
2.5 Discussion
2.5.1 Hydrology and methane fluxes
Wetland areas that experienced a fluctuating water table with periodically exposed soils clearly demonstrated significantly lower efflux of methane compared to 25 permanently inundated areas. As long as soils remained saturated, methane emission
continued to be substantial, indicating the water level needed to fall below the surface of
the soil to reduce methane fluxes. The water table threshold required to completely
inhibit methane flux was a minimum of –20 cm, comparable to inhibition of methane flux
below –15 cm observed in Michigan peatlands (Shannon and White, 1994). Weak
relationships between water table and methane flux when the water table is slightly above
or below the soil surface have been found by other researchers (Dalva et al., 2001),
suggesting that hydrologic pulsing events should be reasonably dramatic in order to reduce methane emissions.
The fact that intermittently flooded wetland zones, when inundated, emitted
significantly less methane than permanently inundated areas indicates a difference in soil
conditions or microbial community structure and dynamics between these areas.
Methanogenic communities may be more sensitive to changes in sediment redox status
than methanotrophs. In a study of relationships between methane dynamics and
hydrology in a boreal bog, methane oxidation resumed less than a day after removal of
anoxic conditions, while methanogenesis was not observed even two days after a return
to anoxic conditions (Whalen and Reeburg, 2000). Periodic soil exposure may restrain
the methanogenic population and favor methanotrophic communities. Similar results
were found in a study of effects of contrasting hydrologic treatments on methane flux
from rice paddies in Louisiana, where periodic drawdowns of the water table reduced methane emissions by over 80% (Sass et al., 1992).
Concurrent studies of soil organic matter distribution in the experimental wetlands
(Anderson et al., 2005) suggest that differences in methane flux among wetland zones is
26 not due to contrasting concentrations of sediment organic carbon in pulsed and permanently inundated wetland zones. However, total accumulation of unconsolidated sediments is greater in permanently inundated areas than in edge zones and may result in greater availability of methanogenic substrates in continuously anoxic areas. In addition, the consistently low oxidation-reduction potential maintained under saturated conditions is a probable explanation for the greater methane production in permanently inundated areas; continuously low redox potential would enable development of a more robust methanogenic population and maintain alternate electron acceptors such as Fe3+ in a depleted state. At higher redox potentials, use of substrates by competing microbial populations decreases substrate availability for methanogens (Segers, 1998), another probable cause of reduced methane flux from intermittently flooded wetland zones.
Oxidation of labile C during periods of soil exposure could lessen available substrates as well.
2.5.2 Emergent macrophytes and methane flux
Methane flux is highest during midday where pressurized convection of the gas through plant tissue is occurring, due to heat and humidity-induced pressure differentials between plant tissues and the atmosphere (Schütz et al., 1991). The lack of diurnal variation in chambers containing emergent macrophytes supports an interpretation that pressurized convection of CH4 through plant tissue was not the dominant pathway for release of the gas, although this mechanism has been demonstrated for genera present in our study (Typha, Sparganium and Eleocharis) (Brix et al., 1992; Vretare Strand, 2002).
The greater average reduction in CH4 flux during drawdown in areas containing
27 macrophytes, compared to areas without emergent vegetation, may be attributed in part to
aeration of the soil due to transport of oxygen to the rhizosphere via aerenchyma.
However, methane flux was observed in areas with emergent macrophytes when the
water table was as low as –17 cm, while in areas without emergent macrophytes methane
flux was observed on only one occasion when the water table was <-15 cm. This
suggests that some methane may have been vented to the atmosphere from plant roots through macrophyte tissue during drawdowns. The variable of ultimate importance in this study was whether plots had been subjected to fluctuating hydrologic conditions, not
whether emergent macrophytes were present. These findings are not consistent with
frequently demonstrated correlations between primary productivity and methane emission
in wetlands (Whalen, 2005). During this study, soils in pulsed chambers remained
exposed for up to two weeks between flood pulses, and intermittently during the low-
flow period from July to December. Primary productivity may be strongly related to
CH4 flux in wetlands with constant positive water tables (Whiting and Chanton, 1993), but in our study of pulsed water levels, duration of flooding was much more important.
2.5.3 Comparison with other wetland methane studies
The flux rates measured in our 10-year-old created wetlands (3.4 – 7.7 mg m-2 hr-1
during the growing season) fall within the range reported for comparable natural wetlands
in which methane flux was assessed using non-steady-state chambers. Comprehensive studies of methane flux rates from floodplains and littoral wetlands across a variety of
-2 -1 climates report average growing season fluxes of 1.3 to 15 mg CH4-C m h in
-2 -1 permanently inundated areas, and -0.2 to 6 mg CH4-C m h in periodically-flooded
28 areas (Devol et al., 1988, Pulliam, 1993; Juutinen et al., 2001). Higher flux rates occur
during waterlogged conditions than during extended drawdowns across wetland
ecosystems (Harris and Sebacher, 1982; Smith et al., 2000). Annual flux rates of 35 g
-2 -1 CH4-C m y are reported for forested swamps and marshes at similar latitudes (Bartlett
-2 -1 and Harris, 1993). Our flux rates, 13 to 42 g CH4-C m y for intermittently and permanently-inundated areas respectively, fall within this range, indicating that after one
decade of development created riparian wetlands compare to natural ecosystems in this
function.
2.6 Conclusions
The frequency and duration of inundation/drawdown events, in combination with vegetation, may be key factors controlling the magnitude of methane flux from wetlands in general. The results of this study suggest that methane emissions from restored and created riparian wetlands can be minimized by the establishment of a pulsed hydroperiod in which the water level is allowed to periodically drop a minimum of 20 cm below the
soil surface during the growing season. Most often this means allowing the water level
to fluctuate naturally, rather than creating constantly flooded wetland pools. In order to
achieve a pulsing dynamic, edge zones—areas that absorb flood pulses and experience periodic drawdown—should be maximized. Such areas support growth of emergent vegetation, which, in combination with sustained periods of soil exposure, can be decoupled from any positive correlation with methane flux to the atmosphere. Although deeper areas that remain permanently inundated can emit substantially more methane
29 than edge zones, inclusion of a moderate proportion of such areas is often desirable for maintaining habitat or for enhancing the sediment retention capacity of the wetland.
2.7 Acknowledgements
Funding for this research was provided by USDA NRI CSREES Award 2003-35102-
13518, by an OARDC Payne Grant, and by the Olentangy River Wetland Research Park.
Olentangy River Wetland Research Park publication number 06-006. Christopher
Anderson, Dr. Virginie Bouchard, Dr. Li Zhang, Daniel Fink and two anonymous reviewers provided helpful critiques of the manuscript.
2.8 Literature Cited
Anderson, C. J., Mitsch, W.J., Nairn, R. J., 2005. Temporal and spatial development of surface soil conditions at two created riverine marshes. Journal of Environmental Quality 34: 2072-2081.
Anderson, C.J. and W.J. Mitsch. 2006. Sediment, carbon, and nutrient accumulation at two 10-year-old created riverine marshes. Wetlands 26: 779-792.
Anderson, K.L., Wheeler, K. A., Robinson, J. A., Tuovinen, O. H., 2002. Atrazine mineralization potential in two wetlands. Water Research 36: 4785—4794.
Bartlett, K. B., Harris, R. C., 1993. Review and assessment of methane emissions from wetlands. Chemosphere 26: 261—320.
Bayley, P. B., 1995. Understanding large river-floodplain ecosystems. Bioscience 45: 153—158.
Bellisario, L. M., Bubier, J. L., Moore, T. R., Chanton, J. P., 1999. Controls on CH4 emissions from a northern peatland. Global Biogeochemical Cycles 13: 81–91.
Brix, H., Sorrell, B. K., Orr, P. T., 1992. Internal pressurization and convective gas flow in some emergent freshwater macrophytes. Limnology and Oceanography 37: 1420—1433.
30 Cao, M., Gregson, K., Marshall, S., 1998. Global methane emission from wetlands and its sensitivity to climate change. Atmospheric Environment 32: 3293—3299.
Dalva, M., Moore, T. R., Arp, P., Clair, T. A., 2001. Methane and soil and plant community respiration from wetlands, Kejimkujik National Park, Nova Scotia: Measurements, predictions, and climatic change. Journal of Geophysical Research 106 (D3): 2955—2962.
Devol, A. H., Richey, J. E., Clark, W. A., King, S. L., 1988. Methane emissions to the troposphere from the Amazon floodplain. J. Geophys. Res. 93 (D2): 1583—1592.
Harris, R. C., Sebacher, D. I., 1982. Methane flux in the Great Dismal Swamp. Nature 297: 673—674.
Harter, S. K., Mitsch, W. J., 2003. Patterns of short-term sedimentation in a freshwater created marsh. Journal of Environmental Quality 32: 325 – 354.
Healy, R. W., Striegl, R. G., Russell, T. F., Hutchinson, G. L., Livingston, G. P., 1996. Numerical evaluation of static-chamber measurements of soil-atmosphere gas exchange: identification of physical processes Soil Science Society of America Journal 60: 740—747.
Hernandez, M. E. H., Mitsch, W. J., 2006. Influence of hydrologic pulses and vegetation on nitrous oxide emissions from created riparian marshes in Midwestern USA. Wetlands, in press.
Hirota, M., Tang, Y., Hu, Q., Hirata, S., Kato, T., Mo, W., Cao, G., and Mariko, S., 2004. Methane emissions from different vegetation zones in a Qinghai-Tibetal Plateau wetland. Soil Biology and Biochemistry 36: 737—748.
Holland, E. A., Robertson, G. P., Greenberg, J., Groffman, P. M., Boone, R. D., Gosz, J. R., 1999. Soil CO2, N2O, and CH4 Exchange. In Robertson, G. P., Coleman, D. C., Bledsoe, C. S. (Eds.), Standard Soil Methods for Long Term Ecological Research. Oxford University Press, New York, New York, pp. 185—201.
IPCC, 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Houghton, J. T., Ding, Y., Griggs, D. J., Noguer, M., van der Linden, P. J., Dai, S., Maskell, K., Johnson, C. A. (Eds.), Cambridge University Press, Cambridge, UK and New York, NY, 881 pp.
Jia, Z. J., Cai, Z. C., Xu, H., Li, X. P., 2001. Effect of rice plants on CH4 production, transport, oxidation and emission in ride paddy soil. Plant and Soil 230: 211–221.
31 Junk, W. J., Bayley, P. B., Sparks, R. E., 1989. The flood pulse concept in river- floodplain systems. In D. P. Dodge, (Ed.), Proceedings of the International Large River Symposium. Can. Spec. Publ. Fish. Aquat. Sci. 106: pp. 110—127.
Juutinen, S., Alm, J., Martikainen, P., 2001. Effects of spring flood and water level draw-down on methane dynamics in the littoral zone of boreal lakes. Freshwater Biology 46: 855—869.
Kang, H., Freeman, C., Lee, D., Mitsch, W. J., 1998. Enzyme activities in constructed wetlands: implication for water quality amelioration. Hydrobiologia 368: 231—235.
Klinger, L. F., Zimmerman, P. R., Greenberg, J. P., Heidt, L. E., Guenther, A. B., 1994. Carbon trace gas fluxes along a successional gradient in the Hudson Bay lowland. Journal of Geophysical Research 99 (D1): 1469—1494.
Koreny, J.S., Mitsch, W. J., Bair, E. S., Wu, X., 1999. Regional and local hydrology of a created riparian wetland system. Wetlands 19: 182—193.
Le Mer, J., Roger, P., 2001. Production, oxidation, emission and consumption of methane by soils: a review. European Journal of Soil Biology 37: 25—50.
Livingston, G. P., Hutchinson, G. L., 1995. Enclosure-based measurements of trace gas exchange: applications and sources of error. In P.A. Matson Harris, R. C., (Eds.), Biogenic Trace Gases: Measuring Emissions from Soil and Water. Blackwell Science, Oxford, pp. 14—51.
Metzker, K., Mitsch, W. J., 1997. Modelling self-design of the aquatic community in a newly created freshwater wetland. Ecological Modelling 100: 61—86.
Mitsch, W. J., Ewel, K. C., 1979. Comparative biomass and growth of cypress in Florida wetlands. American Midland Naturalist 101: 417—426.
Mitsch, W. J., Wu, X., Nairn, R. W., Weihe, P. E., Wang, M., Deal, R., Boucher, C. E., 1998. Creating and restoring wetlands: a whole-ecosystem experiment in self-design. Bioscience 48: 1019—1030.
Mitsch, W. J., Day, Jr., J. W., Gilliam, J. W., Groffman, P. M., Hey, D. L., Randall, G. W., Wang, N., 2001. Reducing nitrogen loading to the Gulf of Mexico from the Mississippi River basin: strategies to counter a persistent problem. Bioscience 51: 373—388.
Mitsch, W. J., Day, J. W., Jr. 2004. Thinking big with whole ecosystem studies and ecosystem restoration—a legacy of H.T. Odum. Ecological Modelling 178: 133— 155.
32 Mitsch, W. J., Day, Jr., J. W., Zhang, L., Lane, R. R., 2005a. Nitrate-nitrogen retention in wetlands in the Mississippi River Basin. Ecological Engineering 24: 267—278.
Mitsch, W.J., Zhang, L., Anderson, C. J., Altor, A., Hernandez, M., 2005b. Creating riverine wetlands: Ecological succession, nutrient retention, and pulsing effects. Ecological Engineering 25: 510-527.
Mitsch, W.J. and J.W. Day, Jr. 2006. Restoration of wetlands in the Mississippi-Ohio- Missouri (MOM) River Basin: Experience and needed research. Ecological Engineering 26: 55-69.
Molles, M. C. Jr., Crawford, C. S., Ellis, L. M., Valett, H. M., Dahm, C. N., 1998. Managed flooding for riparian ecosystem restoration: managed flooding reorganizes riparian forest ecosystems along the middle Rio Grande in New Mexico. Bioscience 48: 749—756.
Nairn, R.W., Mitsch, W. J., 2000. Phosphorus removal in created wetland ponds receiving river overflow. Ecological Engineering 14: 107—126.
Odum, W. E., Odum, E. P., Odum, H. T., 1995. Nature’s pulsing paradigm. Estuaries 18: 547—555.
Pulliam, W. M., 1993. Carbon dioxide and methane exports from a southeastern floodplain swamp. Ecological Monographs 63: 29—53.
Sass, R. L., Fisher, F. M., Wang, Y. B., 1992. Methane emission from rice fields: the effect of floodwater management. Global Biogeochemical Cycles 6: 249—262.
Schipper, L. A., Reddy, K. R., 1994. Methane production and emission from four reclaimed and pristine wetlands of southeastern United States. Soil Science Society of America Journal 58: 1270—1275.
Schütz, H., Schröder, P., Rennenberg, H., 1991. Role of plants in regulating the methane flux to the atmosphere. In Sharkey, T. D., Holland, E. A., and Mooney, H. A., (Eds.), Trace Gas Emissions by Plants. Academic Press, San Diego, California, pp. 29—63.
Segers, R.,1998. Methane production and methane consumption: a review of processes underlying wetland methane fluxes. Biogeochemistry 41: 23—51.
Shannon, R. D., White, J. R., 1994. A three-year study of controls on methane emissions from two Michigan peatlands. Biogeochemistry 27: 35—60.
Selbo, S.M., Snow, A. A., 2004. The potential for hydridization between Typha angustifolia and T. latafolia in a constructed wetland. Aquatic Botany 78: 361—369.
33 Smith, L. K., Lewis, W. M. Jr., Chanton, J. P., Cronin, G., and Hamilton, S. K., 2000. Methane emissions from the Orinoco River floodplain, Venezuela. Biogeochemistry 51: 113—140. Spieles, D.J., Mitsch, W. J., 2000. The effects of season and hydrologic and chemical loading on nitrate retention in constructed wetlands: A comparison of low and high nutrient riverine systems. Ecological Engineering 14: 77—91.
SPSS, 2004. Version 11.0.3. SPSS, Inc. Chicago, Illinois.
Tanner, C. C., Adams, D. D., Downes, M. T., 1997. Methane emissions from constructed wetlands treating agricultural wastewaters. Journal of Environmental Quality 26: 1056–1062.
Tockner, K., Stanford, J. A., 2002. Riverine flood plains: present state and future trends. Environmental Conservation 29: 308—330.
Tuittila, E. S., Komulainen, V. M., Vasander, H., Nykänen, H., Martikainen, P. J., Laine, K., 2000. Methane dynamics of a restored cut-away peatland. Global Change Biology 6: 569—581.
Van der Nat, F. J. W. A., Middelburg, J. J., 2000. Methane emissions from tidal freshwater marshes. Biogeochemistry 49: 103—121.
Vretare Strand, V., 2002. The influence of ventilation systems on water depth penetration of emergent macrophytes. Freshwater Biology 47: 1097—1005.
Waddington, J. M., Roulet, R. T., and Swanson, R. V., 1996. Water table control of CH4 emission enhancement by vascular plants in boreal peatlands. Journal of Geophysical Research 101 (D17): 22775—22785.
Whalen, S. C., 2005. Biogeochemistry of methane exchange between natural wetlands and the atmosphere. Environmental Engineering Science 22: 73—94.
Whalen, S. C., Reeburg, W. S., 2000. Methane oxidation, production, and emission at contrasting sites in a Boreal bog. Geomicrobiology Journal 17: 237—251.
Whiting, G. J. Chanton, J. P., 1993. Primary production control of methane emission from wetlands. Nature 364: 794—785.
Whiting, G. J., Chanton, J. P., 2001. Greenhouse carbon balance of wetlands: methane emission versus carbon sequestration. Tellus 53B: 521—528.
Wu, X., Mitsch, W. J., 1998. Spatial and temporal patterns of algae in newly constructed freshwater wetlands. Wetlands 18: 9—20.
34
Figure 2.1. The Schiermeier Olentangy River Wetland Research Park on the campus of the Ohio State University, Columbus, USA. Research was conducted in the kidney- shaped “experimental wetlands 1 and 2”. 35 Intermittently flooded with emergent macrophytes Intermittently flooded without emergent macrophytes Marsh zone plots Floating chambers were deployed within central, shaded areas
Figure 2.2. Locations of gas-sampling chambers in the two experimental wetlands. Straight lines represent permanent boardwalks.
36
3.0 ) -1 2.5 min 3 2.0
1.5
1.0
0.5 average weekly flow (m 0.0 r ar p ov ec Jan A Oct Feb M May June July Aug Sept N D
Figure 2.3. Inflow to each experimental wetland in 2004. Each bar represents the average pumping rate for a 1-week period.
37
30 A. 25
20
15 Degrees C 10
5
0
100 90 B. 80 70 60 50 40 30 20 10 Percent of intermittently flooded plots Percent of intermittently flooded plots period sampling during each inundated 0
45 C. 40 35 -1
h 30
-2 25 m
-C 20 4 15
mg CH 10 5 0 JFMAMJJASOND Intermittently flooded with emergent macrophytes Intermittently flooded without emergent macrophytes Permanently Flooded
Figure 2.4. A. Average (with standard error) soil temperature between 5-10 cm depths for each wetland zone on each sampling date; B. Percentage of intermittently flooded plots that were inundated on each sampling date; and C. Average (with standard error) CH4-C flux rate for each wetland zone on each sampling date.
38
60 a ) -1 50 y 2
-C m 40 4 a
g CH 30
20 a a a a b
10
Methane flux ( b b 0 Intermittently flooded Intermittently flooded Permanently- with emergent without emergent inundated macrophytes macrophytes Growing season Non-growing season Annual total
Figure 2.5. Seasonal and annual methane flux for each wetland zone. Different letters indicate a significant difference in methane emissions between seasons or wetland zones. Bars represent standard error. Sample sizes are as follows: intermittently flooded with emergent macrophytes—growing season n=185, non-growing season n=87; intermittently flooded without emergent macrophytes —growing season n=115, non- growing season n=57; permanently inundated—growing season n=111, non-growing season n=38.
39 45 A. y = -0.011x2 + 0.2735x + 4.8196 40 R2 = 0.031 -1 35 h
-2 30 25 -C m 4 20 15
mg CH 10 5 0 0204060
B. 25 y = -0.0052x2 + 0.1736x + 3.7133 R2 = 0.1336 20 -1 h -2 15
c -C m 4 10 mg CH 5
0 -40 -20 0 20 40 60
25 C. y = -0.0073x2 + 0.1838x + 4.8082 R2 = 0.1358 20
15
c 10
5
0 -40 -20 0 20 40 60 level of water table (cm) in relation to soil surface
Figure 2.6. Correlation between methane flux and the level of the water table above (positive numbers) and below (negative numbers) the soil surface for A: permanently inundated wetland areas, B: intermittently flooded wetland zones containing emergent macrophytes, and C: intermittently flooded wetland zones in which emergent macrophytes were removed.
40 ) 10 -1 h 9 A.
-2 Inundated average 8 Exposed average -C m
4 7 6 a ac 5 4 bc
flux (mg CH 3 4 b 2 1
Avg. CH 0 Intermittently flooded Intermittently flooded with emergent without emergent macrophytes macrophytes
10 B. b 9 8 7 6 5 a a 4 3 2 1 0 Intermittently Intermittently Permanently flooded with flooded without inundated emergent emergent macrophytes macrophytes
Figure 2.7. Average methane flux rates when soil in pulsed wetland zones was inundated vs. exposed during sampling (A), and overall rates for all wetland zones (B) during the 2004 growing season. Different letters within a panel indicate a significant difference in average methane flux rates between conditions. Bars indicate standard error. The number of samples is as follows: intermittently flooded with emergent macrophytes — inundated n=114, exposed n=71; intermittently flooded without emergent macrophytes— inundated n=70, exposed n=45; permanently inundated n=111.
41
18 Morning b 16 Afternoon 14
-1 Night h
2 12 10 -C m 4 8 a aa 6 a aa a a
mg CH 4 2 0 Intermittently Intermittently Permanently- flooded with flooded without inundated emergent emergent macrophytes macrophytes
Figure 2.8. Diurnal patterns in methane flux for each wetland zone. Different letters indicate a significant difference in methane emissions between time of day or wetland zone. Bars represent standard error. Gas samples were taken within the indicated time intervals. The number of samples is as follows: intermittently flooded with emergent macrophytes —7-11 a.m. n=59, 1-5 p.m. n=61, 8 p.m.-12 a.m. n=65; intermittently flooded without emergent macrophytes—7-11 a.m. n=36, 1-5 p.m. n=37, 8 p.m.-12 a.m. n=42; permanently-inundated—7-11 a.m. n=35, 1-5 p.m. n=38, 8 p.m.-12 a.m. n=38.
42
CHAPTER 3
PULSING HYDROLOGY, AND METHANE AND CARBON DIOXIDE FLUXES IN
CREATED MARSHES: A 2-YEAR ECOSYSTEM STUDY
3.1 Abstract
Short-term methane and carbon dioxide flux rates were measured in two created,
experimental marshes in the Midwestern U.S. over a two-year period (2004-2005) in
which hydrologic conditions were manipulated to simulate flood-pulse and steady-flow
conditions. Gas flux rates were measured in three distinct wetland zones: continuously inundated areas; edge zones with emergent macrophytes; and edge zones in which emergent macrophytes were removed. Methane fluxes between years were not significantly different in edge zones with and without emergent vegetation, but were twice as high in continuously inundated zones during the steady-flow year compared to the flood pulsed year. There was no apparent relationship between emergent vegetation and methane flux, as mean flux rates were not significantly different in either year in edge zones where emergent vegetation was removed, compared to edge zones containing emergent vegetation. Continuously inundated wetland zones emitted methane from
43 summer through fall, while in edge zones methane fluxes were only substantial in spring
and summer. Neither daytime rates of carbon dioxide uptake or nighttime rates of
respiration were significantly different between the years for any wetland zone. When
CO2 flux rates (daytime uptake plus nighttime respiration) were normalized for solar
radiation and day length, solar efficiency was found to be comparable between the steady
flow and pulsed years. Methane fluxes were more strongly affected than carbon dioxide fluxes by the differences in hydrology, but only in the deeper areas of the wetlands.
3.2 Introduction
Hydrology is a critical variable in the biogeochemical functioning of wetlands, and
thus hydrologic dynamics must be carefully planned in created and restored wetlands.
Because wetlands play a major role in the global carbon cycle, it is important to
understand how the design of these ecosystems impacts their storage of carbon and
emissions of greenhouse gases such as nitrous oxide and methane (Whiting and Chanton,
2001). Many natural wetlands experience dynamic hydrologic conditions that vary with
the seasons, including fluctuations in water depth above and below the soil, expanding
and contracting wetland area, or alternating cycles of wetting and drying (Gilvear and
Bradley, 2000). Hydrology, water quality, climate, substrate and biota form the context
for carbon cycling in wetland ecosystems (Christensen et al., 2003; Megonigal et al.,
2004). Under highly reduced conditions, methane (CH4) is produced in wetland
sediments; methane has a global warming potential 21 to 23 times greater than that of
carbon dioxide (IPCC, 2001; Whalen, 2005). Methane can be oxidized to CO2 and H2O by methanotrophic bacteria in the soil, rhizosphere or water column if oxygen is
44 available. Far from simply facilitating greenhouse gas production, reduced conditions in
wetland soils create an important sink for carbon, as complex and recalcitrant organic
matter such as humic acids accumulate under anaerobic conditions (Moore and Turunen,
2004; Mitra et al., 2005; Hernandez and Mitsch, 2007a). This carbon sequestration can
outweigh the production of methane in wetlands even with the greenhouse multiplier of
21-23, particularly in ecosystems such as created wetlands that are in an early stage of
development (Mitsch and Gosselink, 2000).
Spatial and temporal variability in biogeochemical processes is often high in natural
wetlands, and is a function of microtopography, soil physiochemical properties,
vegetation and microbial populations, disturbance, hydrology and availability of nutrients
(Hunt et al., 1997; Vaithiyanathan and Richardson, 1997; Fennessy and Mitsch, 2001;
Owen Coning, 2005). A frequently heard criticism of created and restored wetlands is
that they are not as dynamic in terms of physical and biotic properties as the natural
wetlands they are intended to replace (Zedler and Callaway, 1999; Brooks et al., 2005).
Engineered wetlands often resemble open-water ponds, with a relatively static
hydroperiod. Establishment of ecosystem-appropriate hydrology is cited as a critical
factor in successful wetland restoration projects (McKee and Faulkner, 2000; Montalto
and Steenhuis, 2004).
Because hydrology is a master variable mediating many of the biogeochemical processes in wetlands, a dramatic contrast in hydrological conditions from one year to another in a wetland should result in measurable differences in processes such as trace gas fluxes. Prolonged inundation leads to strongly reduced conditions in sediments, facilitating methane production when appropriate substrates are available. The scarcity
45 of available oxygen in strongly reduced soils limits oxidation of methane by obligately
aerobic methanotrophic bacteria. A fluctuating water table such as occurs in a
hydrologically-dynamic or pulsed system can stimulate complex and interacting
biogeochemical processes. These processes depend on the timing, duration and
magnitude of flood pulses and dry periods, physiology of micro- and macroscopic
organisms, and the stage of ecosystem development (Mitsch and Rust, 1984; Megonigal
et al., 1997). Periodic drawdowns of the water table can result in a reduction of methane
emissions to the atmosphere, due to methanotrophy, inhibition of methanogenesis by
oxygen or oxidation of the simple carbon compounds that serve as methanogenic
substrates (LeMer and Roger, 2001; Freeman et al., 2002). If wetland soils are exposed
during the spring, seeds can germinate and recharge wetland vegetation. Floods can
provide inputs of nutrients, biota and sediments to wetlands, thus acting as a subsidy (or a
stress if invasive biota such as carp are introduced). Floods of large magnitude can
uproot or bury vegetation, thereby decreasing productivity.
Methane fluxes can be affected by wetland vegetation in a variety of ways. Labile
carbon released as exudates or detrital matter from roots and other plant parts provides a substrate for methanogenesis, while oxygen released from root aerenchyma facilitates
methanotrophy (Whalen, 2005). Emergent vegetation has been shown to act as a conduit
for methane, when pressurized ventilation channels gases from sediments to the atmosphere through aerenchyma, in wetlands with standing water (Brix et al., 1996;
Yavitt and Knapp, 1998; Vretare Strand, 2002). Although pressurized ventilation has been well documented for a number of emergent wetland plant species as well as some aquatic plants in wetlands that are continuously inundated, the relationship between
46 vegetation and methane flux in wetlands with a fluctuating water table is less
straightforward (Waddington et al., 1996; Altor and Mitsch, 2006). Diurnal differences
in methane flux rates are often apparent when both pressurized convection and diffusive
transport through plant tissue are occurring. In plant species that demonstrate convective
throughflow, the highest rates of gas flux often occur at mid-morning when increasing
solar radiation activates pressurized ventilation, allowing release of gases stored in plant lacunae (Whiting and Chanton, 1996; Käki et al., 2001). Rates of diffusive gas flow
depend upon soil and air temperature as well as concentration gradients of gas between
the soil, plant and atmosphere. Warmer temperatures, as often occur during midday,
correspond to higher rates of diffusion (Whiting and Chanton, 1996). However,
maximum rates of convective gas flow have also been shown to occur during the
afternoon in some species due to temperature and pressure differentials between plant
tissue and the atmosphere (Brix et al., 1992).
We reported earlier on within-year spatial and seasonal fluxes of methane from two
experimental riparian marshes in the Midwestern USA, according to hydrologic
characteristics and the presence or absence of emergent vegetation (Altor and Mitsch,
2006). During that one-year study (2004), flood pulses were delivered to the wetlands
with the intention of simulating conditions that could be found in a natural riparian marsh
flooded with river water. In this study, gas flux dynamics under flood-pulsed conditions
are evaluated in comparison to gas fluxes from the same wetlands during a year of
steady-flow hydrologic conditions. During 2005, constant, low-flow (“steady-flow”)
conditions were maintained in the experimental wetlands to mimic a less dynamic
hydrologic regime. Methane fluxes were measured using non-steady-state chambers in
47 both years, in wetland areas designated as edge zones with and without emergent
vegetation, and in continuously-inundated zones. The objective was to compare fluxes of
CH4 under these contrasting hydrologic conditions, including an examination of the relationship between emergent vegetation and methane fluxes. A second focus was to
monitor uptake and efflux of carbon dioxide in each wetland zone over the two growing
seasons. The hypotheses for this investigation were that:
1. During the steady-flow year, methane fluxes from edge zones with emergent
vegetation will be higher than fluxes from similar areas without emergent
vegetation. This was expected because ventilation of methane through emergent
macrophytes has been demonstrated for wetlands with continuously standing
water, but not for wetlands with a fluctuating (alternating positive and negative)
water table. In addition, diffusion of methane through vascular plant stems
bypasses potential oxidation at the sediment/water interface and in the water
column.
2. Mean methane flux will be higher from edge topographical/spatial zones of the
wetlands during the steady flow year, compared to the pulsed year, while fluxes
from continuously inundated zones will not be significantly different between the
years. Higher fluxes from edge zones were expected during the steady-flow year
because these wetland zones remained inundated for the entire growing season
during the steady-flow year, in contrast to the flood-pulse year when edge zone
soils were intermittently inundated and exposed.
48 3. Net and gross rates of carbon dioxide uptake in vegetated zones of the wetlands
will be higher during the pulsing year than the steady-flow year, and percent solar
efficiency will be higher in the pulsing year. Greater productivity was expected in
the pulsing year because of the potential subsidy effect of nutrient and sediment
delivery via flood pulses, as well as the occurrence of exposed soils in spring of
the pulsing year to facilitate seed germination.
3.3 Methods
3.3.1 Study site
This research took place in two experimental wetlands at the Wilma H. Schiermeier
Olentangy River Wetland Research Park (ORWRP, Figure 3.1), on the campus of The
Ohio State University, Columbus, OH, USA (40°0′N and 83°1′E). Water depths fluctuate
in the wetlands according to precipitation, pumping rate and topography. The middle
basins of each wetland are deeper water areas that are “continuously inundated”, even at
low pumping rates, whereas the “edge zones” fluctuate between inundated and drained
conditions, depending on the pumping rate and season. Generally at inflows of less than
9 cm day-1, the edge zones will become dry during the growing season. Over the 12
years that the experimental wetlands have existed (1994–2005), the deep zones have
rarely been drained, with the exception of a 2-day annual biomass sampling each August,
and occasional long-term pump failures. Plankton, metaphyton, and submerged and
floating vegetation grow in the deep basins each year, while emergent macrophytes grow
in the shallower “edge zones” of the wetlands.
49 3.3.2 Hydrology
Pump-controlled floods were delivered to the marshes from February through June
2004, with low-flow conditions maintained during summer and fall 2004 (Figure 3.2).
This ‘flood pulse’ hydrology was designed to simulate a hydroperiod typical of riparian
wetlands receiving floodwaters in winter and spring (Mitsch et al., 2005). During the
3 first week of flood pulse months, the wetlands received water at an average rate of 2.8 m
-1 3 min (750 gpm), and during the remainder of each month the flow rate averaged 0.57 m
-1 min (150 gpm). Problems with the pumps in January resulted in the flood pulse being
delivered in the middle of that month. From July through December (low-flow months),
3 -1 the flow rate averaged 0.76 m min (200 gpm). In 2005, the flow rate was maintained
3 -1 at approximately 0.76 m min (200 gpm) each month. To reduce experimental
variability, the hydrologic input into the wetlands in 2005 was designed to be similar to
that in 2004. In total, approximately 43.5 m yr-1 of inflow were pumped into each
wetland during the flood-pulse year (2004) and approximately 39.1 m yr-1 of inflow were pumped into the wetlands during the steady-flow year (2005). Companion studies of other effects of this hydrologic pulsing were undertaken for denitrification (Hernandez and Mitsch, 2007b) and aquatic productivity (Tuttle and Mitsch, in revision).
3.3.3 Gas sampling
Fluxes of CH4 and CO2 were measured at eight locations in each wetland, using non-
steady-state chambers (Figure 3.3). PVC chamber frames and circular, HDPE bases were
left in place and were located close to permanent boardwalks to facilitate movement between sampling locations and to minimize disturbance to sediments during sampling
50 (Figure 3.4a). Wooden pallets (non treated lumber) were placed beside chambers that could not be reached directly from the boardwalks. Edge zone chambers were located a minimum of 10 m from chambers being utilized in a separate study to investigate gaseous nitrogen fluxes (Hernandez and Mitsch, 2006, 2007b). A chamber was located in the upper and lower third of each marsh between the edge and deepwater zones, in areas that had supported growth of emergent macrophytes in previous years (the “marsh zone”).
Floating chambers were deployed in randomly selected permanently inundated areas in the deepwater basins of each marsh. Chamber bases were 0.27 m2 and heights were as follows: marsh zones and edges with emergent macrophtyes, 150 cm; floating chambers and edge zones without emergent macrophytes, 50 cm. As macrophyte shoots emerged in edge zone chambers designed to be free of emergent vegetation, they were gently pulled up from the roots, but rhizomes were left in place to minimize excessive disturbance to the sediments. Submerged, floating and creeping vegetation was not removed from these chambers.
During 2004, gas sampling was conducted on the following dates: February 28,
March 4, March 24, April 8, April 25, May 9, May 23, June 10, June 29, July 7, August
4, August 10, August 25, September 4, September 28, October 26, November 23 and
December 16. Samples were taken in the morning (8:00 – 11:30), afternoon (1:30 –
5:30), and after dark on all dates except Feb. 28 and March 3 (morning and afternoon only), August 10 and September 28 (afternoon and nighttime only), and Dec. 16
(morning only). During 2005, gas sampling was conducted on February 22, March 21,
April 4, April 25, May 9, May 25, June 9, June 27, July 25, August 17, September 29, and October 29. Samples were taken in the morning, afternoon and after dark as in the
51 flood pulsed year, on all dates except for February 22 (afternoon only), March 21
(morning and afternoon only), April 25 (morning and nighttime only) and September 29
(afternoon and nighttime only). Because samples were taken on approximately the same
date in late September 2004 and 2005, only the samples from 28 September 2004 were
included in the between-year comparison. Results from the three sampling dates in
August 2004 were averaged for each chamber in order not to weight 2004 more heavily
than 2005 for August.
During sampling, transparent 4-mil (0.1 mm thick) polyethylene bags were pulled
down over the chamber frames, and attached tightly to the bases with elastic straps
(Figure 3.4b). The top of each bag was embedded with a grey butyl rubber sampling
septa and 3m Tygon vent tube (1.6 mm i.d.) for equilibrating the chamber with
atmospheric pressure. Using a 30ml syringe with 1-way stopcock, five samples were withdrawn from each chamber over approximately 60 minutes in 2004, and over approximately 25 minutes in 2005. Samples were injected into pre-evacuated 10-ml autosampler vials, overpressurizing the vials. Chamber air temperature was recorded when each sample was taken, and soil temperature and water depth were recorded for each chamber, along with percent cover of all plant species present in the chamber by visual estimation.
Samples were stored at 4°C until analysis by gas chromatography, always within one
week. Methane and carbon dioxide concentrations in each sample were quantified with a
single 2-ml injection on a Shimadzu GC 14A with HT200H Autosampler, using a thermal
conductivity detector and flame ionization detector in series. A 1.8 m Porapak Q column
was used for sample separation, and helium (25 ml min-1) was the carrier gas. Oven,
52 injection, TCD and FID temperatures were 40°C, 40°C, 200°C and 150°C respectively.
Four concentrations (5, 10, 15 and 20 ppm) were used to create a calibration curve for
methane, using Matheson Tri-Gas standards balanced with pure nitrogen. Four
concentrations (250, 500, 750 and 1000 ppm) were used to create a calibration curve for
CO2, using standards balanced with helium. GC results were returned in ppm, which were converted to concentration by mass (corrected for chamber surface area and volume) using the ideal gas law, after Healy et al. (1996) and Holland et al., (1999). Flux
-2 -1 rates (mg CH4-C and mg CO2-C m h ) were then determined from regressions between
the concentration of each sample and the time each sample was taken, using Microsoft
Excel. Linearity of each flux rate was examined and non-linear flux rates (R2 < 0.90)
were discarded, unless removing a sample corrected the regression. Flux rates were
determined from a minimum of three samples.
The average sampling time per chamber in 2004 was approximately one hour, while
the average sampling time per chamber in 2005 was approximately 25 minutes. This
difference occurred because in 2004 it took experimentation to determine the most
efficient approach to sample all of the chambers in the least amount of time. To assess
differences in gas flux results according to different sampling times, time trials were
conducted on May 9 and July 25, 2005. CO2 and CH4 fluxes were measured over one
hour by taking samples every 3 – 10 minutes, and regressions were performed between
flux rates within chambers at approximately 25 minutes vs. flux rates at approximately 60
minutes. Relationships between sampling time and gas flux rate were linear (R2 = 0.94).
Shorter vs. longer sampling times did not make a significant difference for methane fluxes, but did for CO2 fluxes. CO2-C flux rates measured over 15 – 30 minutes were
53 approximately twice as high as those measured over 30 – 60 minutes (Figure 3.5) and
2004 CO2 flux data were normalized using this relationship.
Solar radiation (kcal m-2 d-1) was measured on each sampling day using a LI-COR LI-
200SA pyranometer sensor located between the two experimental wetlands. Solar
radiation data were used to determine whether measured differences in CO2 uptake
between years could be due to differences in solar intensity and solar efficiency. The pyranometer was not operational until mid November, 2004. Solar radiation data (kcal
m-2 d-1) was acquired from the nearby town of Delaware, OH recorded by the Ohio
Agricultural Research and Development Center (OARDC) using a Campbell Scientific
datalogger. A regression was performed with these measurements and daily data from
the ORWRP for Nov.15, 2004 through May 2, 2005 to obtain a relationship from which
2004 ORWRP solar data could be calculated. The calculated relationship was strong (R2
= 0.96). In order to verify that the calculated (2004) data were accurate, the same regression equation was used on OARDC data for 2005 sampling dates, and compared with measured ORWRP solar radiation values for the same dates. Solar efficiency was determined for edge zones with emergent macrophytes and for continuously inundated zones as follows:
- mean hourly GPP * # of daylight hours on given sampling date = g C m-2 d-1
- g C m-2 d-1 * 10 = kcal m-2 d-1 (Mitsch and Gosselink, 2000)
- [kcal m-2 d-1 / kcal m-2 d-1 solar radiation] * 100 = solar efficiency (%) (Tuttle and
Mitsch, in revision)
54
3.3.4 Analysis of data
Fluxes of CH4 and CO2 were organized according to the three sampling zones: edge
zones with emergent vegetation (which included the inflow ‘marsh’ chambers), edge
zones without emergent vegetation, and continuously inundated zones (which included
outflow ‘marsh’ chambers). This plot organization was chosen in 2004, when inflow
marsh chambers experienced water level fluctuations similar to the edge zones and also
contained emergent vegetation. Also in 2004, outflow marsh chambers were
continuously inundated and did not contain emergent vegetation. Gas fluxes were
analyzed by sampling zone, according to time of day (morning v. afternoon v. night), and
year (2004 pulsed conditions v. 2005 steady-flow conditions). Methane fluxes were
analyzed for each of the four seasons, and CO2 fluxes were analyzed for the growing
season (April 21 – Sept 21) and non-growing season (late Sept – early April). Methane
flux rates were not normally distributed, and the signed square-root transformation was applied to normalize the distribution of the residuals. CO2 flux rates were normally
distributed. Methane and CO2 fluxes were analyzed with the General Linear Model
univariate analysis, and one way Analysis of Variance with Tukey’s post-hoc test.
Interaction effects between fixed factors (wetland zone, season, time of day, year) and covariates (soil temperature, water depth) were explored. Day and nighttime CO2 flux rates were analyzed with separate models, using only negative fluxes (CO2 uptake) for
the daytime analysis. All analyses were performed in SPSS 10 for Macintosh (SPSS,
2004).
55 3.4 Results
3.4.1 Hydrology
In 2004, the average inflow to Wetlands 1 and 2 was 12 cm day-1, while in 2005 the
average daily inflow into each wetland was 11 cm day-1. The pattern of inflow was
dramatically different between 2004 and 2005, with flood pulses and low flows in 2004,
and a moderate and consistent flow regime in 2005 (Figure 3.2). During 2004 pulsing,
the deepwater basins were inundated to a depth of +7.5 to +30 cm across the growing
season, while the edge zone water depth with and without emergent macrophytes varied
between –20 to +28 cm and –20 to +27 cm respectively. In 2005 steady-flow conditions,
water depth varied between +14 and +18 cm in continuously inundated areas, and from
+2 to +6 cm and +4 to +8 cm in edge zones with and without emergent macrophytes
respectively. In 2004, during low flows, the soil was drained as far as –20 cm below the
soil surface for up to a week at a time in edge zones.
3.4.2 Vegetation
Plant species observed in the sampling areas were similar between the two years
(Figure 3.6). Species found in edge zones with emergent vegetation included
Schoenoplectus tabernaemontani K.C. Gmel., Typha latifolia L., T. angustifolia L.,
Sparganium eurycarpum Engelm, Leersia oryzoides L. Sw., Eleocharis palustris L.,
Saggitaria latifolia Willd, and Alisma plantago-aquatica L. (also known as Alisma subcordatum). Of these species, L. oryzoides and S. tabernaemontani were dominant.
Edge zones in which emergent vegetation was removed contained Lemna minor L.,
Spirodela polyrhiza (L.) Schleiden (formerly known as Lemna major), Ludwigia palustris
56 L., and Algae spp. In edge zones where emergent vegetation was removed, Lemna spp.,
algae spp. and L. palustris were dominant. Plant species in continuously inundated
wetland zones included several benthic and planktonic algal species, Potamogeton natans
L., P. crispus L., P. pectinatus L., Lemna minor L., and S. polyrhiza. In continuously inundated zones, Lemna, Potamogeton and several algal species were dominant. The dominant species within each wetland zone were the same in 2004 and 2005, although the proportions changed slightly (Figure 3.6).
3.4.3 Methane fluxes
Within and between years, there were significant differences in methane flux
according to wetland zone and season (F = 15.898, 951, 23 p < 0.001). In the pulsing year,
significantly more methane was emitted during summer than the other seasons in edge
zones with and without emergent macrophytes (6.61 ± 0.78 and 7.40 ± 2.78 mg CH4-C
-2 -1 -2 -1 m h respectively during summer vs. < 1.5 mg CH4-C m h during the other seasons).
Similar rates of methane flux were measured in spring, summer and fall of the pulsing
year in continuously inundated zones (5.58 ± 3.04, 7.14 ± 1.21, and 5.59 ± 1.16 mg CH4-
C m-2 h-1 in each season respectively) (Figure 3.7). Spring and summertime rates of methane flux were similar between all wetland zones in the pulsing year. The range of methane flux rates from edge zones without emergent macrophytes was greater in the
-2 -1 pulsing year (-1.44 to 71.16 mg CH4-C m h ) than in the steady flow year to (-0.56 to
-2 -1 43.18 mg CH4-C m h ). The range of methane flux rates from edge zones with
emergent macrophytes and continuously inundated wetland zones was higher during the
steady flow than the pulsing year (Table 3.3).
57 In the steady flow year, the greatest rates of methane flux occurred in summer in all
-2 -1 wetland zones (7.18 ± 1.21, 6.25 ± 2.34, and 18.50 ± 3.97 mg CH4-C m h in edge
zones with and without emergent macrophytes, and continuously inundated zones
respectively), and more pronounced differences were observed between the seasons
(Figure 3.8). Between years, the only significant differences in methane flux within
treatments were observed in continuously inundated zones, with mean rates over the
-2 -1 entire year of 5.65 ± 1.01 mg CH4-C m h during the pulsing year and 11.06 ± 2.16 mg
-2 -1 CH4-C m h during the steady flow year (F = 18.631 209, 1 p < 0.001).
There were no differences in methane flux rates within years and wetland zones
according to time of day. Mean water depth within wetland zones was not significantly
different between the pulsed and steady flow years (15.1 ± 1.1 cm vs. 14.3 ± 0.5 cm in
continuously inundated zones, 3.9 ± 1.3 vs. 4.5 ± 0.2 in edge zones with emergent
macrophytes, and 4.1 ± 1.6 vs. 6.7 ± 0.6 cm in edge zones without emergent
macrophytes). Water depth was significantly correlated with methane flux in relation to
wetland zone and year (F = 3.206 561, 5 p = 0.007) (Table 3.1), with the highest methane fluxes measured in the continuously inundated zones (the deepest areas). Soil temperature was positively correlated with methane flux for all wetland zones combined,
2 in both years (r = 0.30, F = 6.597 787, 1 p < 0.011) (Figure 3.9). Soil temperatures were
significantly higher within each wetland zone in summer of both years (F = 40.073 858, 23
p < 0.001), with the exception of the continuously inundated zone in the pulsing year
(Table 3.4).
58 3.4.4 Carbon dioxide fluxes and solar radiation
Edge zones without emergent macrophytes did not produce a clear trend in CO2 fluxes, with about half of the daytime fluxes being positive (respiration > photosynthesis) and half negative (photosynthesis > respiration). CO2 fluxes from these zones were
therefore not included in the analysis. No significant differences in CO2 uptake were found within wetland zones between the pulsing and steady flow years, for either the growing or non-growing season (Figure 3.11a). CO2 uptake was significantly higher in edge zones with emergent macrophytes during the growing season than in continuously inundated zones in both years (F = 30.711 321, 7 p < 0.001). There were no significant
differences in mean rates of CO2 uptake between wetland zones during the non-growing season.
Respiration rates were not significantly different according to wetland zone or
growing vs. non growing season. Soil temperature, water depth and year were the only significant variables in the model for nighttime CO2 flux (Table 3.2). Mean respiration
- rate from all wetland zones combined during the pulsing year was 233 ± 29 mg CO2-C m
2 -1 -2 h , and during the steady flow year the mean respiration rate was 172 ± 37 CO2-C m
-1 h (F = 7.035 177, 1 p = 0.009). Water depth and soil temperature explained approximately
9 and 8 % of nighttime CO2 flux rates respectively (Figures 3.12 and 3.13), with
respiration rates generally increasing with increasing soil temperatures, and decreasing with increasing water depths.
Mean solar radiation data on each sampling date was plotted (Figure 3.12), and paired
t-tests revealed no significant difference in solar radiation between the pulsed and steady flow years. Recalculation of 2005 solar radiation values based on the relationship
59 obtained between OARDC and ORWRP solar radiation found a slight overestimation of
radiation for 2005 compared to actual measurements, but the difference was not
significant at α =0.05. Solar efficiency was comparable between the flood pulsed and
steady flow years in edge zones with emergent macrophytes and continuously inundated
wetland zones. In edge zones with emergent macrophytes, solar efficiency was 2.1 ±
0.7% and 2.1 ± 0.4% in the pulsed and steady flow years respectively, and in
continuously inundated zones solar efficiency was 0.7 ± 0.2 and 0.9 ± 0.2% in the
respective years.
3.5 Discussion
It was expected that periodically aerated conditions in the soil during 2004 would
inhibit methanogenesis and promote methane oxidation, significantly lowering rates of methane flux from edge zones compared to 2005, when soils were not exposed to the air.
Within the pulsing year, methane fluxes in edge zones were previously reported as significantly lower when the water level was below the soil surface than when these zones were inundated (Altor and Mitsch, 2006). However, the edge zones were inundated or saturated a greater percentage of the time than they were dry, and methane fluxes resumed quickly once soils were reflooded. The higher range in methane fluxes observed in edge zones without emergent macrophytes during flood pulsing, as compared to steady-flow conditions, may have been due in part to release of methane from sediments after flood pulses as hydrostatic pressure dropped with the falling water table.
Similar observations were made by Juttinen (2004), who suggested that shrinkage of drying sediments allowed the release of trapped methane.
60 Contrary to our first hypothesis, the presence of emergent vegetation did not appear to
positively influence methane emissions in edge zones under steady-flow conditions, where CH4 emissions were comparable to emissions from adjacent areas where emergent vegetation was removed. There was also no evidence of pressurized ventilation of
methane occurring through vegetation, as there were no differences in methane flux
according to time of day. While labile carbon deposited into sediments by root exudates
can facilitate methanogenesis, oxygen released from plant roots can inhibit
methanogenesis by increasing the availability of alternate electron acceptors including
3+ - Fe and NO3 (Conrad, 2002; Hernandez and Mitsch, 2007b), and can promote methane
oxidation. Areas in which emergent plant species were removed contained wetland vegetation, but species found in these areas lacked the extensive root and rhizome structures of emergent species, and thus may have contributed less oxygen and organic matter to the rhizosphere. On the other hand, high turnover rates of species such as algae and Lemna, which dominated areas where emergent macrophytes were removed, could contribute substantial amounts of carbon to the sediment surface.
While we hypothesized that the most dramatic differences in methane flux between
the two years would occur in the edge zones, CH4 fluxes from continuously inundated
zones were significantly higher in the steady flow year than in the pulsed year. One
possible explanation for this is a smaller range in water level during the steady flow year,
despite the similar average water level. During the pulsing year, methane fluxes reached
their highest rates when the water level was between +5 and +20 cm, in continuously
inundated areas (Altor and Mitsch, 2006). A shallower water depth results in a shorter
diffusive path between the sediments and atmosphere. Methane can be oxidized when
61 passing through the water column (Boon and Lee, 1997); a deeper water column provides
greater potential for methanotrophy as methane diffuses towards the atmosphere. In
addition to water depth, the average soil temperature in continuously inundated zones
during the steady flow year was 24.7 ± 0.5 °C compared with 23.1 ± 0.4 °C in 2004.
Shallower water levels and lower flow rates into the wetlands could have enabled warmer
water and soil temperatures in 2005. Microbial activity is strongly regulated by
temperature, and relatively small increases in sediment temperature have been shown to
correspond positively to rates of methane flux by other researchers (Whiting and
Chanton, 1992). In a concurrent study of sedimentation, Nahlik and Mitsch (in review)
found significantly greater total sedimentation during the pulsing year compared to the
steady flow year. Allochthonus sediments are dominated by mineral, as opposed to
organic, material and thus could have diluted the pool of organic matter available during
the pulsing year (Hernandez and Mitsch, 2007a).
Our third hypothesis, that net and gross rates of CO2 uptake would be higher in the pulsing year, was rejected. The lack of significant difference in rates of primary
productivity and respiration, and the equality of percent solar efficiency between the
years, suggest that wetland productivity was neither positively or negatively affected by
flood pulses. These metabolism results contrast with results from a comparison of
aquatic gross primary productivity (GPP) during the pulsing and steady flow years (Tuttle
and Mitsch, in revision). In that study, aquatic GPP measured in the pulsing year was
5.13 ± 0.3 kcal m-2 d-1, while GPP measured in the steady flow year was 11.4 ± 0.7 kcal
m-2 d-1; however, measurements in the pulsing year were only made during floods, when
plankton and benthic algae were flushed from the system (Tuttle and Mitsch, in revision).
62 A study of aboveground NPP in the same experimental wetlands found an average
productivity of 2278 kg yr-1 in the hydrologic pulsing year and 3530 kg yr-1 in the steady
flow year (Mitsch et al., 2006). In this study, gas sampling measurements were made in the same locations during each year, and measurements were taken throughout the growing (and non-growing) season, not just at peak biomass. This discrepancy in results suggests that it would be useful to conduct a future study in which gas fluxes are measured repeatedly over the year, but at random locations.
Substantial amounts of particulate organic matter (POM) were noted as discharging
from the wetlands from late April through mid June of the pulsing year. While this
outflow of POM was due in part to the activity of a beaver at the outflows of the
wetlands, the scouring action of the flood pulses loosened and carried out dead vegetation
and algae as well. Collection of the POM from the outflow of each wetland on nine
occasions revealed that between 0.1 – 9 kg (dry weight basis) of POM were being lost from the wetlands each day during this time period. The loss of aquatic vegetation
during pulses indicates the importance of taking productivity measurements at intervals
that include between-flood periods.
3.6 Conclusions
Flood pulsing vs. steady flow hydrology did not make a significant difference to CO2 uptake, respiration or solar efficiency, but steady flow hydrology was associated with significantly higher rates of methane flux form the deeper water zones of the wetlands, compared to flood pulse hydrology. Warmer soil temperatures and less fluctuation in water level and inflow rate during the steady flow year are likely to have facilitated
63 greater rates of methanogenesis. Emergent vegetation did not appear to ventilate
methane directly to the atmosphere, although organic matter contributed to the sediments
by emergent macrophytes plays a major role in methanogenesis throughout the various wetland zones. Antecedent conditions can be expected to influence dynamics observed in a given year. Another set of experiments is currently being conducted with steady-
flow conditions preceding hydrologic pulsing conditions. One objective of that
experiment will be to determine whether carbon dioxide and methane dynamics under
flood pulsed conditions will differ when preceded by steady flow hydrologic conditions.
Longer-term studies in a given ecosystem can further establish what differences in carbon
fluxes are due to interannual variability, antecedent conditions, or independent
experimental variables. Experimental manipulations of hydrology enable investigations
into ecosystem resiliency and adaptation of wetland systems to changes in this critical
parameter.
3.7 Acknowledgements
Funding for this research was provided by USDA NRI CSREES Award 2003-35102- 438
13518, by a Payne Grant from the Ohio Agricultural Research and Development Center
(OARDC) at The Ohio State University and by the Schiermeier Olentangy River Wetland
Research Park (ORWRP). Some support was also provided by a Rhonda and Paul Sipp
Wetland Research Award. Olentangy River Wetland Research Publication Number 07-
0xx. Help in the field provided by Kyle Chambers and Maria E. Hernandez is gratefully
acknowledged. Michael Holloman provided very valuable input into the statistical
64 analyses used in this study. Thanks to Jennie Morgan for providing a peer review of this
chapter.
3.8 Literature Cited
Altor, A. E., and W. J. Mitsch, 2006. Methane flux from created riparian marshes: relationship to intermittent vs. continuous inundation and emergent macrophytes. Ecological Engineering 28: 224 – 234.
Boon, P. I., and K. Lee, 1997. Methane oxidation in sediments of a floodplain wetland in south-eastern Australia. Letters in Applied Microbiology 25: 138 – 142.
Brix, H., B. K. Sorrell and P. T. Orr, 1992. Internal pressurization and convective gas flow in some emergent freshwater macrophytes. Limnology and Oceanography 37: 1420-1433.
Brix, H., B. K. Sorrell and H-H. Schierup, 1996. Gas fluxes achieved by in situ convective flow in Phragmites australis. Aquatic Botany 54: 151 – 163.
Brooks, R. P., D. H. Wardrop, C. A. Cole, and D. A. Campbell, 2005. Are we purveyors of wetland homogeneity? A model of degradation and restoration to improve wetland mitigation performance. Ecological Engineering 24: 331–340.
Christensen, T. R., N. Panikov, M. Mastepanov, A. Joabsson, A. Stewart, M. Öquist, M. Sommerkorn, S. Reynaud, and B. Svensson, 2003. Biotic controls on CO2 and CH4 exchange in wetlands – a closed environment study. Biogeochemistry 64: 337 – 354.
Conrad, R., 2002. Control of microbial methane production in wetland rice fields. Nutrient Cycling in Agroecosystems 64: 59-69.
Fennessy, M.S., and W. J. Mitsch, 2001. Effects of hydrology on spatial patterns of soil development in created riparian wetlands. Wetlands Ecology and Management 9: 103–120.
Freeman, C., G. B. Nevison, H. Kang, S. Hughes, B. Reynolds, and J. A. Hudson, 2002. Contrasted effects of simulated drought on the production and oxidation of methane in a mid-Wales wetland. Soil Biology and Biochemistry 34: 61—67.
Gilvear, D. J., and C. Bradley, 2000. Hydrological monitoring and surveillance for wetland conservation and management; a UK perspective. Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere 25: 571—588.
65 Healy, R. W., R. G. Striegl, T. F. Russell, G. L. Hutchinson, and G. P. Livingston, 1996. Numerical evaluation of static-chamber measurements of soil-atmosphere gas exchange: identification of physical processes. Soil Science Society of America Journal 60: 740-747.
Hernandez, M.E. and W.J. Mitsch. 2006. Influence of hydrologic pulses, flooding frequency, and vegetation on nitrous oxide emissions from created riparian marshes. Wetlands 26: 862-877.
Hernandez, M.E. and W.J. Mitsch. 2007a. Denitrification potential and organic matter as affected by vegetation community, wetland age, and plant introduction in created wetlands. Journal of Environmental Quality 36: 333-342.
Hernandez, M.E. and W.J. Mitsch. 2007b. Denitrification in created riverine wetlands: Influence of hydrology and season. Ecological Engineering, in press.
Holland, E. A., G. P. Robertson, J. Greenberg, P. M. Groffman, R. D. Boone and J. R. Gosz, 1999. Soil CO2, N2O, and CH4 Exchange. Chapter 10 in Robertson, G. P., D. C. Coleman, C. S. Bledsoe, and P. Sollins, Eds. Standard Soil Methods for Long Term Ecological Research. Oxford University Press, New York, NY.
Hunt, R.J., Krabbenhoft, D.P., and Anderson, M.P., 1997, Assessing hydrogeochemical heterogeneity in natural and constructed wetlands. Biogeochemistry 39: 271–293.
IPCC, 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Houghton, J. T., Ding, Y., Griggs, D. J., Noguer, M., van der Linden, P. J., Dai, S., Maskell, K., Johnson, C. A. (Eds.), Cambridge University Press, Cambridge, UK and New York, NY, 881 pp.
Juutinen, S., 2004. Methane fluxes and their environmental controls in the littoral zone of boreal lakes. University of Joensuu, PhD Dissertations in Biology, number 25. ISSN 1457-2486. 110 pp.
Käki, T., A. Ojala and P. Kankaala, 2001. Diel variation in methane emissions from stands of Phragmites australis (Cav.) Trin. Ex Steud. and Typha latifolia L. in a boreal lake. Aquatic Botany 71: 259-271.
Le Mer, J., and P. Roger, 2001. Production, oxidation, emission and consumption of methane by soils: A review. European Journal of Soil Biology 37: 25—50.
McKee, K. L., and P. L. Faulkner, 2000. Restoration of biogeochemical function in mangrove forests. Restoration Ecology 8: 247—259.
Megonigal, J.P., W.H. Conner, S. Kroeger and R.R. Sharitz. 1997. Aboveground
66 production in southeastern floodplain forests: a test of the subsidy-stress hypothesis. Ecology 78: 370-384.
Megonigal, J.P., M.E. Hines, and P.T. Visscher, 2004. Anaerobic Metabolism: Linkages to Trace Gases and Aerobic Processes. Pages 317-424 in Schlesinger, W.H. (Editor). Biogeochemistry. Elsevier-Pergamon, Oxford, UK.
Mitra, S., R. Wassmann, and P. L. G. Vlek, 2005. An appraisal of global wetland area and its organic carbon stock. Current Science 88: 25 – 35.
Mitsch, W. J., and W. G. Rust, 1984. Tree growth responses to flooding in a bottomland forest in northeastern Illinois. Forest Science 30: 499—510.
Mitsch, W.J., L. Zhang, C. J. Anderson, A. E. Altor, and M. E. Hernandez, 2005. Creating riverine wetlands: Ecological succession, nutrient retention, and pulsing effects. Ecological Engineering 25: 510 – 527.
Mitsch, W.J., L. Zhang, M.E. Hernandez, A.E. Altor, A.M. Nahlik, C.L. Tuttle, D.F. Fink, and C. J. Anderson, 2006. Importance of hydrologic pulsing on the water quality function of wetlands in Midwestern USA. Abstracts, American Water Resources Association Annual Meeting, Baltimore, MD.
Mitsch, W.J. and J.G. Gosselink. 2007. Wetlands, 4th edition. John Wiley & Sons, Inc., New York.
Montalto, F. A., and T. S. Steenhuis, 2004. The link between hydrology and restoration of tidal marshes in the New York/New Jersey estuary. Wetlands 24: 414—425.
Moore, T. R., and J. Turunen, 2004. Carbon accumulation and storage in mineral subsoil beneath peat. Soil Science Society of America Journal 68: 690 – 696.
Nahlik, A.M. and W.J. Mitsch. In review. The effect of river pulsing on sedimentation in created riparian wetlands. Journal of Environmental Quality.
Owen Koning, C., 2005. Vegetation patterns resulting from spatial and temporal variability in hydrology, soils, and trampling in an isolated basin marsh, New Hampshire, USA. Wetlands 25: 239-251.
SPSS, 2004. Version 11.0.3. SPSS, Inc. Chicago, Illinois.
Tuttle, C.L. and W.J. Mitsch. In revision. Aquatic metabolism as an indicator of the ecological effects of hydrologic pulsing in flow-through wetlands. Ecological Indicators.
67 Vaithiyanathan, P., and C. J. Richardson, 1997. Nutrient profiles in the everglades: examination along the eutrophication gradient. The Science of the Total Environment 205: 81 –95.
Vretare Strand, V., 2002. The influence of ventilation systems on water depth penetration of emergent macrophytes. Freshwater Biology 47: 1097—1105.
Waddington, J. M., N. T. Roulet, and R. V. Swanson, 1996. Water table control of CH4 emission enhancement by vascular plants in boreal peatlands. Journal of Geophysical Research 101 (D17): 22,775—22,785.
Whalen, S. C., 2005. Biogeochemistry of methane exchange between natural wetlands and the atmosphere. Environmental Engineering Science 22: 73—94.
Whiting, G. J. and J. P. Chanton, 1992. Plant-dependent CH4 emission in a subarctic Canadian fen. Global Biogeochemical Cycles 6: 225-231.
Whiting, G. J. and J. P. Chanton, 1996. Control of the diurnal pattern of methane emission from emergent aquatic macrophytes by gas transport mechanisms. Aquatic Botany 54: 237, 253.
Whiting, G. J. and J. P. Chanton, 2001. Greenhouse carbon balance of wetlands: methane emission versus carbon sequestration. Tellus 53B: 521-528.
Yavitt, J. B., and A. K. Knapp, 1998. Aspects of methane flow from sediment through emergent cattail (Typha latifolia) plants. New Phytologist 139: 495—503.
Zedler, J. B., and J. C. Callaway, 1999. Tracking wetland restoration: Do mitigation sites follow desired trajectories? Restoration Ecology 7: 69—73.
68
Figure 3.1. Wilma H. Schiermeier Olentangy River Wetland Research Park (ORWRP). Research described in this study was conducted in ‘experimental wetland 1’ and ‘experimental wetland 2.’
69 160 Wetland 1 Wetland 2
120 ) -1
80 Infow (cm day 40
0 J- F- M- A- M- J- J- A- S- O- N- D- J- F- M- A- M- J- J- A- S- O- N- D- 04 04 04 04 04 04 04 04 04 04 04 04 05 05 05 05 05 05 05 05 05 05 05 05 Figure 3.2. Hydrology over the two-year study period illustrating flood pulses in 2004 and steady-flow conditions in 2005.
Figure 3.3. Schematic of the two experimental wetlands at the ORWRP at The Ohio State University, Columbus, and the three wetland zones in which gas sampling was carried out. “Edge” zones contained chambers with and without emergent macrophytes; “Marsh” zones in the inflow exhibited dynamics comparable to edge zone chambers with emergent macrophytes; “Marsh” zones in the outflow were comparable to chambers in continuously-inundated zones. “CI” refers to continuously inundated zones in which floating chambers were deployed. Straight lines represent permanent boardwalks located throughout the wetlands. 70
Figure 3.4. Non-steady-state chamber design, illustrating PVC frame left in wetlands for duration of experiment, and placement of plastic bag over chamber before gas sampling.
1000 y = 2.0787x - 13.454 900 R2 = 0.9432 800
700
600
500
400
300
200
100
0 0 100 200 300 400 500
-2 -1 mg CO2-C m h , 30 - 60 min
Figure 3.5. Carbon dioxide flux rates measured within non-steady-state chambers according to sampling time. Each point represents a flux rate measured after approximately a half hour (Y-axis) and an hour (X axis) of closing an individual chamber. This relationship was used to normalize 2004 data before they were compared to 2005 data.
71
Figure 3.6. Percent cover of dominant plant species in each wetland zone during the pulsing year (2004) and the steady-flow year (2005).
72
12 Fall Spring Summer c 10 Winter abc c
8 bc -1
h ac -2 6 -C m 4
4
mg CH ad ad 2
d d d d d 0 Continuously Edge zones with Edge zones without inundated zones emergent emergent macrophytes macrophytes -2
Figure 3.7. Mean methane flux rates within each wetland zone during the flood pulsed year (2004). Different letters represent a significant difference (p<0.05) between wetland zones or seasons. Bars represent standard error.
73 25 b Fall Spring 20 Summer Winter
-1 15 h -2
d
-C m ad 4 10 d mg CH 5 c ac ac c c c c c 0 Continuously Edge zones with Edge zones without inundated zones emergent macrophytes emergent macrophytes
-5
Figure 3.8. Mean methane flux rates within each wetland zone during the steady flow year (2005). Different letters represent a significant difference (p<0.05) between wetland zones or seasons. Bars represent standard error.
74
Figure 3.9. Relationship between soil temperature (°C) and methane flux rates (mg CH4- C m-2 h-1) for all wetland zones during the flood pulsed and steady flow years.
Figure 3.10. Relationship between soil temperature (°C) and nighttime carbon dioxide -2 -1 flux rates (mg CO2-C m h ) for all wetland zones during the flood pulsed and steady flow years.
75
Figure 3.11. Relationship between water depth (cm above or below soil surface) and -2 -1 nighttime carbon dioxide flux rates (mg CO2-C m h ) for all wetland zones during the flood pulsed and steady flow years.
76 0
-100 b b b -200 -1
h bc
-2 bc -300 -C m 2 -400 c 2004
mg CO -500 2005 -600 a a -700 Edge zones with Edge zones with Continuously Continuously emergent emergent inundated inundated macrophytes, macrophytes, zones, growing zones, non Growing season Non growing season growing season season
Figure 3.12. Mean growing- and non-growing season carbon dioxide uptake for the pulsing and steady-flow years. Different letters represent a significant difference (p<0.05) between wetland zones or years. Bars represent standard error. Flux rates are shown as negative numbers to represent removal of carbon dioxide from the chamber environment.
77 2004 (Calculated) 9000 2005 (Actual) 2005 (Calculated) 8000
7000
6000
5000
4000
3000
2000
1000
0 100 150 200 250 300 Julian day
Figure 3.13. Average daily solar radiation on each sampling day for both years of the study. 2004 data were calculated from the relationship between solar radiation at ORWRP and Delaware, OH data for the period Nov 25, 2004 – May 2, 2005. The equation obtained from this relationship was used to calculate solar radiation for each sampling date in 2005 in order to check the accuracy of the equation.
78
Dependent Variable: Signed square root, methane flux rate
df F-value p-value
Corrected Model 21 9.319 <0.001 Wetland zone 2 6.512 0.002 Season 3 4.81 0.003 Year 1 1.272 0.26 Soil temperature 1 6.597 0.011 Water depth 1 2.264 0.133 Wetland zone * year * water depth 5 3.206 0.007 Wetland zone* season * year * soil temperature 14 3.873 <0.001
Table 3.1. General Linear Model for methane flux from each wetland zone, for the pulsed and steady flow years.
Dependent Variable: CO2 Flux df F-value p-value
Corrected Model 3 7.812 < 0.001 Year 1 7.035 0.009 Soil temperature 1 6.1 0.015 Water depth 1 11.224 0.001
Table 3.2. General Linear Model for nighttime CO2 flux from edge zones with emergent macrophytes and continuously inundated wetland zones, for the pulsed and steady flow years.
79
Wetland Zone* N Mean SE‡ Min. Max.
C. I., fall 2004 37 5.59 1.16 -0.36 25.9 C. I., spring 2004 34 5.58 3.05 0 98.06
C. I., summer 2004 41 7.14 1.21 0.17 34.27
C. I., winter 2004 10 0.01 0.01 0 0.11 C. I., fall 2005 15 0.97 0.39 -1.7 4.57 C. I., spring 2005 29 6.53 2.26 -0.51 63.6 C. I., summer 2005 41 18.50 3.98 0 147.49 C. I., winter 2005 2 0.00 0.00 0 0 Edge E.M., fall 2004 37 0.41 0.13 0 3.22 Edge E.M., spring 2004 105 1.88 0.43 -0.25 34.36 Edge E.M., summer 2004 53 6.61 0.78 0.05 19.69 Edge E.M., winter 2004 24 -0.01 0.01 -0.15 0 Edge E.M., fall 2005 26 0.18 0.07 -0.5 0.8
Edge E.M., spring 2005 152 1.26 0.22 -7.01 15.99
Edge E.M., summer 2005 79 7.18 1.22 -0.49 43.18 Edge E.M., winter 2005 8 -0.09 0.32 -2.12 1.11 Edge no E.M., fall 2004 24 0.14 0.10 -1.44 1.5 Edge no E.M., spring 2004 67 2.24 0.69 -0.1 33.42 Edge no E.M., summer 2004 32 7.40 2.78 0.02 71.16 Edge no E.M., winter 2004 21 0.00 0.00 -0.03 0.03 Edge no E.M., fall 2005 14 0.11 0.06 -0.12 0.63 Edge no E.M., spring 2005 65 1.31 0.34 -0.56 15.99 Edge no E.M., summer 2005 24 6.25 2.34 0 43.18 Edge no E.M., winter 2005 12 0.03 0.03 0 0.32
Total 952
* C. I. = Continuously inundated; Edge E.M. = Edge zones with emergent macrophytes; Edge no E.M. = Edge zones without emergent macrophytes ‡ SE = Standard error of the mean
-2 -1 Table 3.3. Mean, minimum and maximum methane fluxes (mg CH4-C m h ) from each wetland zone in each season in the pulsing (2004) and steady flow (2005) years.
80
Wetland Zone* N Mean SE‡ Min. Max.
C. I., fall 2004 39 17.49 0.96 11.75 27.25 C. I., spring 2004 33 19.65 0.79 9.9 28
C. I., summer 2004 23 24.98 0.32 21 27.25
C. I., winter 2004 3 4.79 0.76 3.38 6 C. I., fall 2005 16 13.16 0.21 11.75 14.75 C. I., spring 2005 31 18.13 1.27 5.5 30.5 C. I., summer 2005 38 27.14 0.32 23.5 30.75 C. I., winter 2005 2 6.50 0.50 6 7 Edge E.M., fall 2004 37 16.71 0.60 11.25 24.5 Edge E.M., spring 2004 108 18.96 0.50 7.5 26.35 Edge E.M., summer 2004 54 23.74 0.24 20.3 28.5 Edge E.M., winter 2004 4 2.25 0.18 2 2.75 Edge E.M., fall 2005 25 9.72 0.45 6.25 14
Edge E.M., spring 2005 154 17.31 0.57 4.25 34.5
Edge E.M., summer 2005 73 26.71 0.29 21.25 31.63 Edge E.M., winter 2005 10 5.70 0.25 3.5 6 Edge no E.M., fall 2004 23 16.96 0.73 11.25 24 Edge no E.M., spring 2004 69 18.35 0.63 7.6 26.35 Edge no E.M., summer 2004 29 23.54 0.44 18.75 27.88 Edge no E.M., winter 2004 11 5.42 0.56 1 7.8 Edge no E.M., fall 2005 9 9.08 0.63 6.25 13 Edge no E.M., spring 2005 59 17.43 0.91 5 30.15 Edge no E.M., summer 2005 28 26.84 0.50 22.25 31.63 Edge no E.M., winter 2005 4 6.00 0.00 6 6
Total 882
* C. I. = Continuously inundated; Edge E.M. = Edge zones with emergent macrophytes; Edge no E.M. = Edge zones without emergent macrophytes ‡ SE = Standard error of the mean
Table 3.4. Mean, minimum and maximum soil temperature (°C) from each wetland zone in each season in the pulsing (2004) and steady flow (2005) years.
81
CHAPTER 4
METHANE EMISSIONS AND CARBON DIOXIDE FLUXES FROM WETLAND
MESOCOSMS: EFFECTS OF HYDROLOGIC REGIME AND HYDRIC SOILS
4.1 Abstract
The hydrology and physiochemical properties of soils forming the foundation for created
and restored wetlands determine what processes are likely to occur in these systems. In
this study we investigated effects of intermittent vs. continuous inundation, and hydric
and non-hydric soils, on fluxes of the greenhouse gas methane from 20 wetland
mesocosms. Fluxes of carbon dioxide and relationships between emergent vegetation
and methane emissions were also examined. The hydrologic treatments represented contrasting wetland restoration scenarios, and the soil treatments represented newly created and established wetlands. Hydric soils and continuously inundated treatments exhibited the greatest methane flux, while intermittently inundated conditions reduced methane fluxes significantly from hydric soils. Methane fluxes were not affected significantly by hydrologic treatment in mesocosms containing non-hydric soils. No
relationship was observed between emergent vegetation and methane flux, and carbon
82 dioxide and methane fluxes were not directly correlated. However, the highest rates of both CO2 uptake and CH4 flux were observed in treatments with continuously inundated hydrology. As part of soil organic matter determination, we evaluated the time of combustion and found a significant difference in soil organic matter from one to three hours of combustion in both hydric and non-hydric soils combusted at 550°C.
Microbially available organic carbon content was significantly greater in hydric soils than non-hydric soils, despite similar organic matter contents in the contrasting soil types.
Methane fluxes from created wetland mesocosms fell within the ranges reported for comparable, natural wetlands.
4.2 Introduction
Wetlands perform complex and important biogeochemical functions in landscapes, including the production, sequestration and release of carbon compounds and gases. As a general rule, wetlands act as both sinks for atmospheric carbon dioxide and sources of the greenhouse gas methane (Mitsch and Wu, 1995; Bouchard and Cochran, 2002; Whalen,
2005). In addition to climate, hydrology and soils play major roles in determining the nature of wetland carbon dynamics, in both natural and created or restored ecosystems.
Riparian wetlands, located within the original flood zone of rivers and streams, receive water in pulses during floods. Between floods, these wetlands may dry out or become partially unsaturated. Numerous non-riverine wetland ecosystems, including prairie potholes (van der Valk and Pederson, 2003), impounded marshes (Johnson Randall and
Foote, 2005), vernal pools (Brooks and Hayashi, 2002), and swamps (Carter et al., 1994) experience a dynamic hydroperiod defined by seasonal or annual patterns of inundation
83 and drying. These changes in water level are driven by precipitation, runoff and evapotranspiration, in combination with the water table location, the position of the wetland in the landscape, and the nature of the substrate underlying the ecosystem
(Carter, 1996).
Periodic drawdowns of the water table during the growing season, such as often
occurs in natural or restored riparian wetlands, have been demonstrated to reduce
methane flux significantly on the ecosystem scale (Altor and Mitsch, 2006). Flood
pulses have been both positively (Sommer et al., 2001, Ahearn et al., 2006) negatively
(Burke et al., 1999) and neutrally (Megonigal et al., 1997) correlated with primary
productivity, and it is recognized that gradients of subsidies (e.g. nutrients, sediments)
and stresses (e.g. anoxia, dessication, turbulence) interact with biological processes in
ecosystems that receive flood pulses (Odum et al., 1979, Burke et al., 1999). The
successional stage of riparian ecosystems in part determines whether flood pulses act as a subsidy or stress. Productivity in early successional systems dominated by phytoplankton and emergent macrophytes may be enhanced by flood pulses (Ahearn et al., 2006), while later successional stages dominated by trees may be stressed (Burke et al., 1999) or neutrally affected by these pulses. In all cases, spatial and biogeochemical heterogeneity
will generally lead to a gradient of responses to flooding and dry down within wetland
ecosystems (Mitsch, 1988, Megonigal et al., 1997).
Major biogeochemical functions in wetlands take place in the context of the soil or
substrate that forms the foundation of the system. When wetlands are constructed in
areas lacking hydric soils, soils that have never been extensively flooded, or at least not
recently so, are subjected to prolonged saturated conditions. Among the many changes
84 that can occur in soils upon flooding are reduction and mobilization of metal cations
3+ 4+ - (particularly Fe and Mn ) and nutrient anions (eg NO3 ), anaerobic production and
release of methane (CH4) and other reduced gases (e.g. N2, N2O, H2S), gradual
accumulation of refractory organic matter, and changes in microbial community
composition (Craft, 2001; Vepraskas and Faulkner, 2001). These changes occur at
varying time scales, depending on water chemistry and the physical and chemical
composition of the soil prior to flooding. The development of hydric soil characteristics
including mottles, oxidized pore linings and redox depletions is an important indication
that wetland hydrology is present (National Research Council, 1995).
Vegetation plays a variety of roles in wetland carbon cycling. Besides
photosynthetic uptake of atmospheric CO2, some wetland plant species transport O2 from the atmosphere to the rhizosphere (root zone), and deliver gases from the soil to the atmosphere. Pressurized ventilation of methane and other gases, in which pressure and temperature differentials drive convective flows between the atmosphere and soil through vascular plant tissue, have been demonstrated for a variety of emergent and submerged plant species (Schütz et al., 1991; Brix et al., 1996; Grosse, 1996). When methane is being released to the atmosphere by pressure-driven convection, the highest rates of gas flux generally occur in the morning when accumulated gases are released from lacunae as plant tissue warms up and convection starts (Whiting and Chanton, 1996), or during midday when light intensity and temperature are at a maximum (Brix et al., 1992). Some researchers have postulated that a consistent proportion of CO2 taken up by wetland
plants may be returned to the atmosphere as CH4 (Whiting and Chanton, 1993).
Vegetation provides the foundation for autochthonous carbon inputs into wetlands,
85 carbon that can contribute to both the stable, sequestered pool and the labile, microbially active pool.
The purpose of this study was to examine relationships between pulsed vs. continuously inundated hydrology, hydric vs. non-hydric soils, vegetation, methane and carbon dioxide fluxes in replicated wetland mesocosms. The utility of mesocosm and microcosm experiments to test parameters of restoration design has been recognized by other researchers (Calloway et al., 1997; Catallo and Junk, 2003). Our research objectives were to assess how methane and carbon dioxide fluxes and soil properties can be expected to vary between newly created vs. established created wetlands. The effect of ponded (continuously inundated) vs. wet/dry hydrologic conditions on macrophyte productivity and methane fluxes were examined in light of potential implications for wetland restoration. Soil parameters considered relevant to methane flux and wetland age included total carbon, nitrogen and organic matter contents, Munsell color and redoximorphic features, and labile water-soluble carbon as determined by cold and hot water extractions. Temporal patterns in methane and CO2 fluxes were monitored to evaluate primary productivity and the influence of vegetation on methane flux. In addition, metholodogical issues relating to soil physiochemical analyses commonly performed on wetland soils were explored, including combustion time for soil organic matter, and acid treatment for carbonate removal.
86 4.3 Methods
4.3.1 Site description
The study was conducted at the Schiermeier Olentangy River Wetland Research Park,
on the campus of The Ohio State University (OSU), Columbus, Ohio USA (40°0’N,
83°1’E). The site houses an outdoor mesocosm compound that includes four sets of
twenty 540 L high-density black plastic tubs, buried in the ground. Other research
conducted using mesocosms on this site has included investigations into the growth of
Typha latifolia L. and Schoenoplectus tabernaemontani K. C. Gmel under conditions of nutrient enrichment (Svengsouk and Mitsch, 2001), scaling issues and use of flue gas desulfurization material as a wetland liner (Ahn and Mitsch, 2002a,b), and response of vegetation to pulsing vs. continuously inundated hydrology (Anderson and Mitsch,
2005). The mesocosm tubs are designed to be flow-through systems, with a drainage outlet at the end of each tub. Drainage also occurs through French drains that discharge water after it has seeped through the soil profile. Water levels are controlled using
standpipes of various heights placed into the French drains (Fig. 1a).
4.3.2 Study design
Twenty mesocosm tubs, containing rinsed pea gravel to a depth of 10 cm depth, were
filled with approximately 35 cm of soil in autumn 2003. Half of the tubs received non-
hydric soil retrieved from excavation of a building site on the OSU campus, and half of the tubs received hydric soil removed with a backhoe from the upper 10 cm of an onsite oxbow wetland that was created in 1997. The hydric soils were taken from an area of the oxbow that was sparsely vegetated, and were examined for Munsell hue, value and
87 chroma prior to excavation to verify that they met hydric criteria. The non-hydric soils
were classified as Ross silt loam - mixed, mesic Cumulic Hapludolls. Cumulic
Hapludolls, of the Mollisol order, formed under humid, temperate climate conditions at
the bottom of slopes and on floodplains; they are often calcareous and have generally
been used as cropland (USDA, 1999). The Ross series are well-drained, moderately
permeable soils originating from alluvium (SCS, 1980). The type of soil each tub would receive (hydric vs. non-hydric) was chosen randomly, the only criteria being that ten tubs would receive each soil type. Soils were allowed to settle until spring 2004, at which point each tub was planted with three rhizomes of Sparganium eurycarpum Engelm, three rhizomes of Schoenoplectus tabernaemontani K.C. Gmel, and seeds of Leersia oryzoides L. Sw. Each of these plant species had been dominant in onsite experimental wetlands in 2003 (Mitsch et al., 2005).
After planting, all mesocosms were maintained under saturated conditions throughout
the 2004 growing season to allow vegetation to establish. No weeding was conducted;
non-planted species were considered part of the potential plant community for each soil
type. In March 2004, the first set of soil cores was taken; a second set was taken in
March 2005. Hydrologic treatments were assigned randomly to the mesocosms, within
soil types. Five tubs with each soil type were assigned pulsed hydrology, and five were
assigned continuously inundated hydrology, resulting in four treatments of five tubs each.
Experimental hydrology and gas sampling were initiated in April 2005, and continued
through September 2005.
88 4.3.3 Hydrology
The objectives for each hydrologic treatment were to 1) maintain saturated or
inundated conditions at all times for continuously inundated conditions; 2) create a
minimum of one cycle of alternating inundated/dry conditions each month for pulsed
conditions. Experimental hydrologic conditions were established in the mesocosms in
April 2005. A groundwater pump was used to deliver water to each mesocosm via a
hose, with approximately the same cumulative volume delivered to each mesocosm over
the course of each month. Continuously inundated mesocosms received water daily,
except when precipitation was adequate to maintain inundated conditions, and pulsed
mesocosms received the same amount of water as continuously inundated mesocosms,
but at higher rates (“flood pulses”) and for fewer days. Daily temperature and
precipitation dynamics determined how many inundated/dry cycles could be achieved in
pulsed mesocosms in a given month.
Porewater was sampled from each mesocosm standpipe on June 1 and August 4,
2005, to examine total organic and inorganic carbon content. The volume of each
standpipe was calculated, and that amount of water was removed and discarded back into the mesocosm before the sample was collected, to ensure that the water collected for porewater samples had not been stagnant in the pipes, but rather was emerging freshly from infiltration through the soil. Porewater samples were collected using a rubber suction bulb attached to Tygon tubing into acid-washed Nalgene bottles, and stored at
4°C until analysis on a Shimadzu TOC5050A with ASI5000A autosampler.
89 4.3.4 Gas sampling
Gas sampling was conducted using non-steady-state chambers (Fig. 1b) designed
after Klinger et al. (1994) and Altor and Mitsch (2006). Using a jigsaw, the bottoms were removed from opaque, 51- by 36-cm rectangular plastic tubs. The tubs had flat sides to facilitate clean cuts with the jigsaw. Transparent, 0.1-mm (4-mil) polyethylene bags (approximately 1.5-m tall) were attached to the plastic bases using weatherproof transparent tape applied inside and outside of the bag. The top of each bag was cut open, and the bags were kept rolled down around the chamber bases between sampling events.
One plastic base, with bag attached, was inserted 3-5 cm into the soil in the center of each mesocosm in April 2005. Once in place, chamber bases were not removed for the remainder of the study. 130-cm tall PVC frames (made from 3.8-cm dia pipe) were installed in each mesocosm. A rectangular PVC support, with a wire for attaching a thermometer, was placed onto the tops of the legs to complete the frame. These frames served as supports for the bags, and were left in place throughout the study.
During gas sampling, the bags were rolled up around the chamber frames, and sealed
at the top with 0.5 cm diameter rubber bands. The top of each bag was affixed with a
grey butyl rubber sampling port and 2-m Tygon tubing for equilibrating the chamber with
atmospheric pressure. Sampling was conducted one day before flood pulses were
delivered to pulsed tubs, and one day after, as well as on numerous occasions between
flood pulses. Sampling sessions were conducted over approximately 1.5 hours in the
morning (between 7:30-10:30), afternoon (between 12:30-4:30) and after dark. Five gas
samples were collected from the headspace of each mesocosm chamber over 20-30
minutes, into pre-evacuated 10 ml autosampler vials. The mesocosms at which gas
90 sampling was started were chosen randomly each sampling day. Methane fluxes were
measured directly in the created oxbow on June 13, June 19 and July 7 2006, using the
same type of chamber as used in the mesocosms; this extra sampling was conducted to
determine whether methane emissions from hydric soils varied significantly between the
natural and the mesocosm environments. Gas sampling chambers were placed in the
oxbow wetland close to the area from which the soils were taken for hydric mesocosms.
Environmental parameters measured in each mesocosm during gas sampling included
soil temperature at 5 and 10 cm depths, temperature within the chamber when each gas
sample was withdrawn, and water level. Percent cover of plant species in the chamber
area of each mesocosm was estimated approximately two times per month, on or close to
sampling days. Dominant plant species were determined for each treatment using the
50/20 rule (USACOE, 1987) applied as follows: 1) mean percent cover of each species for all sampling dates was obtained and multiplied by the number of sampling dates on which the species was observed; 2) total percent cover for all species was summed and multiplied by 50% (0.5) and 20% (0.2); 3) the species with the greatest percent cover were summed until the value obtained for 50% of the total percent cover was reached; these species were the dominants in a given treatment. No individual plant species attained 20% cover during this study.
Gas samples were analyzed on a Shimadzu GC 14A equipped with an HTA
Autosampler, with a thermal conductivity detector (TCD) and flame ionization detector
(FID) in series. A 1.8-m Porapaq-Q column was used for sample separation, with helium
(approximately 25 ml min-1) as the carrier gas. The GC oven and injection temperatures
were maintained at 40°C; detector temperatures were 200°C (TCD) and 150°C (FID).
91 Four-point calibration curves were prepared for each GC run using Matheson gas
standards. The calibration curves consisted of 5, 10, 15 and 20 ppm CH4, with ultrapure
N2 as the balance, and 250, 500, 750 and 1000 ppm CO2 balanced with helium. Check
standards were injected during each run to verify consistency of the analysis. Gas
samples were stored at 4°C until analysis, and were analyzed within one week of
collection. Concentration by volume of methane and carbon dioxide in each sample were
-2 -1 converted to flux rates (mg CH4-C and mg CO2-C m h ), corrected for chamber volume
and temperature (Healy et al., 1996). Regressions were performed on each flux rate in
Microsoft Excel™ to determine linearity of flux. Non-linear (R2 < 0.88) flux rates were
discarded, or recorded as zero as appropriate. Where removing a sample corrected a poor
correlation to > 0.90, the sample was eliminated from the calculation (Holland et al.,
1999).
4.3.5 Soil sampling
In March 2004, three soil cores were taken from each mesocosm using a stainless
steel soil corer, (two cm inner diameter) pushed into the soil profile as far as possible.
The length of each core was recorded, and Munsell color, hue and chroma were
determined in the field for the matrix and for mottles. Cores were taken near the center
of the tubs, at each end and middle. The stainless steel corer was rinsed and dried
between samples. Because the soil was not naturally stratified, having been shoveled into the tubs, cores were not analyzed by depth interval. Instead, the lengths of the three cores
taken from each tub were measured to determine volume, cores combined into one plastic
bag and stored in a cooler (4°C) until further analysis. The soil combined soil cores from
92 each tub were dried at 105°C in the laboratory until constant weight to determine bulk
density (USDA, 2004). The soil samples were then ground in a mortar and pestle and
sieved to <2 mm diameter. One subsample (approximately 5 g) from each mesocosm tub
was measured into a 10 ml porcelain crucible, and combusted in a Fisher Scientific
Isotemp forced-draft furnace for one hour at 550°C for determination of percent soil
organic matter (SOM) (Anderson et al., 2005). A subsample of each original
homogenized soil core was analyzed for available iron by DTPA extraction at STARLab,
Wooster, OH. DTPA, diethylenetriaminepentaacetic acid, is an organic chelate used to
bind certain metals (e.g. Mn4+, Fe3+) under neutral and alkaline conditions, which keeps
the metals from precipitating with hydroxides or phosphates, maintaining them in
solution and available for plant uptake (Elrashidi et al., 2003). The affinity of Mn4+ and
Fe3+ for the DPTA is stronger than their attraction to the soil matrix, making the chelate
an effective means of extraction (Hong et al., 2002). Subsamples were also analyzed for
total carbon and nitrogen content by combustion in an Elementar America’s VarioMAX
analyzer at STARLab.
To determine whether carbonates were present, a subsample of each composited
sample was treated with approximately 0.5 ml 10N hydrochloric acid (fizzing indicates that soil contains carbonates), and samples were dried at 105°C until constant weight
(0.001 g) as measured on a Mettler balance. 20% of the samples were acid-treated in duplicate. The reaction between HCl and CaCO3 is as follows;
CaCO3 + 2HCl → CaCl2 + CO2 + H2O
When Cl2 replaces CO3 bound to the metal cation, a net gain in weight of 10.89 g
-1 mol occurs, assuming that CO2 and H2O are lost to the atmosphere during drying.
93 A second set of soil cores was taken in March 2005 using the same soil corer, after the mesocosms were saturated continuously for one year. Three cores were taken from
each tub, in the center and near each end. One 2-cm section was cut from each core and placed in a plastic bag for laboratory analysis. If there was a visible difference in texture or color along the length of the core, a 2-cm section was taken from each portion. The
soil samples from each mesocosm were not combined as they had been in 2004. Soil
samples were kept in the shade and brought into the laboratory within one hour of
collection. Each of the three or more samples from each tub was analyzed separately
using the methods described above for bulk density. Soil samples were combusted for
three hours, in one hour intervals, at 550°C (Anderson et al., 2005). After each hour, combusted soils were cooled to 105°C, placed in a dessicator for approximately 30
minutes, and reweighed to determine the percent soil organic matter content (SOM) by
loss on ignition.
Water extractable organic matter (WEOM) < 0.45 μm diameter was determined by
sequential cold and hot water extractions on each of the 2005 soil samples, after Nguyen
(2000). WEOM in this size class (dissolved) is that fraction considered to be potentially
available to microbes (Zsolnay, 2003), because microbial metabolism and physical
processes are dependent upon an aquatic environment (Marschner and Kalbitz, 2003).
WEOM is generally considered to be labile (Sparling et al., 1998; Chantigny, 2003). Two
grams of dry soil were combined with 30 ml of deionized water (20°C) in 50-ml
centrifuge tubes, placed horizontally on the shaker tray of a controlled temperature
Precision Reciprocal Shaking Bath Model 25 and shaken at 120 rpm for 18 hours.
Samples were then centrifuged at 2600 rpm for 15 minutes, and the supernatant was 94 filtered through sterile 0.45 μm Whatman polyethersulfone membrane syringe filters.
Filtrate was analyzed for total carbon (TC) and total inorganic carbon (TIC) on the
Shimadzu TOC5050A. 30 ml of deionized water was then added to each centrifuge tube,
and the slurries were shaken at 120 rpm for 18 hours at 80°C. Each sample was
centrifuged, filtered and analyzed as above, for hot-water extractable carbon. 20% of the
soil samples were tested in duplicate for both cold and hot-water extractable organic
carbon. Blank samples of deionized water were analyzed for background concentrations
of TC and TIC. WESOM values were obtained by subtracting TIC from TC, and
multiplying by the van Bemmelen organic matter factor 1.72, dividing by 2 (two g of soil
were used for the extraction) and multiplying by 30 (30-mls of water for the extraction).
Values thus obtained were reported as mg WESOM kg-1 soil.
4.3.6 Data analysis
Methane fluxes were analyzed according to the four treatments: non-hydric soil with
pulsed hydrology, non-hydric soil with continuously inundated hydrology, hydric soil with pulsed hydrology, and hydric soil with continuously inundated hydrology. Data were analyzed with the General Linear Model (GLM) univariate analysis, with CH4 flux as the dependent variable, and soil type, soil temperature, hydrology, season, time of day and water depth as independent variables. Interactions between the terms were explored, and residuals of the dependent variable were analyzed for normality. Carbon dioxide fluxes were also analyzed with GLM univariate analysis, separately for daytime and nighttime flux rates. Independent variables were the same as those for methane flux, and interaction effects were explored. Relationships between CO2 and CH4 fluxes were 95 compared with linear regression verifying that CH4 residuals were normally distributed.
Significance was defined at α = 0.05.
Soil carbon and nitrogen contents of hydric and non-hydric soils (determined before
establishing hydrology in the mesocosms), bulk densities, soil organic matter and
dissolved carbon in porewater for each soil type were analyzed for normality and
compared with one way ANOVA. Cold and hot water extractable organic matter content
for each treatment were also compared with one-way ANOVA. Time of combustion for
SOM (1 h v. 2 h v. 3 h) was compared separately for hydric and non-hydric soils, using
paired t-tests assuming equal variance. T-tests assuming unequal variance were used to
compare the total amount of water delivered to each continuously inundated and pulsed
mesocosm each month. All statistical analyses assumed a confidence interval of 95% (α
= 0.05) and were performed in SPSS 11 for Mac (SPSS, 2004).
4.4 Results
4.4.1 Hydrology, soil type and methane flux
Water depth in continuously inundated mesocosms averaged 4.0±0.09 cm for the
duration of the experiment, while the water depth in pulsed tubs varied from 5 to –30 cm.
With the exception of April and May, pulsed treatments experienced at least two cycles of drawdown and inundation each month. The mean (± standard error) quantity of water delivered to each mesocosm was approximately 400 ± 80 L month-1, with less water
delivered during months with higher volumes of precipitation. Hydrologic treatment had
a significant effect on methane flux (F=20.055, 736,1, p=0.00), with continuously
96 inundated treatments emitting significantly more methane than pulsed treatments (0.43 ±
0.06 vs. 0.74 ± 0.09 mg m-2 h-1 respectively).
Soil type alone did not significantly affect methane flux rates. However the
interaction between soil type and hydrology had a significant effect on methane fluxes (F
= 5.135, 354,1, p = 0.024, Table 4.1). For mesocosms with non-hydric soils, continuously inundated vs. pulsed hydrology did not make a significant difference (0.32±0.07 vs.
-2 -1 0.17±0.04 mg CH4-C m h respectively), whereas mesocosms with hydric soils emitted
significantly more methane under pulsing conditions than under continuously inundated
-2 -1 conditions (1.25±0.17 vs. 0.66±0.10 mg CH4-C m h respectively, Figure 4.2).
There was more variability in methane flux from pulsed mesocosms with hydric soils
than from continuously inundated mesocosms with hydric soils (Fig. 4.3). The former
exhibited rising and falling rates of CH4 emission that lagged changes in the position of
the water table (either positive or negative) by approximately one week. After 25 days of
inundated conditions from mid May to mid June, methane fluxes from pulsed mesocosms
-2 -1 with hydric soils approached the maximum for this treatment (~2 mg CH4-C m h ).
The highest rates of methane flux from the hydric pulsed treatment occurred after the mesocosms had been flooded continuously for at least 3 weeks. A two-week dry down occurred in mid-June, during which methane emissions from this treatment dropped to <
-2 -1 0.5 mg CH4-C m h . Three shorter intervals (max 2-weeks each) of flooding/dry down
occurred between July 19-mid September. Methane flux from the hydric pulsed
treatment during these shorter periods of inundation was minimal (Fig. 4.3a).
Continuously inundated mesocosms with hydric soils demonstrated a less variable pattern
97 in methane flux than the hydric pulsed treatment, with the highest fluxes (up to 5.3 mg
-2 -1 CH4-C m h ) occurring from late June through early August (Fig. 4.3c).
The hydrologic pattern in pulsed mesocosms with non-hydric soils was roughly
identical to that of pulsed mesocosms with hydric soils, yet methane flux in the non-
hydric treatment showed no apparent relationship to the dynamic hydrology (Fig. 4.3c).
Methane emissions from the non-hydric pulsed treatment continued at the average rate
-2 -1 (0.17 ± 0.05 mg CH4-C m h ) even when the soils experienced a two-week dry down
(Fig. 4.3d). Methane fluxes from the non-hydric continuously inundated treatment were
slightly more variable than those from the non-hydric pulsed treatment (-0.15 to 1.11 mg
-2 -1 -2 -1 CH4-C m h vs. –0.23 to 0.40 mg CH4-C m h ). Higher methane fluxes from the non- hydric continuously inundated treatment corresponded to water levels ≥ 4 cm deep.
Interaction between hydrology and season impacted methane flux significantly (F =
5.239 422, 1, p= 0.023). Pulsed mesocosms emitted approximately the same amount of
methane in spring and summer, while continuously inundated mesocosms emitted
significantly more methane in summer than in spring (1.10 ± 0.16 vs. 0.41 ± 0.07 mg
-2 -1 CH4-C m h respectively). Mean soil temperature over the study period was
significantly warmer (F = 7.346 738,1 p = 0.007) in mesocosms with nonhydric soils than
in mesocosms with hydric soils (23.16 ± 0.24 vs. 22.64 ± 0.25 respectively).
4.4.2 Soil physical properties and methane flux
Munsell color chart analysis of the soils, performed after filling and prior to flooding
the mesocosms, verified hydric and nonhydric characteristics for the two source soils
(Table 4.2). Non-hydric source soils generally had chromas of 3, with a few samples of
98 chroma 2. Minimal redoximorphic features, such as mottles, oxidized pore linings or
redox depletions, were observed in non-hydric soils. The majority of the hydric source
soils had chromas of 1 or 2, but a few samples had chromas of 3 or 4. Hydric soils
contained numerous (>5%) redoximorphic features, mottles being the most frequently
observed.
Bulk densities of each soil type after filling the tubs and prior to inundation were not
significantly different (1.25 ± 0.02 and 1.29 ± 0.02 g cm-3 for hydric and non-hydric soils
respectively). After one year of inundation, bulk densities of hydric and non-hydric soils
respectively were 0.56 ± 0.01 and 0.42 ± 0.01 g cm-3, and the difference between the soil types was significant (F = 73.535 64,1 p<0.01, Table 4.2). Total carbon content of hydric
vs. non-hydric soils was significantly different (1.56±0.06 and 3.68 ± 0.04 %C
respectively, F = 1000.735 19,1 p < 0.01), while total nitrogen content was nearly identical
(~0.16%N, Table 4.2).
Percent SOM (combustion time = 1 h) in hydric and non-hydric soils was comparable
before, and one year after, inundating the mesocosms (4.6 – 4.7 %). Time of combustion
made a significant difference in SOM values for both hydric and non-hydric soils, but
%SOM was nearly identical between the soil types for all combustion times (Fig. 4.4).
While combustion for two hours resulted in higher SOM values than combustion for one hour, the differences were not significant. Combustion for three hours produced significantly higher SOM values for both soil types (p=0.03 and p=0.00 respectively).
Significantly greater quantities of cold and hot water extractable soil organic matter were extracted from hydric soils (Fig. 4.5) compared to non-hydric soils. Cold water extractions produced 0.53 ± 0.02 and 0.47 ± 0.02 g CWEOM per kg dry soil for hydric
99 and non-hydric soils respectively (F = 4.960 48, 1 p=0.031). Hot water extractions
produced 0.86 ± 0.17 and 0.73 ± 0.12 g HWEOM per kg dry soil for hydric and non- hydric soils respectively (F = 9.216 48, 1 p = 0.004). Duplicate extractions produced
experimental results that were not significantly different.
No fizzing was observed in any of the hydric soil samples after addition of hydrochloric acid, indicating that the hydric soils did not contain a substantial quantity of carbonates. To the contrary, vigorous bubbling occurred when HCl was added to each of the non-hydric soil samples, indicating the presence of carbonates. The percentage of inorganic C in non-hydric soils (three hour combustion time) was 1.09±0.04%. Inorganic
C contents of porewaters from non-hydric and hydric soils were not significantly different on either sampling occasion, with respective values of 114 ± 4 vs. 125 ± 4 mg L-
1 on June 2, and 135±12 vs. 136±14 mg L-1 on August 4. Inorganic C content in
groundwater samples taken on the same dates averaged 84.23 ± 1.08 mg L-1.
At the start of gas sampling in April 2005, soil temperatures between 5-10 cm depth
averaged 17-18°C for all treatments. Soils warmed consistently through mid July (27.5 -
29.5°C), after which temperatures began to decline again. There was one aberration in
the warming trend, with mean soil temperatures of 12.3 – 12.7°C on the May 4 sampling
date. At the end of the study in late August, mean soil temperatures were 21.8 - 22.9°C.
Rates of methane emission and methane consumption generally increased with increasing
soil temperature (F = 46.91 656, 1 p = 0.00). Soil temperature explained approximately
30% of methane flux in the hydric, continuously-inundated treatment (F = 65.864 152, 1 p
= 0.00), but was a small to insignificant explanatory variable in the other treatments
(Table 4.3). Soil temperature in all mesocosms combined increased significantly from
100 morning to afternoon to night (19.25 ± 0.43, 22.73 ± 0.24, and 25.05 ± 0.19 respectively,
F = 109.299 737, 2 p = 0.00).
4.4.3 Vegetation and carbon dioxide fluxes in relation to methane flux
Twenty-three plant species were identified in the combined treatments, and a minimum of 16 species was identified in each treatment (Table 4.4). Hydric and non- hydric soils supported similar, diverse plant communities, but none contained more than four dominant species. Percent cover was dominated by obligate wetland species in all treatments, but the non-hydric pulsed treatment contained a FACU clover species
(Trifolium repens L.) as a dominant. Eleocharis spp. (spikerushes) were dominant in all treatments, despite not being planted. Both emergent and submerged (Algae spp.) or floating vegetation (e.g., Potamogeton spp.) was dominant in all treatments except for the non-hydric pulsed mesocosms, which contained only emergent species. Of the three planted species, only Schoenoplectus tabernaemontani and Leersia oryzoides were dominant in any of the treatments.
There were no significant differences in methane flux according to time of day, within treatment types, but the mean fluxes from all treatments combined increased significantly from morning to night (F = 4.872 736,2 p = 0.008). No relationship was observed between rates of methane flux and daytime CO2 uptake for any treatment type, and relationships between methane flux and nighttime CO2 efflux were also not significantly related.
101 4.4.4. Factors influencing carbon dioxide fluxes
Rates of CO2-C uptake were not significantly different between spring (sampled Apr
6 - June 14) and summer (sampled June 21 – Sept 20) within treatments. However, CO2-
C uptake was significantly higher in summer than in spring for the combined pulsed treatments (F = 4.838 200, 1 p = 0.029) compared to the combined continuously inundated
-2 -1 treatments (394 ± 15 vs. 340 ± 30 mg CO2-C m h ) (Table 4.5). Hydrologic treatment
alone had no effect on day or nighttime CO2 flux. CO2-C uptake was explained in part by
soil temperature (F = 82.88 369, 1 p = 0.001), with increasing soil temperatures
corresponding to increasing CO2-C uptake. Significantly more (F = 4.206 415, 1 p = 0.041)
carbon dioxide was taken up during the afternoon than the morning during spring and
-2 -1 summer combined (418 ± 12 vs. 365 ± 20 mg CO2-C m h ).
Over the course of the entire study, the mean rates of CO2-C uptake for each
treatment were 467 ± 18 mg m-2 h-1 in non-hydric continuously inundated mesocosms,
429 ± 19 mg m-2 h-1 in hydric continuously inundated mesocosms, 410±23 mg m-2 h-1 in non-hydric pulsed mesocosms, and 382 ± 16 mg m-2 h-1 in hydric pulsed mesocosms
(Figure 4.7). Mean CO2 uptake rates in pulsed mesocosms with hydric and non hydric
soils were significantly lower than CO2 uptake in continuously inundated mesocosms
with non-hydric soils (F = 5.301 411, 3 p = 0.001). Nighttime CO2 efflux (respiration) was
also significantly higher in summer than in spring for the combined treatments (F = 5.883
183, 1 p=0.016). Nighttime respiration was significantly higher (F = 5.314 182, 1 p = 0.022)
from mesocosms with hydric soils than from mesocosms with non-hydric soils (187 ± 25
-2 -1 vs. 120 ± 23 mg CO2-C m h ), but respiration rates between individual treatments were
not significantly different (Figure 4.7).
102 4.5 Discussion
4.5.1 Methane fluxes in relation to hydrology and hydric vs. non-hydric soils
Both soil type and hydrology were important in determining differences in methane
flux between the four treatments, although soil type was only important in combination
with hydrology and soil temperature. Low rates of methane emissions from non-hydric
-2 -1 treatments (0.17 to 0.32 mg CH4-C m h ) are still within the range reported for
floodplain wetlands in ecosystem studies. Boon et al. (1997) documented fluxes of 0.1 ±
-2 -1 -2 -1 0.02 mg CH4-C m h to 0.7±0.2 mg CH4-C m h from Ryans 5 Billabong, Australia, after two separate flood events. Rates of CH4 flux from mesocosms with hydric soils
-2 -1 (0.66 – 1.25 mg CH4-C m h ) were comparable to or lower than average rates reported in the literature for other floodplain wetlands. In Venezuela’s Orinoco River floodplain,
-2 -1 areas with macrophytes emitted 0.78 ± 0.04 mg CH4-C m h , and forested areas emitted
-2 -1 3.40 ± 0.04 mg CH4-C m h during flooded conditions (Smith et al., 2000). During low
water conditions in the Amazon floodplain, areas containing aquatic macrophytes
-2 -1 produced average methane fluxes of 3.83 ± 0.14 mg CH4-C m h , while during high
-2 -1 water such areas produced 10.13 ± 0.28 mg CH4-C m h (Melack et al., 2004).
Experimental wetlands located at the ORWRP that were inundated for 12 years emitted
-2 -1 3.40 ± 0.47 mg CH4-C m h in intermittently flooded areas, and 7.66 ± 1.78 mg CH4-C m-2 h-1 in continuously inundated areas during a year in which flood pulses were
delivered to the wetlands (Altor and Mitsch, 2006).
Rates of methane flux measured directly in the oxbow wetland varied between 0.28 –
-2 -1 -2 2.63 mg CH4-C m h , (n = 10; average and standard error = 1.15 ± 0.27 mg CH4-C m h-1) which corresponded to the rates observed in mesocosms with hydric soils. This 103 result reassured us that the methane flux rates observed in hydric soil mesocosms were
not an artifact of the mesocosm physical structure, or of the use of groundwater in place
of river water in this experiment. The organic matter content of the mesocosm soils was
comparable to that measured in surface soils from the same site in 1993, before the
experimental wetlands were inundated (Anderson et al., 2005). The relatively low
methane flux rates measured in the mesocosms can be attributed in part to low levels of
organic matter accretion in the soils, a parameter that also reflects, in part, the amount of microbial biomass present in the soils.
Low rates of methane flux from both hydrologic treatments with non-hydric soils is
probably due to a less developed community of methanogens, which could be expected to
develop if the soils were to remain inundated. Although methanotrophy (aerobic,
microbial oxidation of methane) was not directly measured in this study, it is likely that
the intermittently aerated soil conditions present in pulsed treatments facilitated oxidation of methane. A mesocosm experiment conducted by Boon et al. (1997) attributed
methanotrophy to a low availability of organic carbon or to a lack of strongly reducing
conditions.
4.5.2 Importance of dynamic hydrology in relation to methane flux
Establishment of appropriate hydrologic conditions in created or restored wetlands is
critical to achieving optimal functionality. If a wetland being restored or mitigated for
was characterized by pulsed hydrologic conditions, establishing prolonged inundated or
ponded conditions will not likely result in comparable biogeochemical dynamics. With
the central objective of establishing flooded conditions, wetland mitigation and
104 restoration designs often result in less dynamic and more ponded conditions than the
natural wetlands they are designed to replace or restore (Cole et al., 2001). Recent
Supreme Court decisions (04-1034, Rapanos v. United States and 04-1384, Carabell v.
US Army Corps of Engineers) reflect a lack of scientific and technical expertise in
forming judgments about the fate of many of the Nation’s wetlands, favoring wetlands
with a water table that does not fluctuate obviously. A key point of the 2006 majority
Court decision was that “the waters of the United States,” under which wetlands are
legally categorized through the Clean Water Act, “includes only those relatively
permanent, standing or continuously flowing bodies of water…that are described in
ordinary parlance as ‘streams,’ ‘oceans, rivers, [and] lakes’…and does not include
channels through which water flows intermittently or ephemerally” (547 U.S., 2006). We
have shown here that wetlands that do not perennially contain standing water emit less
methane than those that do.
4.5.3 Soil physiochemical properties in relation to methane flux
The percent carbon content of the hydric soils was surprisingly low, equal to the %C
value of the ORWRP surface soils prior to flooding (Nairn, 1996). Despite efforts to
remove only the upper 10 cm of soils from the created oxbow, such precision was
difficult to achieve with the backhoe, and it is possible that some of the underlying soils were removed as well. However, the lack of variation in %C and SOM values for hydric
mesocosms implies consistency in the soils used for the hydric treatments. Hydric soils
were taken from a sparsely vegetated area of the oxbow wetland; organic matter content
can thus be expected to fall on the low end of the continuum. Because soil nitrogen
105 content was similar in hydric and non-hydric soils, hydric soils had a low C/N ratio
compared to non-hydric soils (~10 and 23 respectively). The low C/N ratio in hydric soils implied that N was not limiting for microbial growth and respiration; to the contrary
N can be released into the soil solution when C/N ratios are low, as approximately 8 parts of carbon are needed for every part nitrogen for microbial growth (Brady and Weil,
1999). Because respiration uses about two thirds of the carbon taken up by microorganisms, the C/N ratio in non-hydric soils was relatively ideal for microbial growth and metabolism. The C/N ratio in hydric soils suggests that a substantial portion
of the SOM itself could be composed of microbial biomass. Unless it was broken down
during the extraction process, most of this biomass would not have been accounted for in
the hot and cold water extractions because the filter size (0.45 μm) would have blocked passage of microorganisms, the majority of which are a minimum of 0.5μm diameter
(Dusenberry, 1997; National Research Council, 1999). Piñeiro et al. (2006) have noted that the C/N ratio of whole-soil SOM does not necessarily reflect the C/N ratio of the various SOM pools (e.g. labile, slow and refractory). While total water-soluble extractable organic matter was about 20% greater in hydric than in non-hydric soils, mean methane flux from hydric mesocosms over the course of the study was four times greater than mean methane flux from non-hydric soils. It is reasonable to suspect that a greater microbial biomass, and in particular a more developed methanogen community, is part of the explanation for the lower C/N ratio and higher methane flux in hydric soils compared to non-hydric soils.
4.5.4 Relationship between vegetation, carbon dioxide and methane flux 106 The diurnal variation in methane flux, (flux rates increasing from morning to night), coincided with increasing soil temperatures. Active ventilation of methane by wetland vegetation was not apparent. If active ventilation were occurring, the highest rates of methane flux would be expected in the morning or afternoon (Chanton, 2005), when sunlight-induced pressure differentials between plant aerenchyma and the external atmosphere would trigger the release of gases from plant culms. A complementary ecosystem-scale study conducted at this research site concluded that convective gas flow
through emergent macrophytes was not a dominant mechanism of CH4 release to the atmosphere (Altor and Mitsch, 2006). While this mechanism has been well-described by numerous other researchers for individual plant species and natural wetland ecosystems with standing water, convective flow of methane through vascular plant tissue has not been reported extensively for wetlands with a fluctuating water table.
Hydric and non-hydric soils supported similar, diverse plant communities, but none
had more than four dominant species. While this could be attributed in part to the small
volume:surface area ratio of the mesocosm tubs, full scale wetlands at the same research
site have generally been dominated by four or fewer species, despite being colonized by a
diversity of species (Mitsch et al., 2005). The majority of herbaceous species in each
treatment were FACW or wetter status, indicating that all treatments met the vegetation
requirements for classification as jurisdictional wetlands in the United States. The fact
that only two of the six UPL or FACU species were observed in hydric soil treatments
suggests that either the hydric soil seedbank contained fewer upland plant propagules, or
that upland species had become unviable since the creation of the oxbow wetland in
1997. Not surprisingly, the greatest percentage of herbaceous upland species was found
107 in the non-hydric pulsed treatment, where soils were only inundated during a portion of
the study period. The highest species richness was observed in this treatment as well, but
this greater richness was due entirely to the presence of the six upland species.
A number of researchers have found a positive relationship between primary
productivity and methane emissions (Whiting and Chanton, 1993; Bellisario et al., 1999).
No such relationship was found from chamber measurements made in this study.
Although the CO2 fluxes reported here do not represent comprehensive rates of photosynthesis or respiration, they do illustrate general trends in productivity.
Continuously inundated mesocosms demonstrated greater CO2 uptake than pulsed
treatments, suggesting that the frequent and occasionally prolonged drawdowns of the
water table in pulsed mesocosms may have acted as a stress to the vegetation. While
periodic drying of sediments may inhibit methane flux, primary productivity may also be
impeded. Given the short-term nature of this study, it cannot be concluded that
diminished primary productivity is a predictable result of pulsing hydrologic dynamics.
With time, plant communities adapt to their given hydrologic environment (De Steven
and Toner, 2004).
4.5.5 Methodological issues for soil analyses
Percent loss of inorganic C was derived from the difference between LOI values of
paired acid-treated and non-acid treated soil samples, with acid-treated samples losing
proportionately more weight than non-acid treated soils. Others have reported that metal
oxides can lose structural water at temperatures well below those used for combustion of
organic matter (Heiri et al., 2001). If this were the case, non acid-treated soils should
108 have lost a similar proportion of weight as acid-treated soils, as metal hydroxide (e.g.
Fe(OH)3) content should not differ substantially among paired samples. Additionally, hydric soils contained significantly more available iron and manganese than non-hydric soils, but hydric soils lost less weight during combustion than non-hydric soils. These results suggest that some loss of chloride from CaCl2 may be occurring during the combustion process. Although the melting point of CaCl2 is 772°C, decomposition of this compound has been observed at lower temperatures, especially in the presence of water (Partanen, 2004). Dessicated (combusted) soil will absorb water from the atmosphere between combustion intervals. Due to the observations described above it would be recommendable to maintain samples at 105°C after weighing and before each period of combustion.
Soils were combusted in one-hour intervals for three hours to determine how much change in weight, if any, took place over a three-hour combustion period.
Time/temperature trials conducted by Nairn (1996) demonstrated that one hour of combustion at 550°C was appropriate for soils of newly-created wetlands. Some studies have suggested that a longer combustion time may be appropriate for wetland and lacustrine soils/sediments, which generally contain more organic matter than upland soils
(Updegraff et al., 1995; Heiri et al., 2001). We found the greatest correlation for duplicate samples of both soil types (n=9) combusted for 3 hours (R=0.99, compared with R=0.86 and R=0.90 for 1 and 2 hours respectively), and given the significant difference in SOM according to combustion time, it can be concluded that three hours will provide more reproducible and accurate results than one-hour combustion.
109 Numerous researchers have demonstrated correlations between hot water extractable
carbon, labile carbon and microbial biomass (Sparling et al., 1998; Ghani et al., 2003;
Jinbo et al., 2006). Specific microbial processes, including denitrification potential and
anaerobic mineralizable carbon (Hernandez and Mitsch, 2007) have been positively
correlated with CWEOM. In this study, CWEOM and HWEOM from non hydric soils were 88 and 85% of that from hydric soils, much less than the difference in methane fluxes from non-hydric vs. hydric treatments. WEOM extractions were performed on all
soils after one year of maintaining saturated hydrology and establishing vegetation. It is
possible that the values for WEOM may have been different at the end of the second
growing season and the gas sampling period. The greater availability of iron and
manganese in hydric soils suggests that ferric and manganic oxides were reduced and
released into solution over the years of flooded conditions, making them available for
microbial or plant uptake or metabolism. When the soils periodically became oxidized,
the metals again precipitated. Wetland soils are often a sink for metals due to sediment
accumulation, precipitation of oxidized metals, and the high cation exchange capacity of
clay and organic matter (Dixon, 1997).
4.6 Conclusions
The results of this study suggest that wetlands created on non-hydric soils emit
modest quantities of methane, and methane flux rates will increase with time as hydric
soil characteristics develop. Establishing dynamic hydrology such as is found in natural
floodplain wetlands may help to minimize methane fluxes as a created or restored
wetland ages. Carbon dioxide uptake in wetland ecosystems may be negatively affected
110 by pulsing hydrology if the plant community is adapted to continuously flooded or
saturated conditions. Flood-pulse hydrology is likely to have little effect on methane flux
in the early stages of wetland soil development, but pulsing may facilitate establishment of a diverse plant community that will increase heterogeneity of the system. Created
wetlands located in landscapes that are open to propagule introduction, for example by
river flooding, waterfowl or mammal use, may not need to be seeded or planted with
macrophytes. Planting is costly in terms of money and effort, and in many cases may not
be necessary (see Mitsch et al., 1998). After development of hydric characteristics and
accumulation of available organic matter, pulsed hydrology can reduce methane flux
significantly from created or restored wetlands in comparison with continuously
inundated conditions. Despite lacking a continuous surface water connection to streams
and rivers, wetlands that experience periodic dry cycles perform critical functions in the
landscape, and emit less methane than permanently inundated wetlands. Minimizing
methane flux is one potential objective for created and restored wetlands.
Microtopographical relief, ephemeral connections to rivers and streams, and appropriate
control structures can provide the proper conditions for minimizing methane emissions
from created and restored wetlands while providing deeper water refuge areas for aquatic
species, and increasing spatial and functional diversity.
4.7 Acknowledgements
Funding for this research was provided by USDA NRI CSREES Award 2003-35102- 438
13518, and by the Schiermeier Olentangy River Wetland Research Park. The help of
111 Kyle Chambers in the field and lab is gratefully acknowledged. Chris Holloman provided very valuable consulting on statistical analysis. Olentangy River Wetland
Research Publication Number 07-0xx.
4.8 Literature Cited
547 U. S., 2006. Supreme Court of the United States, Nos. 04 – 1034, John A. Rapanos, et ux., et al., Petitioners v. United States, and 04 – 1384, June Carabell et al., Petitioners v. United States Army Corps of Engineers et al., on writs of certiorari to the United States Court of Appeals for the Sixth Circuit.
Ahearn, D. S., J. H. Viers, J. F. Mount, and R. A. Dahlgren, 2006. Priming the productivity pump: flood pulse driven trends in suspended algal biomass distribution across a restored floodplain. Freshwater Biology 51: 1417-1433.
Ahn, C. and W.J. Mitsch. 2002a. Scaling considerations of mesocosm wetlands in simulating large created freshwater marshes. Ecological Engineering 18: 327-342.
Ahn, C. and W.J. Mitsch. 2002b. Evaluating the use of recycled coal combustion products in constructed wetlands: An ecologic-economic modeling approach. Ecological Modelling 150: 117-140.
Altor, A. E., and W. J. Mitsch, 2006. Methane flux from created riparian marshes: relationship to intermittent vs. continuous inundation and emergent macrophytes. Ecological Engineering 28: 224-234.
Anderson, C.J. and W.J. Mitsch. 2005. Effect of pulsing on macrophyte productivity and nutrient uptake: A wetland experiment. American Midland Naturalist 154: 305-319.
Anderson, C.J., W.J. Mitsch, and R.W. Nairn. 2005. Temporal and spatial development of surface soil conditions at two created riverine marshes. Journal of Environmental Quality 34: 2072-2081.
Bellisario, L. M., J. L. Bubier, T. R. Moore and J.P Chanyon, 1999. Controls on CH4 emissions from a northern peatland. Global Biogeochemical Cycles 13: 81 – 91.
Boon, P. I., A. Mitchell and K. Lee, 1997. Effects of wetting and drying on methane emissions from ephemeral floodplain wetlands in south-eastern Australia. Hydrobiologia 357: 73 – 87.
112 Boon, P. I., and K. Lee, 1997. Methane oxidation in sediments of a floodplain wetland in south-eastern Australia. Letters in Applied Microbiology 25: 138 – 142.
Bouchard, V., and M. Cochran, 2002. Wetland and Carbon Sequestration. Pp. 1416 – 1419 in R. Lal (Editor), Encyclopedia of Soil Science. Marcel Dekker, Inc.
Brady, N. C., and R. R. Weil, 1999. The Nature and Properties of Soils. 12th Edition. Prentice Hall, Upper Saddle River, New Jersey, USA. 881 pp.
Brix, H., B. K. Sorrell and P. T. Orr, 1992. Internal pressurization and convective gas flow in some emergent freshwater macrophytes. Limnology and Oceanography 37: 1420-1433.
Brix, H., B. K. Sorrell and H-H. Schierup, 1996. Gas fluxes achieved by in situ convective flow in Phragmites australis. Aquatic Botany 54: 151 – 163.
Brooks, R. T., and M. Hayashi, 2002. Depth-area-volume and hydroperiod relationships of ephemeral (vernal) forest pools in southern New England. Wetlands 22: 247— 255.
Burke, M. K., B. G. Lockaby, and W. H. Conner, 1999. Aboveground production and nutrient cirulation along a flooding gradient in a South Carolina Coastal Plain forest. Canadian Journal of Forest Research 29: 1402 – 1418.
Calloway, J. C., J. B. Zedler, and D. L. Ross, 1997. Using tidal salt marsh mesocosms to aid wetland restoration. Restoration Ecology 5:135—146.
Carter, V., 1996. Wetland hydrology, water quality, and associated functions. In National Water Summary on Wetland Resources, United States Geological Survey Water Supply Paper 2425. Available on line at http://water.usgs.gov/nwsum/WSP2425/.
Carter, V., P. T. Gammon, and M. K. Garrett, 1994. Ecotone dynamics and boundary determination in the Great Dismal Swamp. Ecological Applications 4: 189—203.
Catallo, J., and T. Junk, 2003. Effects of static vs. tidal hydrology on pollutant transformation in wetland sediments. Journal of Environmental Quality 32: 2421— 2427.
Chantigny, M. H., 2003. Dissolved and water-extractable organic matter in soils: a review on the influence of land use and management practices. Geoderma 113: 357- 380.
Chanton, J. P., 2005. The effect of gas transport on the isotope signature of methane in wetlands. Organic Geochemistry 36: 753 – 768.
113 Cole, C. A., R. P. Brooks, and D. H. Wardrop, 2001. Assessing the relationship between biomass and soil organic matter in created wetlands of central Pennsylvania, USA. Ecological Engineering 17: 423-428.
Craft, C. B., 2001. Biology of wetland soils. Chapter 5 In J. L. Richardson and M. J. Vepraskas (Editors): Wetland Soils: Genesis, Hydrology, Landscape and Classification. CRC Press, Boca Raton, FL, U.S.A.
De Steven, D., and M. M. Toner, 2004. Vegetation of upper coastal plain depression wetlands: Environmental templates and wetland dynamics within a landscape framework. Wetlands 24: 23-42.
Dixon, K. L., 1997. Background concentrations of metals in wetland soils on and near the Savannah River site. USDOE Technical report number WSRC-MS--97-00692- Rev.1, Washington, DC.
Dusenberry, D. B., 1997. Minimum size limit for useful locomotion by free-swimming microbes. Proceedings of the National Academy of Sciences 94: 10949 – 10954.
Elrashidi, M. A., M. D. Mays, and C. W. Lee, 2003. Assessment of Mehlich3 and ammonium bicarbonate-DTPA extraction for simultaneous measurement of fifteen elements in soils. Communications in Soil Science and Plant Analysis 34: 2817- 2838.
Ghani, A., M. Dexter, and K. W. Perrott, 2003. Hot-water extractable carbon in soils: a sensitive measurement for determining impacts of fertilization, grazing and cultivation. Soil Biology & Biochemistry 35: 1231 – 1243.
Grosse, W., 1996. Pressurised ventilation in floating-leaved aquatic macrophytes. Aquatic Botany 54: 137–150.
Heiri, O., A. F. Lotter, and G. Lemcke, 2001. Loss on ignition as a method for estimating organic and carbonate content in sediments: reproducibility and comparability of results. Journal of Paleolimnology 24: 101 – 110.
Hernandez, M.E. and W.J. Mitsch. 2007. Denitrification potential and organic matter as affected by vegetation community, wetland age, and plant introduction in created wetlands. Journal of Environmental Quality 36: 333-342.
Hong, P. K. A., C. Li, S. K. Banerji, and Y. Wang, 2002. Feasibility of metal recovery from soil using DTPA and its biostability. Journal of Hazardous Materials B94: 253 – 272.
114 Jinbo, Z., S. Changchun, and YU. Wenyan, 2006. Land use effects on the distribution of labile organic carbon fractions through soil profiles. Soil Science Society of America Journal 70: 660 – 667. Johnson Randall, L. A., and A. L. Foote, 2005. Effects of managed impoundments and herbivory on wetland plant production and stand structure. Wetlands 25: 38—50.
Klinger, L. F., P. R. Zimmerman, J. P. Greenberg, L. E. Heidt, and A. B. Guenther, 1994. Carbon trace gas fluxes along a successional gradient in the Hudson Bay lowland. Journal of Geophysical Research 99 (D1): 1469—1494.
Marschner, B., and K. Kalbitz, 2003. Controls of bioavailability and biodegradability of dissolved organic matter in soils. Geoderma 113: 211–235.
Megonigal, J.P., W.H. Conner, S. Kroeger and R.R. Sharitz. 1997. Aboveground production in southeastern floodplain forests: a test of the subsidy-stress hypothesis. Ecology 78: 370-384.
Melack, J. M., L. L. Hess, M. Gastil, B. R. Forsberg, S. K. Hamilton, I. B. T. Lima, and E. M. L. M. Novo, 2004. Regionalization of methane emissions in the Amazon Basin with microwave remote sensing. Global Change Biology 10: 530 – 544.
Mitsch, W.J. 1988. Productivity - hydrology - nutrient models of forested wetlands. Pp. 115-132 in W.J. Mitsch, M. Straskraba and S.E. Jørgensen (Editors), Wetland Modelling. Elsevier, Amsterdam.
Mitsch, W.J. and X. Wu. 1995. Wetlands and Global Change. Pp. 205-230 in R. Lal , J. Kimble, E. Levine, and B.A. Stewart (Editors.), Advances in Soil Science, Soil Management and Greenhouse Effect. Boca Raton, Florida: Lewis Publishers.
Mitsch, W.J., X. Wu, R.W. Nairn, P.E. Weihe, N. Wang, R. Deal, C.E. Boucher. 1998. Creating and restoring wetlands: A whole-ecosystem experiment in self-design. BioScience 48: 1019-1030.
Mitsch, W.J., L. Zhang, C.J. Anderson, A. Altor, and M. Hernandez. 2005. Creating riverine wetlands: Ecological succession, nutrient retention, and pulsing effects. Ecological Engineering 25: 510-527.
Nairn, R., 1996. "Biogeochemistry of newly created riparian wetlands: evaluation of water quality changes and soil development " Ph.D. dissertation, Environmental Science Graduate Program, The Ohio State University.
National Research Council, 1995. Wetlands: Characteristics and Boundaries. National Academy Press, Washington, DC.
115 National Research Council, 1999. Size limits of very small microorganisms: proceedings of a workshop. National Academy of Sciences. Space Studies Board, National Research Council, 2101 Constitution Avenue, NW, Washington, DC.
Odum, E. P., J. T. Finn, and E. H. Franz, 1979. Perturbation theory and the subsidy- stress gradient. Bioscience 29: 349 – 352.
Partanen, J., 2004. Chemistry of HCl and limestone in fluidized bed combustion. Ph.D. dissertation, Inorganic Chemistry, The Åbo Akademi Process Chemistry Centre, Finland.
Piñeiro, G., M. Oesterheld, W. B. Batista, and J. M. Paruelo, 2006. Opposite changes of whole-soil vs. pools C : N ratios: a case of Simpson’s paradox with implications on nitrogen cycling. Global Change Biology 12: 804 – 809.
Reed, P. B. Jr., 1988. National list of plant species that occur in wetlands: Northeast (Region 1). U.S. Fish and Wildlife Service Biological Report 88 (26.1). 111 pp.
Schütz, H., Schröder, P., Rennenberg, H., 1991. Role of plants in regulating the methane flux to the atmosphere. Pp. 29-63 in Sharkey, T. D., Holland, E. A., and Mooney, H. A., (Editors), Trace Gas Emissions by Plants. Academic Press, San Diego, California.
SCS, 1980. Soil Survey of Franklin County, Ohio. U.S. Department of Agraculture, Soil Conservation Service. Washington, DC, 122 pp.
Smith, L. K., W. M. Lewis Jr., J. P. Chanton, G. Cronin and S. K. Hamilton, 2000. Methane emissions from the Orinoco River floodplain, Venezuela. Biogeochemistry 51: 113-140.
Sommer, R., B. Harrell, M. Nobriga, R. Brown, P. Moyle, W. Kimmerer, and L. Schemel, 2001. California’s Yolo Bypass: evidence that flood control can be compatible with fisheries, wetlands, wildlife, and agriculture. Fisheries 26: 6 – 16.
Sparling, G., M. Vojvodic-Vukovic, and L. A. Schipper, 1998. Hot-water soluble C as a simple measure of labile soil organic matter: the relationship with microbial biomass C. Soil Biology and Biochemistry 30: 1469-1472.
SPSS, 2004. Version 11.0.3. SPSS, Inc. Chicago, Illinois.
Svengsouk, L. and W.J. Mitsch. 2001. Dynamics of mixtures of Typha latifolia and Schoenoplectus tabernaemontani in nutrient-enrichment wetland experiments. American Midland Naturalist 145: 187-210.
116 Updegraff, K., J. Pastor, S. D. Bridgham, and C. A. Johnston, 1995. Environmental and substrate controls over carbon and nitrogen mineralization in northern wetlands. Ecological Applications 5: 151 – 163.
USACOE, 1987. Corps of Engineers Wetlands Delineation Manual. Technical Report Y-87-1, U. S. Army Corps of Engineers Waterways Experiment Station, Vicksburg, MS.
USDA, 1999. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. Second Edition, Soil Survey Staff, United States Department of Agriculture, Natural Resources Conservation Service, Agriculture Handbook Number 236. 871 pp.
USDA, 2004. Soil Survey Laboratory Methods Manual. Soil Survey Investigations Report No. 42, Version 4.0, Burt, R., Editor. United States Department of Agriculture Natural Resources Conservation Service. van der Valk, A., and R. L. Pederson, 2003. The SWANCC decision and its implications for prairie potholes. Wetlands 23: 590—596.
Vepraskas, M. J., and S. P. Faulkner, 2001. Redox chemistry of hydric soils. Chapter 4 In J. L. Richardson and M. J. Vepraskas (Editors): Wetland Soils: Genesis, Hydrology, Landscape and Classification. CRC Press, Boca Raton, FL, U.S.A.
Whalen, S. C., 2005. Biogeochemistry of methane exchange between natural wetlands and the atmosphere. Environmental Engineering Science 22: 73 – 94.
Whiting, G. J. and J. P. Chanton, 1993. Primary production control of methane emission from wetlands. Nature 364: 794 – 795.
Whiting, G. J. and J. P. Chanton, 1996. Control of the diurnal pattern of methane emission from emergent aquatic macrophytes by gas transport mechanisms. Aquatic Botany 54: 237—253.
Zsolnay, A., 2003. Dissolved organic matter: artifacts, definitions and functions. Geoderma 113: 178-210.
117 a.
b.
Figure 4.1. Experimental mesocosm design, illustrating a) dimensions, French drains, stand pipes for adjusting water level, and outflow opening; and b) gas sampling chamber design (illustration 4.1.b. by Jung-Chen Huang).
118 1.60 b 1.40
1.20 -1 h
-2 1.00 c 0.80 -C m 4 0.60 a
mg CH 0.40 a 0.20
0.00 Non-hydric Non-hydric Hydric Hydric Pulsed Continuously Pulsed Continuously Inundated Inundated
Figure 4.2. Mean methane emissions over the entire study period for each treatment. Number of samples is as follows: non-hydric continuously inundated: 214; non-hydric pulsed: 171; hydric continuously inundated: 176; hydric pulsed: 180. Bars represent standard error.
119 10 2.5 a. 8 6 b. 5 2 7 ) 5 ) 0 -1 -1 h h 1.5 -2
6 -2 -5 4 -C m 4 -C m
1 4 -10 5 3 -15 0.5 4 -20 2 0 3 Water Level (cm) Level Water Water Level (cm) -25 1 -0.5 2 -30 Methane Flux (mg CH Methane Flux (mg CH -1 1 0 -35 -40 -1.5 0 -1
10 2.5 8 6 d. c. ) ) -1 5 2 7 -1 h 5 h -2 -2 0 1.5 6 4 -C m -C m 4 4 -5 1 5 3 -10 0.5 4 2 -15 0 3
Water Level (cm) Water Level 1 Water Level (cm) -20 -0.5 2 0 -25 -1 1 Methane Flux (mg CH (mg Methane Flux Methane Flux (mg CH Methane Flux (mg -30 -1.5 0 -1 1-Jul 1-Apr 3-Jun 6-Aug 2-Sep 10-Jul 19-Jul 28-Jul 7-May 1-Jul 10-Apr 19-Apr 28-Apr 12-Jun 21-Jun 1-Apr 3-Jun 15-Aug 24-Aug 11-Sep 6-Aug 2-Sep 16-May 25-May 10-Jul 19-Jul 28-Jul 7-May 10-Apr 19-Apr 28-Apr 12-Jun 21-Jun 15-Aug 24-Aug 11-Sep 16-May 25-May Water Level Avg CH4-C Flux Water Level Mean Methane Flux
Figure 4.3. Mean daily water level, and mean diurnal methane flux on each sampling date for each treatment: a: Hydric pulsed; b: Hydric continuously inundated; c: Non- hydric pulsed; d: Non-hydric continuously inundated.
120 5.20 5.10 b b ab 5.00 4.90 ab 4.80 a a 4.70 4.60
%SOM 4.50 4.40 4.30 4.20 4.10 NON-HYDRIC HYDRIC
1 HR 2 HR 3 HR
Figure 4.4. Percent soil organic matter (SOM) according to number of hours combusted at 550°C. Different letters indicate a significant difference in SOM between combustion times. Bars represent standard error.
1.00 Hydric Soil 0.90 Non-Hydric Soil
0.80
0.70
0.60
0.50
0.40
g WEOM kg dry soil-1 0.30
0.20
0.10
0.00 CWEOM HWEOM
Figure 4.5. Cold-water extractable organic matter (CWEOM) and hot-water extractable organic matter (HWEOM) for hydric and non-hydric soils, measured one year after mesocosms were inundated. Different letters indicate a significant (p ≤ 0.05) difference among treatments. Bars represent standard error.
121 Mean CO2 Uptake Mean CO2 Efflux 600 a 500 ab b 400 b
300 -1
h 200 -2 100 -C m 2 0
-100 mg CO
-200 a -300 a a -400 a -500 Non-hydric, Hydric, Non-hydric, Hydric, Pulsed Continuously Continuously Pulsed inundated inundated
Figure 4.6. Mean rates of daytime CO2-C uptake and nighttime efflux for each treatment over the entire gas-sampling period (April – September 2005). Different letters indicate a significant difference between treatments for either uptake or efflux.
Parameter F-value p-value Corrected Model 16.992 0.00 Hydrologic treatment x Season 5.239 0.023 Soil type x Soil temperature 7.346 0.007 Soil type x Hydrologic treatment 5.135 0.024 Hydrologic treatment 20.055 0.00 Soil temperature 38.803 0.00 Time of day 4.872 0.008 Season 0.945 0.332 Soil type 0.927 0.336
Table 4.1 Univariate general linear model for methane flux from mesocosms, indicating significant terms and interactions.
122 Munsell Redoximorphic Db‡ %SOM %C %N profile* features Hydric soil 10YR 4/3 – Mottles >5% 2004: 2004: 1.56±0.06 0.15±0.01 treatments 10YR 3/1 1.25±0.02 4.77±0.12 82% with 5YR 4/5 – 10YR 2005: 2005: chroma ≤ 2 5/8 0.56±0.01 4.60±0.14 Non-hydric 10YR 4/3 – Mottles ≤ 2% 2004: 2004: 3.68±0.04 0.16±0.00 soil 10YR 3/2 1.29±0.02 4.65±0.08 treatments 91% with 10YR 6/7 – 10YR 2005: 2005: chroma ≥ 3 6/8 0.42±0.01 4.62±0.07 * Hue (e.g. 10YR), value (#/), chroma (/#) ‡ Bulk density (± standard error) after soil settled, before mesocosms were flooded (2004), and one year after flooding (2005)
Table 4.2. Physical and chemical characteristics of hydric and non-hydric soils.
REGRESSION ANOVA Treatment r2 F-value p-value df Hydric, Continuously inundated 0.304 65.864 0.00 152 Hydric, Pulsed 0.043 7.065 0.009 157 Non hydric, Continuously inundated 0.061 12.577 0.00 194
Table 4.3. Significant results from regressions between soil temperature and methane flux from mesocosm treatments, and analysis of variance results for each regression.
123 Hydric Non-hydric Hydric Non-hydric Pulsed Pulsed Continuously Continuously inundated inundated Emergent wetland species *Schoenoplectus tabernaemontani X x x X K.C. Gmel (OBL)
*Sparganium eurycarpum Engelm. x x x x (OBL) Eleocharis spp. (OBL) X X X X *Leersia oryzoides L. Sw. (OBL) X x x x Typha latifolia L. (OBL) x X x x
Carex spp. (FACW) x x x Eupatorium perfoliatum L. x (FACW+) Lycopus americanus L. (OBL) x x x x Mimulus ringens L. (OBL) x x x x Asclepius incarnata L. (OBL) x Verbena hastata L. (FACW+) x x Submerged and floating species Algae spp. (OBL) x x X X Ludwigia palustris (L.) Elliott (OBL) x x x x Potamogeton spp. (OBL) X x X x Herbaceous upland species Solidago sp. x Echinochloa crusgalli L. Beauv. x (FACU) Plantago major L. (FACU) x x x x Taraxacum officinale Weber ex. x x x Wiggers (FACU) Hibiscus trionum L. (UPL) x x Trifolium repens L. (FACU-) X x Table 4.4 continued on next page
Table 4.4. Plant species identified in each treatment (wetland indicator status in parentheses‡) during the gas sampling period. An asterisk indicates the species was planted in all mesocosms; capitalized, bold X indicates the species was dominant in a given treatment.
124 Table 4.4. (continued)
Hydric Non-hydric Hydric Non-hydric Pulsed Pulsed Continuously Continuously inundated inundated Woody species Acer negundo L. (FAC+) x Salix nigra Marshall (FACW+) x x x x Populus deltoides Bertram ex x x x x Marshall (FAC) Acer rubrum L. (FAC) x x x x
‡ OBL: Probability of occurring in wetlands = >99%; FACW: Probability of occurring in wetlands = 67 to 99%; FAC: Probability of occurring in wetlands = 34 to 66%; FACU: Probability of occurring in wetlands = 1 to 33%; UPL: Probability of occurring in wetlands = <1% (Reed, 1988). A + or – indicates the plant falls at the upper or lower end, respectively, of the probability range.
Parameter F-value p-value Corrected Model 15.583 0.00 Hydrologic treatment x Season 4.838 0.029 Soil temperature x Water depth 11.429 0.001 Hydrologic treatment 0.045 0.832 Time of day 4.206 0.041 Season 14.931 0.00 Soil temperature 82.88 0.00 Water depth 16.845 0.00
Table 4.5. Univariate general linear model for daytime carbon dioxide uptake in mesocosms indicating significant terms and interactions.
125
Parameter F-value p-value Corrected Model 7.535 0.00 Season x Soil temperature 7.178 0.008 Soil type 5.314 0.022 Season 5.883 0.016 Soil temperature 3.066 0.082
Table 4.6 Univariate general linear model for nighttime carbon dioxide efflux from mesocosms, indicating significant terms and interactions.
126
CHAPTER 5
CONCLUSIONS
Some general conclusions and practical applications have been derived from this disseration research. Results include implications for the impact of flood pulses on carbon cycling in created and restored wetlands and suggestions for refining methodologies in wetland biogeochemical studies.
5.1 Hydrology and methane fluxes in experimental wetlands and mesocosms
• Hydrologic pulsing corresponded to significantly lower methane fluxes from
continuously inundated wetland areas, compared with steady flow hydrology.
The difference in methane flux between the years can be explained in part by
slightly higher average soil temperatures and a greater range in water depths in
continuously inundated areas during the steady flow year.
• Shallower water levels were also associated with higher rates of methane flux
during the pulsing year, presumably because of less opportunity for methane
oxidation and more solar radiation reaching the sediments.
127 • During the pulsing year, methane fluxes from intermittently inundated wetland
areas with emergent macrophytes were significantly lower when soils were not
saturated or submerged, compared to when they were inundated. The average
methane flux from edge zones during the pulsed year was not significantly
different from methane flux during the steady flow year, because soils were
inundated/saturated more than they were exposed during the growing season of
the pulsed year.
• Prolonged drawdowns have a significant dampening effect on wetland methane
fluxes, probably due to methanotrophy and suppression of methanogenesis by
oxygenated conditions in the sediment. However, methane emissions resume
upon reflooding. In areas without emergent vegetation, drawdown of the water
table can initially lead to increased methane fluxes by reducing hydrostatic
pressure on sediments and allowing ebullition or more rapid diffusion across the
sediment/air interface.
• Methane flux rates in wetland edge zones without emergent vegetation did not
differ significantly within the pulsing year when soils were exposed vs. inundated
or saturated. More prolonged exposure may result in diminished methane fluxes
from areas without emergent vegetation. The water table can be expected to drop
more slowly in areas without emergent plants during the growing season because
macrophyte transpiration draws water up from deeper in the soil profile than
surface evaporation.
128
• Methane fluxes from continuously inundated wetland zones were significantly
higher than fluxes from edge zones under both pulsing and steady flow
conditions. Persistent anaerobic conditions in continuously inundated areas
results in maintenance of a low redox potential conducive to methanogenesis.
• The stage of soil development (hydric vs. non-hydric) is more important to the
magnitude of methane flux than hydrologic treatment in newly established
wetlands. Hydrology becomes important once hydric characteristics and labile
organic matter content have developed in soils.
5.2 Hydrology and carbon dioxide fluxes in experimental wetlands and mesocosms
• Hydrologic pulsing affected carbon dioxide fluxes mainly through interaction
with stage of soil development. At the mescosocm scale, hydrologic treatment did
not make a significant difference to rates of photosynthesis except between the
two most contrasting treatments: steady flow mesocosms with non-hydric soils vs.
pulsed mesocosms with hydric soils. Steady flow conditions were conducive to
higher rates of photosynthesis in mesocosms with non-hydric soils compared to
pulsed treatments with hydric soils. Hydric substrates that develop under
conditions of prolonged inundation support vegetation that is adapted to thrive in
oxygen poor sediments, and such macrophytes are not necessarily more
productive when the soil in which they grow is periodically drained.
129 5.3 Vegetation and diurnal methane fluxes in experimental marshes and mesocosms
• Under conditions of both steady flow and flood-pulsed hydrologic conditions,
mean growing season methane fluxes from wetland edge zones with emergent
vegetation and edge zones in which emergent vegetation was removed were
nearly identical, and there were no differences in methane flux according to time
of day. A lack of diurnal difference in methane flux in any of the mesocosm
treatments reinforces the interpretation that emergent vegetation was not
exhibiting pressurized ventilation, but rather acted as a conduit for diffusion of
methane.
• Positive relationships between emergent vegetation species and methane flux
observed in wetlands with a continuously positive water table may not hold in
wetlands in which sediments are periodically exposed during the growing season.
5.4 Methodological issues for wetland studies
• The non-steady-state chamber design utilized for this dissertation research is
relatively novel and can be recommended for use in experiments that include
multiple sampling sites, sites containing vegetation, and frequent sampling.
While the chamber bases and frames could be left in place, the bags that formed
the enclosures were portable. Two designs were used in the various studies: one
in which bags were pulled down over the tops of the chamber frames and secured
around the bases. This design (described in Chapters 2 and 3) is preferable if
water levels fluctuate substantially and may rise up close to the top of the
130 chamber base. If water levels rise and fall within a narrow range, it is
recommendable to use the second design (described in Chapter 4), where the bags
remain attached to the chamber base. The advantage of this design is that rolling
bags up around vegetation is much simpler and less destructive than pulling them
down around the plants. In addition, it is more time effective. The bags must
remain out of the water.
• Methods of analysis for wetland soils vary to the extent that it is sometimes
difficult to compare results from different researchers’ projects. A key example is
the range of temperature and combustion times reported in the literature for
organic matter determination. It was found during the course of this dissertation
research that the time of combustion makes a significant difference in quantity of
organic matter reported for both hydric and non-hydric soils. It is arguable that
one hour of combustion at 550°C is not sufficient for determining soil organic
matter content. For purposes of comparison with other studies, combustion time
and temperature should be consistent. Soils should be combusted for additional
hours to ensure all organic matter is accounted for. Continued experimentation
with combustion time should be carried out; individual comparative studies such
as this one can help to inform standardized methods for wetland soil analyses.
5.5. Suggestions for future research
• A useful future study would be the pairing of steady-state and non-steady-state
chambers for sampling of CO2 fluxes, to compare flux rate results. While all CO2
131 fluxes used in the analyses presented here were linear, higher CO2 flux rates were
associated with shorter sampling times. Experimenting with even shorter
sampling times under steady state chamber conditions would provide additional
information on methodology for capturing CO2 fluxes in site and sampling
intensive studies.
• The pulsing regime utilized in the ecosystem-scale experiment was conducive to
scheduling sampling events, and simulated high flow conditions that occur in
spring in the Midwest. A future pulsing study could time hydrologic pulses to
precipitation and runoff events, to even further capture sediment and nutrient
loads from runoff in the watershed.
• The reasons behind the lower methane emission rates from wetland edge zones
compared to continuously inundated zones should be further elucidated. In order
to do so, multiple-year, paired comparisons could be made between edge zones
maintained under continuously inundated conditions, edge zones in which the
water table fluctuates above and below the soil, and continuously inundated
zones. By including areas with and without emergent macrophytes, and soil
organic matter analysis, further understanding about the interactions between
hydrology, vegetation, soils and carbon fluxes could be attained.
• An additional means of determining what role emergent macrophytes play in
methane flux from the wetlands would be to measure gas fluxes from small areas
132 enclosing macrophytes, and then immediately cut the macrophyte stems below the
water surface, replace the chamber and measure the flux rates again. Doing so
would reveal how much of the methane is being delivered via diffusion (or
ventilation) through plant tissue, and how much is escaping directly from the
sediment and water column (Jeff Chanton, personal communication).
• It would be worthwhile to examine how much methane is actually retained within
the chamber structure by adhering to chamber materials, thereby obtaining an
estimate of percent recovery of methane during sampling. To do so, one could
use a chamber with a sealed base in the laboratory. After injecting a known
concentration of methane into the chamber, chamber air samples should be
removed at time intervals mimicking those used in the field. Accounting for
background concentrations, the percent recovered vs. percent expected could be
used to estimate error in field sampling measurements (Warren Dick, personal
communication).
133
BIBLIOGRAPHY
547 U. S., 2006. Supreme Court of the United States, Nos. 04 – 1034, John A. Rapanos, et ux., et al., Petitioners v. United States, and 04 – 1384, June Carabell et al., Petitioners v. United States Army Corps of Engineers et al., on writs of certiorari to the United States Court of Appeals for the Sixth Circuit.
Ahearn, D. S., J. H. Viers, J. F. Mount, and R. A. Dahlgren, 2006. Priming the productivity pump: flood pulse driven trends in suspended algal biomass distribution across a restored floodplain. Freshwater Biology 51: 1417-1433.
Ahn, C. and W.J. Mitsch. 2002a. Scaling considerations of mesocosm wetlands in simulating large created freshwater marshes. Ecological Engineering 18: 327-342.
Ahn, C. and W.J. Mitsch. 2002b. Evaluating the use of recycled coal combustion products in constructed wetlands: An ecologic-economic modeling approach. Ecological Modelling 150: 117-140.
Altor, A. E., and W. J. Mitsch, 2006. Methane flux from created riparian marshes: relationship to intermittent vs. continuous inundation and emergent macrophytes. Ecological Engineering 28: 224-234.
Anderson, C. J., Mitsch, W.J., Nairn, R. J., 2005. Temporal and spatial development of surface soil conditions at two created riverine marshes. Journal of Environmental Quality 34: 2072-2081.
Anderson, C.J. and W.J. Mitsch. 2006. Sediment, carbon, and nutrient accumulation at two 10-year-old created riverine marshes. Wetlands 26: 779-792.
Anderson, K.L., Wheeler, K. A., Robinson, J. A., Tuovinen, O. H., 2002. Atrazine mineralization potential in two wetlands. Water Research 36: 4785-4794.
Bartlett, K. B., Harris, R. C., 1993. Review and assessment of methane emissions from wetlands. Chemosphere 26: 261-320.
Bayley, P. B., 1995. Understanding large river-floodplain ecosystems. Bioscience 45: 153-158.
134 Bellisario, L. M., Bubier, J. L., Moore, T. R., Chanton, J. P., 1999. Controls on CH4 emissions from a northern peatland. Global Biogeochemical Cycles 13: 81-91.
Boon, P. I., and K. Lee, 1997. Methane oxidation in sediments of a floodplain wetland in south-eastern Australia. Letters in Applied Microbiology 25: 138-142.
Boon, P. I., A. Mitchell and K. Lee, 1997. Effects of wetting and drying on methane emissions from ephemeral floodplain wetlands in south-eastern Australia. Hydrobiologia 357: 73-87.
Bouchard, V., and M. Cochran, 2002. Wetland and Carbon Sequestration. Pp. 1416- 1419 in R. Lal (Editor), Encyclopedia of Soil Science. Marcel Dekker, Inc.
Brady, N. C., and R. R. Weil, 1999. The Nature and Properties of Soils. 12th Edition. Prentice Hall, Upper Saddle River, New Jersey, USA. 881 pp.
Bridgham, S. C., J. P.. Megonigal, J. K. Keller, N. B. Bliss, and C. Trettin, 2006. The carbon balance of North American wetlands. Wetlands 26: 889-916.
Brix, H., Sorrell, B. K., Orr, P. T., 1992. Internal pressurization and convective gas flow in some emergent freshwater macrophytes. Limnology and Oceanography 37: 1420- 1433.
Brix, H., B. K. Sorrell and H-H. Schierup, 1996. Gas fluxes achieved by in situ convective flow in Phragmites australis. Aquatic Botany 54: 151-163.
Brooks, R. T., and M. Hayashi, 2002. Depth-area-volume and hydroperiod relationships of ephemeral (vernal) forest pools in southern New England. Wetlands 22: 247-255.
Brooks, R. P., D. H. Wardrop, C. A. Cole, and D. A. Campbell, 2005. Are we purveyors of wetland homogeneity? A model of degradation and restoration to improve wetland mitigation performance. Ecological Engineering 24: 331-340.
Bunce, J. A., 2005. Response of respiration of soybean leaves grown at ambient and elevated carbon dioxide concentrations to day-to-day variation in light and temperature under field conditions. Annals of Botany 95: 1059-1066.
Burke, M. K., B. G. Lockaby, and W. H. Conner, 1999. Aboveground production and nutrient cirulation along a flooding gradient in a South Carolina Coastal Plain forest. Canadian Journal of Forest Research 29: 1402-1418.
Calloway, J. C., J. B. Zedler, and D. L. Ross, 1997. Using tidal salt marsh mesocosms to aid wetland restoration. Restoration Ecology 5:135-146.
135 Cao, M., Gregson, K., Marshall, S., 1998. Global methane emission from wetlands and its sensitivity to climate change. Atmospheric Environment 32: 3293-3299.
Carter, V., 1996. Wetland hydrology, water quality, and associated functions. In National Water Summary on Wetland Resources, United States Geological Survey Water Supply Paper 2425. Available on line at http://water.usgs.gov/nwsum/WSP2425/.
Carter, V., P. T. Gammon, and M. K. Garrett, 1994. Ecotone dynamics and boundary determination in the Great Dismal Swamp. Ecological Applications 4: 189-203.
Catallo, J., and T. Junk, 2003. Effects of static vs. tidal hydrology on pollutant transformation in wetland sediments. Journal of Environmental Quality 32: 2421- 2427.
Chantigny, M. H., 2003. Dissolved and water-extractable organic matter in soils: a review on the influence of land use and management practices. Geoderma 113: 357- 380.
Chanton, J. P. and J. W. H. Dacey, 1991. Effects of vegetation on methane flux, reservoirs, and carbon isotopic composition. Pp 29-63 In: T. D. Sharkey, E. A. Holland, and H. A. Mooney (Editors), Trace Gas Emissions by Plants. Academic Press, Inc. San Diego.
Christensen, T. R., N. Panikov, M. Mastepanov, A. Joabsson, A. Stewart, M. Öquist, M. Sommerkorn, S. Reynaud, and B. Svensson, 2003. Biotic controls on CO2 and CH4 exchange in wetlands – a closed environment study. Biogeochemistry 64: 337-354.
Cole, C. A., R. P. Brooks, and D. H. Wardrop, 2001. Assessing the relationship between biomass and soil organic matter in created wetlands of central Pennsylvania, USA. Ecological Engineering 17: 423-428.
Conrad, R., A. Bollmann, H. Yao and R. Roy, 1999. Effect of water management on soil microbial communities and atmospheric trace gases. Microbial Biosystems: New Frontiers. Proceedings of the 8th International Symposium on Microbial Ecology. C. R. Bell, M. Brylinsky, and P. Johnson-Green, Editors. Atlantic Canada Society for Microbial Ecology, Halifax, Canada.
Conrad, R., 2002. Control of microbial methane production in wetland rice fields. Nutrient Cycling in Agroecosystems 64: 59-69.
Craft, C. B., 2001. Biology of wetland soils. Chapter 5 In J. L. Richardson and M. J. Vepraskas (Editors): Wetland Soils: Genesis, Hydrology, Landscape and Classification. CRC Press, Boca Raton, FL, U.S.A.
136 D’Angelo, E. M., and K. R. Reddy, 1999. Regulators of heterotrophic microbial potentials in wetland soils. Soil Biology and Biochemistry 31: 815-830.
Dalva, M., Moore, T. R., Arp, P., Clair, T. A., 2001. Methane and soil and plant community respiration from wetlands, Kejimkujik National Park, Nova Scotia: Measurements, predictions, and climatic change. Journal of Geophysical Research 106 (D3): 2955-2962.
Devol, A. H., Richey, J. E., Clark, W. A., King, S. L., 1988. Methane emissions to the troposphere from the Amazon floodplain. J. Geophys. Res. 93 (D2): 1583-1592.
Dixon, K. L., 1997. Background concentrations of metals in wetland soils on and near the Savannah River site. USDOE Technical report number WSRC-MS--97-00692- Rev.1, Washington, DC.
Dusenberry, D. B., 1997. Minimum size limit for useful locomotion by free-swimming microbes. Proceedings of the National Academy of Sciences 94: 10949-10954.
Elrashidi, M. A., M. D. Mays, and C. W. Lee, 2003. Assessment of Mehlich3 and ammonium bicarbonate-DTPA extraction for simultaneous measurement of fifteen elements in soils. Communications in Soil Science and Plant Analysis 34: 2817- 2838.
Fennessy, M.S., and W. J. Mitsch, 2001. Effects of hydrology on spatial patterns of soil development in created riparian wetlands. Wetlands Ecology and Management 9: 103-120.
Freeman, C., G. B. Nevison, H. Kang, S. Hughes, B. Reynolds, and J. A. Hudson, 2002. Contrasted effects of simulated drought on the production and oxidation of methane in a mid-Wales wetland. Soil Biology and Biochemistry 34: 61-67.
Galat, D. L., L. H. Fredrickson, D. D. Humburg, K. J. Bataille, J. R. Bodie, J. Dohrenwend, G. T. Gelwicks, J. E. Havel, D. L. Helmers, J. B. Hooker, J. R. Jones, M. F. Knowlton, J. Kubisiak, J. Mazourek, A. C. McColpin, R. B. Renken, and R. D. Semlitsch, 1998. Flooding to restore connectivity of regulated, large-river wetlands: natural and controlled flooding as complementary processes along the lower Missouri River. Bioscience 48: 721-733.
Ghani, A., M. Dexter, and K. W. Perrott, 2003. Hot-water extractable carbon in soils: a sensitive measurement for determining impacts of fertilization, grazing and cultivation. Soil Biology & Biochemistry 35: 1231-1243.
Gilvear, D. J., and C. Bradley, 2000. Hydrological monitoring and surveillance for wetland conservation and management; a UK perspective. Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere 25: 571-588.
137
Grosse, W., 1996. Pressurised ventilation in floating-leaved aquatic macrophytes. Aquatic Botany 54: 137-150.
Harris, R. C., Sebacher, D. I., 1982. Methane flux in the Great Dismal Swamp. Nature 297: 673-674.
Harter, S. K., Mitsch, W. J., 2003. Patterns of short-term sedimentation in a freshwater created marsh. Journal of Environmental Quality 32: 325-354.
Healy, R. W., Striegl, R. G., Russell, T. F., Hutchinson, G. L., Livingston, G. P., 1996. Numerical evaluation of static-chamber measurements of soil-atmosphere gas exchange: identification of physical processes Soil Science Society of America Journal 60: 740-747.
Heiri, O., A. F. Lotter, and G. Lemcke, 2001. Loss on ignition as a method for estimating organic and carbonate content in sediments: reproducibility and comparability of results. Journal of Paleolimnology 24: 101-110.
Hernandez, M. E. H., Mitsch, W. J., 2006. Influence of hydrologic pulses and vegetation on nitrous oxide emissions from created riparian marshes. Wetlands 26: 862-877.
Hernandez, M.E. and W.J. Mitsch. 2007. Denitrification potential and organic matter as affected by vegetation community, wetland age, and plant introduction in created wetlands. Journal of Environmental Quality 36: 333-342.
Hirota, M., Tang, Y., Hu, Q., Hirata, S., Kato, T., Mo, W., Cao, G., and Mariko, S., 2004. Methane emissions from different vegetation zones in a Qinghai-Tibetal Plateau wetland. Soil Biology and Biochemistry 36: 737-748.
Holland, E. A., Robertson, G. P., Greenberg, J., Groffman, P. M., Boone, R. D., Gosz, J. R., 1999. Soil CO2, N2O, and CH4 Exchange. In Robertson, G. P., Coleman, D. C., Bledsoe, C. S. (Eds.), Standard Soil Methods for Long Term Ecological Research. Oxford University Press, New York, New York, pp. 185-201.
Hong, P. K. A., C. Li, S. K. Banerji, and Y. Wang, 2002. Feasibility of metal recovery from soil using DTPA and its biostability. Journal of Hazardous Materials B94: 253- 272.
Hunt, R.J., Krabbenhoft, D.P., and Anderson, M.P., 1997, Assessing hydrogeochemical heterogeneity in natural and constructed wetlands. Biogeochemistry 39: 271–293.
IPCC, 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Houghton, J. T., Ding, Y., Griggs, D. J., Noguer, M., van der Linden, P. J., Dai, S.,
138 Maskell, K., Johnson, C. A. (Eds.), Cambridge University Press, Cambridge, UK and New York, NY, 881 pp.
Jia, Z. J., Cai, Z. C., Xu, H., Li, X. P., 2001. Effect of rice plants on CH4 production, transport, oxidation and emission in ride paddy soil. Plant and Soil 230: 211–221.
Jinbo, Z., S. Changchun, and YU. Wenyan, 2006. Land use effects on the distribution of labile organic carbon fractions through soil profiles. Soil Science Society of America Journal 70: 660-667.
Joabsson, A., and T. R. Christensen, 2001. Wetlands and methane emission. Pp. 1429- 1432 in: R. Lal, (Editor), Encyclopedia of Soil Science. Marcel Dekker, Inc. New York, USA.
Johnson Randall, L. A., and A. L. Foote, 2005. Effects of managed impoundments and herbivory on wetland plant production and stand structure. Wetlands 25: 38-50.
Junk, W. J., Bayley, P. B., Sparks, R. E., 1989. The flood pulse concept in river- floodplain systems. In D. P. Dodge, (Ed.), Proceedings of the International Large River Symposium. Can. Spec. Publ. Fish. Aquat. Sci. 106: pp. 110-127.
Juutinen, S., Alm, J., Martikainen, P., 2001. Effects of spring flood and water level draw-down on methane dynamics in the littoral zone of boreal lakes. Freshwater Biology 46: 855-869.
Juutinen, S., 2004. Methane fluxes and their environmental controls in the littoral zone of boreal lakes. University of Joensuu, PhD Dissertations in Biology, number 25. ISSN 1457-2486. 110 pp.
Käki, T., A. Ojala and P. Kankaala, 2001. Diel variation in methane emissions from stands of Phragmites australis (Cav.) Trin. Ex Steud. and Typha latifolia L. in a boreal lake. Aquatic Botany 71: 259-271.
Kang, H., Freeman, C., Lee, D., Mitsch, W. J., 1998. Enzyme activities in constructed wetlands: implication for water quality amelioration. Hydrobiologia 368: 231-235.
Kim, J., S. B. Verma, and D. P. Billesbach, 1998. Seasonal variation in methane emission from a temperate Phragmites-dominated marsh: effect of growth stage and plant-mediated transport. Global Change Biology 5: 433-440.
Klinger, L. F., Zimmerman, P. R., Greenberg, J. P., Heidt, L. E., Guenther, A. B., 1994. Carbon trace gas fluxes along a successional gradient in the Hudson Bay lowland. Journal of Geophysical Research 99 (D1): 1469-1494.
139 Koreny, J.S., Mitsch, W. J., Bair, E. S., Wu, X., 1999. Regional and local hydrology of a created riparian wetland system. Wetlands 19: 182-193.
Kutzbach, L., D. Wagner, and E-M. Pfeiffer, 2004. Effect of microrelief and vegetation on methane emission from wet polygonal tundra, Lena Delta, Northern Siberia. Biogeochemistry 69: 341-362.
Lal, R., and M. K. Shukla, 2004. Principles of Soil Physics. Marcel Dekker, Inc., New York, NY. 716 pp.
Le Mer, J., Roger, P., 2001. Production, oxidation, emission and consumption of methane by soils: a review. European Journal of Soil Biology 37: 25-50.
Livingston, G. P., Hutchinson, G. L., 1995. Enclosure-based measurements of trace gas exchange: applications and sources of error. In P.A. Matson Harris, R. C., (Eds.), Biogenic Trace Gases: Measuring Emissions from Soil and Water. Blackwell Science, Oxford, pp. 14-51.
Marschner, B., and K. Kalbitz, 2003. Controls of bioavailability and biodegradability of dissolved organic matter in soils. Geoderma 113: 211–235.
McKee, K. L., and P. L. Faulkner, 2000. Restoration of biogeochemical function in mangrove forests. Restoration Ecology 8: 247-259.
Megonigal, J.P., W.H. Conner, S. Kroeger and R.R. Sharitz. 1997. Aboveground production in southeastern floodplain forests: a test of the subsidy-stress hypothesis. Ecology 78: 370-384. Megonigal, J.P., M.E. Hines, and P.T. Visscher, 2004. Anaerobic Metabolism: Linkages to Trace Gases and Aerobic Processes. Pages 317-424 in Schlesinger, W.H. (Editor). Biogeochemistry. Elsevier-Pergamon, Oxford, UK.
Melack, J. M., L. L. Hess, M. Gastil, B. R. Forsberg, S. K. Hamilton, I. B. T. Lima, and E. M. L. M. Novo, 2004. Regionalization of methane emissions in the Amazon Basin with microwave remote sensing. Global Change Biology 10: 530-544.
Metzker, K., Mitsch, W. J., 1997. Modelling self-design of the aquatic community in a newly created freshwater wetland. Ecological Modelling 100: 61-86.
Middleton, B., 1999. Wetland Restoration, Flood Pulsing, and Disturbance Dynamics. John Wiley and Sons, New York.
Mitra, S., R. Wassmann, and P. L. G. Vlek, 2005. An appraisal of global wetland area and its organic carbon stock. Current Science 88: 25-35.
140 Mitsch, W.J. 1988. Productivity - hydrology - nutrient models of forested wetlands. Pp. 115-132 in W.J. Mitsch, M. Straskraba and S.E. Jørgensen (Editors), Wetland Modelling. Elsevier, Amsterdam.
Mitsch, W. J., Ewel, K. C., 1979. Comparative biomass and growth of cypress in Florida wetlands. American Midland Naturalist 101: 417-426.
Mitsch, W. J., and W. G. Rust, 1984. Tree growth responses to flooding in a bottomland forest in northeastern Illinois. Forest Science 30: 499-510.
Mitsch, W.J. and X. Wu. 1995. Wetlands and Global Change. Pp. 205-230 in R. Lal , J. Kimble, E. Levine, and B.A. Stewart (Editors.), Advances in Soil Science, Soil Management and Greenhouse Effect. Boca Raton, Florida: Lewis Publishers.
Mitsch, W. J., Wu, X., Nairn, R. W., Weihe, P. E., Wang, M., Deal, R., Boucher, C. E., 1998. Creating and restoring wetlands: a whole-ecosystem experiment in self-design. Bioscience 48: 1019-1030.
Mitsch, W. J., and J. G. Gosselink, 2000. Wetlands, 3rd Edition. John Wiley & Sons, Inc, New York.
Mitsch, W. J., Day, Jr., J. W., Gilliam, J. W., Groffman, P. M., Hey, D. L., Randall, G. W., Wang, N., 2001. Reducing nitrogen loading to the Gulf of Mexico from the Mississippi River basin: strategies to counter a persistent problem. Bioscience 51: 373-388. Mitsch, W. J., Day, J. W., Jr. 2004. Thinking big with whole ecosystem studies and ecosystem restoration—a legacy of H.T. Odum. Ecological Modelling 178: 133-155.
Mitsch, W. J., and S. E. Jorgensen, 2004. Ecological Engineering and Ecosystem Restoration. John Wiley & Sons, Inc. Hoboken, NJ. 411 pp.
Mitsch, W. J., Day, Jr., J. W., Zhang, L., Lane, R. R., 2005a. Nitrate-nitrogen retention in wetlands in the Mississippi River Basin. Ecological Engineering 24: 267-278.
Mitsch, W.J., Zhang, L., Anderson, C. J., Altor, A., Hernandez, M., 2005b. Creating riverine wetlands: Ecological succession, nutrient retention, and pulsing effects. Ecological Engineering 25: 510-527.
Mitsch, W.J. and J.W. Day, Jr. 2006. Restoration of wetlands in the Mississippi-Ohio- Missouri (MOM) River Basin: Experience and needed research. Ecological Engineering 26: 55-69.
Mitsch, W.J., L. Zhang, M.E. Hernandez, A.E. Altor, A.M. Nahlik, C.L. Tuttle, D.F. Fink, and C. J. Anderson, 2006. Importance of hydrologic pulsing on the water
141 quality function of wetlands in Midwestern USA. Abstracts, American Water Resources Association Annual Meeting, Baltimore, MD.
Molles, M. C. Jr., Crawford, C. S., Ellis, L. M., Valett, H. M., Dahm, C. N., 1998. Managed flooding for riparian ecosystem restoration: managed flooding reorganizes riparian forest ecosystems along the middle Rio Grande in New Mexico. Bioscience 48: 749-756.
Montalto, F. A., and T. S. Steenhuis, 2004. The link between hydrology and restoration of tidal marshes in the New York/New Jersey estuary. Wetlands 24: 414-425.
Moore, T. R., and J. Turunen, 2004. Carbon accumulation and storage in mineral subsoil beneath peat. Soil Science Society of America Journal 68: 690-696.
Nahlik, A.M. and W.J. Mitsch. In review. The effect of river pulsing on sedimentation in created riparian wetlands. Journal of Environmental Quality.
Nairn, R., 1996. "Biogeochemistry of newly created riparian wetlands: evaluation of water quality changes and soil development " Ph.D. dissertation, Environmental Science Graduate Program, The Ohio State University.
Nairn, R.W., Mitsch, W. J., 2000. Phosphorus removal in created wetland ponds receiving river overflow. Ecological Engineering 14: 107-126.
National Research Council, 1995. Wetlands: Characteristics and Boundaries. National Academy Press, Washington, DC. National Research Council, 1999. Size limits of very small microorganisms: proceedings of a workshop. National Academy of Sciences. Space Studies Board, National Research Council, 2101 Constitution Avenue, NW, Washington, DC.
Odum, E. P., J. T. Finn, and E. H. Franz, 1979. Perturbation theory and the subsidy- stress gradient. Bioscience 29: 349-352.
Odum, W. E., Odum, E. P., Odum, H. T., 1995. Nature’s pulsing paradigm. Estuaries 18: 547-555.
Owen Koning, C., 2005. Vegetation patterns resulting from spatial and temporal variability in hydrology, soils, and trampling in an isolated basin marsh, New Hampshire, USA. Wetlands 25: 239-251.
Partanen, J., 2004. Chemistry of HCl and limestone in fluidized bed combustion. Ph.D. dissertation, Inorganic Chemistry, The Åbo Akademi Process Chemistry Centre, Finland.
142 Piñeiro, G., M. Oesterheld, W. B. Batista, and J. M. Paruelo, 2006. Opposite changes of whole-soil vs. pools C : N ratios: a case of Simpson’s paradox with implications on nitrogen cycling. Global Change Biology 12: 804-809.
Pulliam, W. M., 1993. Carbon dioxide and methane exports from a southeastern floodplain swamp. Ecological Monographs 63: 29-53.
Reed, P. B. Jr., 1988. National list of plant species that occur in wetlands: Northeast (Region 1). U.S. Fish and Wildlife Service Biological Report 88 (26.1). 111 pp.
Roulet, N. T., 2000. Peatlands, carbon storage, greenhouse gases, and the Kyoto Protocol: Prospects and significance for Canada. Wetlands 20: 605-615.
Sass, R. L., Fisher, F. M., Wang, Y. B., 1992. Methane emission from rice fields: the effect of floodwater management. Global Biogeochemical Cycles 6: 249-262.
Schipper, L. A., Reddy, K. R., 1994. Methane production and emission from four reclaimed and pristine wetlands of southeastern United States. Soil Science Society of America Journal 58: 1270-1275.
Schlesinger, W. H., 1997. Biogeochemistry, An Analysis of Global Change. 2nd Ed. Academic Press, San Diego, CA. 588 pp.
Schütz, H., Schröder, P., Rennenberg, H., 1991. Role of plants in regulating the methane flux to the atmosphere. Pp. 29-63 in Sharkey, T. D., Holland, E. A., and Mooney, H. A., (Editors), Trace Gas Emissions by Plants. Academic Press, San Diego, California.
SCS, 1980. Soil Survey of Franklin County, Ohio. U.S. Department of Agraculture, Soil Conservation Service. Washington, DC, 122 pp.
Segers, R.,1998. Methane production and methane consumption: a review of processes underlying wetland methane fluxes. Biogeochemistry 41: 23-51.
Selbo, S.M., Snow, A. A., 2004. The potential for hydridization between Typha angustifolia and T. latafolia in a constructed wetland. Aquatic Botany 78: 361-369.
Shannon, R. D., White, J. R., 1994. A three-year study of controls on methane emissions from two Michigan peatlands. Biogeochemistry 27: 35-60.
Smith, L. K., Lewis, W. M. Jr., Chanton, J. P., Cronin, G., and Hamilton, S. K., 2000. Methane emissions from the Orinoco River floodplain, Venezuela. Biogeochemistry 51: 113-140.
143 Sommer, R., B. Harrell, M. Nobriga, R. Brown, P. Moyle, W. Kimmerer, and L. Schemel, 2001. California’s Yolo Bypass: evidence that flood control can be compatible with fisheries, wetlands, wildlife, and agriculture. Fisheries 26: 6-16.
Sparling, G., M. Vojvodic-Vukovic, and L. A. Schipper, 1998. Hot-water soluble C as a simple measure of labile soil organic matter: the relationship with microbial biomass C. Soil Biology and Biochemistry 30: 1469-1472.
Spieles, D.J., Mitsch, W. J., 2000. The effects of season and hydrologic and chemical loading on nitrate retention in constructed wetlands: A comparison of low and high nutrient riverine systems. Ecological Engineering 14: 77-91.
Spink, A., R. E. Sparks, M. van Oorschot, and J. T. A. Verhoeven, 1998. Nutrient dynamics of large river floodplains. Regulated Rivers: Research and Management 14: 203-216. SPSS, 2004. Version 11.0.3. SPSS, Inc. Chicago, Illinois.
Svengsouk, L. and W.J. Mitsch. 2001. Dynamics of mixtures of Typha latifolia and Schoenoplectus tabernaemontani in nutrient-enrichment wetland experiments. American Midland Naturalist 145: 187-210.
Tanner, C. C., Adams, D. D., Downes, M. T., 1997. Methane emissions from constructed wetlands treating agricultural wastewaters. Journal of Environmental Quality 26: 1056-1062.
Tiner, R. W., 2005. Assessing cumulative loss of wetland functions in the Nanticoke River watershed using enhanced National Wetlands Inventory data. Wetlands 25: 405-419.
Tockner, K., Stanford, J. A., 2002. Riverine flood plains: present state and future trends. Environmental Conservation 29: 308-330.
Tuittila, E. S., Komulainen, V. M., Vasander, H., Nykänen, H., Martikainen, P. J., Laine, K., 2000. Methane dynamics of a restored cut-away peatland. Global Change Biology 6: 569-581.
Tuttle, C.L. and W.J. Mitsch. In revision. Aquatic metabolism as an indicator of the ecological effects of hydrologic pulsing in flow-through wetlands. Ecological Indicators.
Updegraff, K., J. Pastor, S. D. Bridgham, and C. A. Johnston, 1995. Environmental and substrate controls over carbon and nitrogen mineralization in northern wetlands. Ecological Applications 5: 151-163.
144 USACOE, 1987. Corps of Engineers Wetlands Delineation Manual. Technical Report Y-87-1, U. S. Army Corps of Engineers Waterways Experiment Station, Vicksburg, MS.
USDA, 1999. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. Second Edition, Soil Survey Staff, United States Department of Agriculture, Natural Resources Conservation Service, Agriculture Handbook Number 236. 871 pp.
Vaithiyanathan, P., and C. J. Richardson, 1997. Nutrient profiles in the everglades: examination along the eutrophication gradient. The Science of the Total Environment 205: 81-95.
Van der Nat, F. J. W. A., Middelburg, J. J., 2000. Methane emissions from tidal freshwater marshes. Biogeochemistry 49: 103-121.
van der Valk, A., and R. L. Pederson, 2003. The SWANCC decision and its implications for prairie potholes. Wetlands 23: 590-596.
Vepraskas, M. J., and S. P. Faulkner, 2001. Redox chemistry of hydric soils. Chapter 4 In J. L. Richardson and M. J. Vepraskas (Editors): Wetland Soils: Genesis, Hydrology, Landscape and Classification. CRC Press, Boca Raton, FL, U.S.A.
Vretare Strand, V., 2002. The influence of ventilation systems on water depth penetration of emergent macrophytes. Freshwater Biology 47: 1097-1005.
Waddington, J. M., Roulet, R. T., and Swanson, R. V., 1996. Water table control of CH4 emission enhancement by vascular plants in boreal peatlands. Journal of Geophysical Research 101 (D17): 22775-22785.
Walter, B. P., M. Heimann, R. D. Shannon and J. R. White, 1996. A process-based model to derive methane emissions from natural wetlands. Geophysical Research Letters 23: 3731-3734.
Whalen, S. C., 2005. Biogeochemistry of methane exchange between natural wetlands and the atmosphere. Environmental Engineering Science 22: 73-94.
Whalen, S. C., Reeburg, W. S., 2000. Methane oxidation, production, and emission at contrasting sites in a Boreal bog. Geomicrobiology Journal 17: 237-251.
Whiting, G. J. Chanton, J. P., 1993. Primary production control of methane emission from wetlands. Nature 364: 794-785.
145 Whiting, G. J. and J. P. Chanton, 1996. Control of the diurnal pattern of methane emission from emergent aquatic macrophytes by gas transport mechanisms. Aquatic Botany 54: 237-253.
Whiting, G. J., Chanton, J. P., 2001. Greenhouse carbon balance of wetlands: methane emission versus carbon sequestration. Tellus 53B: 521-528.
Wu, X., Mitsch, W. J., 1998. Spatial and temporal patterns of algae in newly constructed freshwater wetlands. Wetlands 18: 9-20.
Yavitt, J. B., and A. K. Knapp, 1998. Aspects of methane flow from sediment through emergent cattail (Typha latifolia) plants. New Phytologist 139: 495-503.
Zedler, J. B., and J. C. Callaway, 1999. Tracking wetland restoration: Do mitigation sites follow desired trajectories? Restoration Ecology 7: 69-73.
Zsolnay, A., 2003. Dissolved organic matter: artifacts, definitions and functions. Geoderma 113: 178-210.
146
APPENDIX A
CH4 AND CO2 FLUX DATA, 2004
147 Apr. May May June June July Aug. Aug. Aug. Sept. Sept. 25-26 9-10 23-24 10-11 29-30 7-8 4-5 10 24-25 4-5 28-29 Wetland 1 a.m.: 0.00 0.11 0.00 0.75 3.37 XX 1.42 0.00 0.07 3.22 Inflow with noon: 0.00 0.32 0.99 0.66 1.98 3.48 n.l 0.00 0.10 0.18 0.00 macrophytes night: 0.00 0.04 0.27 0.81 0.30 5.40 1.08 0.00 0.13 0.00 0.00 Wetland 1 a.m.: 0.04 0.00 4.10 6.73 12.60 14.42 2.26 0.51 5.19 0.00 Inflow noon: 0.41 0.29 6.14 8.64 10.67 14.16 2.16 1.74 0.52 13.22 0.70 marsh zone night: 0.10 0.24 3.73 7.26 11.31 17.32 1.45 0.58 0.54 5.15 0.20 Wetland 1 a.m.: 0.06 0.19 2.28 13.36 12.47 15.25 0.37 -0.15 0.62 1.24 Outflow with noon: 0.00 0.52 6.39 7.87 19.69 12.35 0.52 0.00 0.46 0.00 0.00 macrophytes night: 0.11 0.25 4.87 7.39 14.79 n.l 0.45 0.11 0.41 0.39 1.84 Wetland 2 a.m.: 0.80 1.99 15.41 0.80 11.53 14.94 2.11 1.76 1.66 0.00 Inflow with noon: 1.01 2.27 14.21 n.l 8.81 14.24 2.94 0.41 1.45 0.00 macrophytes night: 1.25 1.45 34.36 1.47 10.53 16.10 2.69 0.31 1.45 2.14 1.04 Wetland 2 a.m.: 1.19 0.00 2.07 0.70 7.11 6.45 4.05 0.76 2.48 0.60 Inflow noon: 0.35 0.07 3.17 0.00 11.68 7.29 3.53 1.34 0.90 2.39 marsh zone night: 2.76 0.31 2.41 2.35 4.00 9.01 4.78 0.45 0.99 2.55 1.20 Wetland 2 a.m.: 0.89 0.33 1.75 n.l 5.12 2.32 0.81 0.46 0.35 0.87 Outflow with noon: 0.62 0.26 6.19 6.06 4.70 10.59 0.70 3.90 0.53 0.32 0.00 macrophytes night: 0.68 0.00 3.03 1.10 3.62 8.38 0.85 2.46 0.48 0.46 0.59 * n.l. = flux rate was non-linear (R2 < 0.90) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.1. Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2004.
148 Feb. Mar. Mar. Apr. Oct. Nov. Dec. 28 3-4 24-25 8-9 26 23-24 16 Wetland 1 a.m.: 0.00 0.00 0.00 0.00 Inflow with noon: 0.00 0.00 -0.25 -0.06 1.27 0.00 macrophytes night: 0.00 -0.06 0.00 0.00 Wetland 1 a.m.: 0.00 0.00 0.00 0.00 Inflow noon: 0.00 0.00 0.03 0.32 0.00 0.00 marsh zone night: 0.00 0.35 0.00 0.00 Wetland 1 a.m.: 0.00 0.00 0.00 0.00 Outflow with noon: 0.00 0.00 0.00 0.00 0.00 0.00 macrophytes night: n.l 0.08 0.00 0.00 Wetland 2 a.m.: 0.00 0.00 0.00 0.00 0.00 0.00 Inflow with noon: 0.00 0.00 0.00 0.00 n.l macrophytes night: 0.00 0.00 0.06 n.l Wetland 2 a.m.: 0.00 0.00 0.06 0.20 0.00 0.00 Inflow noon: 0.00 0.00 0.00 0.00 0.00 marsh zone night: 0.00 0.08 0.06 0.00 0.00 Wetland 2 a.m.: 0.00 0.00 0.06 0.00 0.00 0.00 Outflow with noon: 0.00 -0.05 0.05 0.00 0.00 macrophytes night: 0.00 0.05 0.09 0.00 0.00 * n.l. = flux rate was non-linear (R2 < 0.90) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.2. Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2004.
149 Apr. May May June June July Aug. Aug. Aug. Sept. Sept. 25-26 9-10 23-24 10-11 29-30 7-8 4-5 10 24-25 4-5 28-29 Wetland 1 a.m.: 0.02 0.00 0.10 0.12 0.13 n.l. 0.20 0.00 0.23 0.70 Inflow without noon: 0.00 0.18 9.69 0.00 0.31 1.30 0.06 0.03 0.02 0.32 0.00 macrophytes night: 0.03 0.06 0.20 0.24 0.67 0.85 0.18 0.00 0.04 0.00 0.00 Wetland 1 a.m.: 0.07 0.21 1.81 n.l. 1.46 2.73 0.11 0.00 0.00 0.00 Outflow without noon: 0.04 0.25 8.80 0.24 0.02 4.70 0.02 0.25 0.04 0.35 1.50 macrophytes night: 0.09 0.13 3.63 1.87n.l. 7.66 0.00 0.23 0.04 0.00 -1.44 Wetland 2 a.m.: 1.23 0.66 18.02 n.l. 0.60 4.58 4.26 0.59 6.05 n.l. Inflow without noon: 0.72 0.62 0.14 n.l. 13.95 0.16 0.12 0.69 n.l. macrophytes night: 1.95 1.21 5.53 1.90 10.75 54.14 2.41 0.32 0.74 2.09 0.44 Wetland 2 a.m.: 1.03 0.00 8.77 8.35 16.07 23.95 0.06 0.04 0.24 0.65 Outflow without noon: 3.51 n.l. 33.42 8.27 n.l. 71.16 0.08 1.85 0.04 1.52 0.22 macrophytes night: 0.49 0.04 23.71 2.07 4.06 12.20 0.05 1.44 0.05 0.66 * n.l. = flux rate was non-linear (R2 < 0.90) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.3. Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the growing season, 2004.
Feb. Mar. Mar. Apr. Oct. Nov. Dec. 28 3-4 24-25 8-9 26 23-24 16 Wetland 1 a.m.: 0.00 0.00 0.02 0.03 Inflow without noon: 0.00 0.00 0.12 0.07 0.00 0.40 macrophytes night: 0.03 0.00 0.00 -0.27 Wetland 1 a.m.: 0.00 0.00 0.00 -0.10 Outflow without noon: 0.00 0.00 0.00 0.00 0.00 0.00 macrophytes night: 0.00 0.00 n.l. 0.40 Wetland 2 a.m.: 0.00 -0.02 0.00 0.00 0.00 0.00 Inflow without noon: 0.00 0.03 0.00 0.04 0.00 macrophytes night: 0.00 0.03 0.05 0.00 Wetland 2 a.m.: 0.00 0.00 0.04 n.l. 0.00 0.00 Outflow without noon: 0.00 0.00 0.10 0.00 0.00 macrophytes night: 0.00 0.16 0.30 0.00 0.00 * n.l. = flux rate was non-linear (R2 < 0.90) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.4. Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2004.
150 Apr. May May June June July Aug. Aug. Aug. Sept. Sept. 25-26 9-10 23-24 10-11 29-30 7-8 4-5 10 24-25 4-5 28-29 Wetland 1 a.m.: 0.87 0.00 7.05 1.38 n.l. 0.17 2.65 0.68 0.77 9.51 Outflow noon: 7.73 0.11 40.51 n.l. 1.77 2.27 5.86 0.91 0.31 3.88 marsh zone night: 1.97 0.10 3.32 0.00 6.04 14.57 7.23 0.56 0.89 1.17 0.98 Wetland 2 a.m.: 0.39 3.46 5.12 1.80 3.91 2.86 3.09 3.02 6.80 11.88 Outflow noon: 1.34 6.31 98.06 2.57 3.47 30.33 4.51 3.29 2.28 18.50 25.90 marsh zone night: n.l. 1.44 3.89 1.73 7.53 5.91 14.31 8.73 4.03 3.40 Wetland 1 a.m.: 2.82 1.41 8.77 2.27 Inflow C.I. noon: 5.16 67.61 0.93 6.18 n.l. night: 8.15 17.28 1.02 7.79 2.00 Wetland 1 a.m.: 1.34 4.67 4.63 0.41 Outflow C.I. noon: 5.37 20.37 18.10 12.72 0.66 night: 5.07 5.06 10.62 2.62 1.11 Wetland 2 a.m.: 4.57 5.13 7.94 8.77 Inflow C.I. noon: 7.02 30.23 9.10 1.01 night: 2.08 9.66 5.98 8.91 3.87 Wetland 2 a.m.: 1.18 21.72 8.90 1.65 Outflow C.I. noon: 0.74 6.72 20.54 11.75 6.18 night: 2.41 0.97 n.l. 14.80 - n.l. = flux rate was non-linear (R2 < 0.90) - C.I. = continuously inundated zone - red type indicates the water level was below the soil surface in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.5. Methane fluxes (mg CH4-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the growing season, 2004. Fluxes in red on August 10 were not used in the analysis.
151
Feb. Mar. Mar. Apr. Oct. Nov. Dec. 28 3-4 24-25 8-9 26 23-24 16 Wetland 1 a.m.: 0.00 0.00 0.00 0.01
Outflow noon: 0.00 0.00 0.00 0.06 -0.36 1.17 marsh zone night: 0.00 0.15 n.l. 0.00
Wetland 2 a.m.: 0.00 0.00 0.05 0.11 n.l. 0.00 Outflow noon: 0.00 0.00 0.00 0.23 3.05
marsh zone night: 0.00 0.09 0.00 0.89 1.37 Wetland 1 a.m.: Inflow C.I. noon: 2.76 1.26
night: n.l. 0.00 Wetland 1 a.m.:
Outflow C.I. noon: 0.00 n.l. night: 0.67 0.00
Wetland 2 a.m.: 13.61 0.00 Inflow C.I. noon: 4.25 night: 3.66 20.54
Wetland 2 a.m.: 7.32 0.40 Outflow C.I. noon: 23.35
night: 6.32 18.79 - n.l. = flux rate was non-linear (R2 < 0.90) - C.I. = continuously inundated zone - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.6. Methane fluxes (mg CH4-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the non-growing season, 2004.
152
Apr. May May June June July Aug. Aug. Aug. Sept. Sept. 25-26 9-10 23-24 10-11 29-30 7-8 4-5 10 24-25 4-5 28-29 Wetland 1 a.m.: 18 n.l. -448 n.l. -494 653 94 n.l. 292 -208 Inflow with noon: -325 -255 n.l. -862 -465 -1450 -550 -429 -843 -291 macrophytes night: 71 94 395 139 19 182 291 418 508 667 248 Wetland 1 a.m.: -606 n.l. n.l. -140 -221 n.l. -785 143 Inflow noon: 109 -302 -747 -876 n.l. -460 -687 n.l. n.l. -664 -204 marsh zone night: 118 135 319 291 302 292 262 553 478 409 n.l. Wetland 1 a.m.: -90 -365 -485 n.l. -452 -681 n.l. n.l. -421 -104 Outflow with noon: -255 -381 -257 n.l. -548 -620 -452 n.l. -211 -1192 -349 macrophytes night: 98 92 354 132 204 n.l. 255 328 249 274 208 Wetland 2 a.m.: -259 -280 -78 0 n.l. n.l. n.l. 113 -194 0 Inflow with noon: -124 -257 -498 -556 45 -26 -284 n.l. n.l. macrophytes night: 122 74 444 178 351 348 293 590 419 527 429 Wetland 2 a.m.: 1 -55 -383 n.l. -681 -647 n.l. n.l. -1383 -303 Inflow noon: 0 -244 n.l. -593 -664 -334 -494 -302 143 marsh zone night: 24 22 59 n.l. 90 187 294 440 357 455 n.l. Wetland 2 a.m.: 48 -20 -271 n.l. -174 -334 -787 -868 0 Outflow with noon: n.l. -203 n.l. n.l. -315 -186 -265 -1086 -469 macrophytes night: 77 n.l. 109 n.l. 106 120 180 340 226 285 277 - n.l. = flux rate was non-linear (R2 < 0.88) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.7. Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time.
153
Feb. Mar. Mar. Apr. Oct. Nov. Dec. 28 3-4 24-25 8-9 26 23-24 16 Wetland 1 a.m.: n.l. -109 -244 Inflow with noon: 0 n.l. -63.3 n.l. 0 macrophytes night: -35 -19 0 65 Wetland 1 a.m.: 0 n.l. -40.5 Inflow noon: 0 -294 17 59 48 marsh zone night: 52 12 n.l. Wetland 1 a.m.: 17 25 12 Outflow with noon: 16 43 -111 -163 0 macrophytes night: n.l. 0 -135 0 Wetland 2 a.m.: -71.7 -59.2 9 0 n.l. Inflow with noon: 55.14 65.54 0 -679 macrophytes night: 19 n.l. -27 8 Wetland 2 a.m.: -124 0 36 0 72 Inflow noon: -34.2 -105 -57.1 -552 marsh zone night: -13 9 n.l. 172 0 Wetland 2 a.m.: 19 1 n.l. -13.5 144 Outflow with noon: 24 29 -32.2 -294 macrophytes night: 49 39 n.l. 41 0 - n.l. = flux rate was non-linear (R2 < 0.88) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.8. Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time.
154 Apr. May May June June July Aug. Aug. Aug. Sept. Sept. 25-26 9-10 23-24 10-11 29-30 7-8 4-5 10 24-25 4-5 28-29 Wetland 1 a.m.: 115 24 11 11 23 14 21 16 33 20 Inflow without noon: 227 -32 213 -20 -49 -32 -105 94 -103 0 36 macrophytes night: 190 13 233 21 16 39 17 145 42 43 161 Wetland 1 a.m.: n.l. -68 33 8 -49 16 n.l. 27 n.l. 75 Outflow without noon: -105 -70 26 n.l. n.l. 12 -70 150 26 53 70 macrophytes night: 72 22 136 15 24 61 36 161 49 58 -103 Wetland 2 a.m.: 14 13 -61 n.l. -47 15 n.l. 38 103 111 Inflow without noon: -82 2 n.l. -57 20 n.l. 202 n.l. 104 macrophytes night: 56 30 90 18 48 87 90 273 105 99 60 Wetland 2 a.m.: 10 n.l. -84 n.l. 12 -186 -74 n.l. -84 121 Outflow without noon: -68 -28 -29 n.l. n.l. n.l. -55 237 28 0 -89 macrophytes night: 45 7 116 18 45 84 73 206 62 150 * n.l. = flux rate was non-linear (R2 < 0.88) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.9. Carbon dioxide fluxes (mg CO2-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time.
Feb. Mar. Mar. Apr. Oct. Nov. Dec. 28 3-4 24-25 8-9 26 23-24 16 Wetland 1 a.m.: 0 17 -36 Inflow without noon: 0 20 n.l. n.l. -161 macrophytes night: 32 0 0 -24 Wetland 1 a.m.: 9 n.l. 0 Outflow without noon: 7 15 -167 n.l. n.l. macrophytes night: 25 5 0 n.l. Wetland 2 a.m.: 0 n.l. 12 0 34 Inflow without noon: n.l. 55 -32 21 macrophytes night: 1 7 16 39 Wetland 2 a.m.: -47 20 5 0 41 Outflow without noon: -55 23 6 70 macrophytes night: 13 25 11 36 107 * n.l. = flux rate was non-linear (R2 < 0.88) - red type indicates the water level was below the soil surface in that chamber during sampling - blue type indicates the soil was saturated in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.10. Carbon dioxide fluxes (mg CO2-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time. 155
Apr. May May June June July Aug. Aug. Aug. Sept. Sept. 25-26 9-10 23-24 10-11 29-30 7-8 4-5 10 24-25 4-5 28-29 Wetland 1 a.m.: 39 8 -97 n.l. -581 -47 -115 -68 -205 81 Outflow noon: 31 n.l. -111 -55 -57 -284 -250 -146 -173 marsh zone night: 47 123 95 n.l. 102 90 118 37 60 224 Wetland 2 a.m.: 0 0 -109 n.l. -512 -373 -344 -211 -352 163 Outflow noon: n.l. -34 n.l. n.l. -290 -273 -196 190 -190 -232 -380 marsh zone night: n.l. 23 35 25 60 103 150 262 246 428 Wetland 1 a.m.: n.l. n.l. 8 33 Inflow C.I. noon: n.l. -28 0 -53 -31 night: n.l. n.l. 15 26 268 Wetland 1 a.m.: 1 38 45 49 Outflow C.I. noon: -53 n.l. -26 n.l. 0 night: n.l. 31 0 n.l. Wetland 2 a.m.: 0 11 -103 56 Inflow C.I. noon: n.l. 103 -61 41 night: n.l. 45 33 41 36 Wetland 2 a.m.: 0 n.l. 22 45 Outflow C.I. noon: n.l. 39 0 -51 -24 night: 10 29 18 -50 - n.l. = flux rate was non-linear (R2 < 0.88) - C.I. = continuously inundated zone - red type indicates the water level was below the soil surface in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.11. Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time.
156
Feb. Mar. Mar. Apr. Oct. Nov. Dec. 28 3-4 24-25 8-9 26 23-24 16 Wetland 1 a.m.: n.l. n.l. n.l. Outflow noon: 0 0 27 0 -144 marsh zone night: 0 37 -215 0 Wetland 2 a.m.: -122 0 0 n.l. Outflow noon: n.l. 26 -119 marsh zone night: -77 34 111 0 Wetland 1 a.m.: Inflow C.I. noon: 0 42 night: 217 0 Wetland 1 a.m.: Outflow C.I. noon: 0 34 night: n.l. 0 Wetland 2 a.m.: 0 76 Inflow C.I. noon: 0 night: 54 0 Wetland 2 a.m.: 0 0 Outflow C.I. noon: 0 night: 0 0 - n.l. = flux rate was non-linear (R2 < 0.88) - C.I. = continuously inundated zone - red type indicates the water level was below the soil surface in that chamber during sampling - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table A.12. Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the non-growing season, 2004. Daytime rates of CO2 uptake are corrected for sampling time.
157
APPENDIX B
CH4 AND CO2 FLUX DATA, 2005
158
Apr. May May June June July Aug. Sept. 25-26 9-10 25-26 9 27-28 25-26 17 29-30 Wetland 1 a.m.: -7.01 1.31 0.89 4.71 3.98 9.55 0.30 Inflow with noon: n.l. 0.95 0.99 3.01 7.94 n.l. 0.71 macrophytes night: 0.00 0.75 1.37 5.81 9.88 13.00 -0.49 -1.53 Wetland 1 a.m.: 0.32 0.59 0.00 3.53 3.55 9.41 0.00 Outflow with noon: 0.00 0.00 1.24 4.46 0.00 8.93 n.l. n.l. macrophytes night: -1.25 0.00 0.13 4.93 4.32 1.21 Wetland 2 a.m.: 0.69 0.57 1.06 7.54 18.23 29.57 3.12 Inflow with noon: n.l. 0.00 5.13 8.08 32.49 2.97 n.l. macrophytes night: -1.44 -0.41 0.90 12.55 7.09 1.24 n.l. Wetland 2 a.m.: n.l. 0.00 0.59 6.05 28.50 23.62 0.00 Outflow with noon: 0.30 1.14 4.57 5.25 35.02 2.56 1.30 macrophytes night: -1.09 0.95 0.94 40.77 13.90 0.57 0.53 Wetland 1 a.m.: 0.10 1.41 1.93 4.69 6.39 30.11 1.13 Inflow noon: 1.00 1.41 2.34 3.87 n.l. 2.85 marsh zone night: -3.27 1.57 1.44 5.80 5.46 15.74 0.00 0.75 Wetland 2 a.m.: n.l. 1.01 0.77 1.35 4.22 12.87 n.l. Inflow noon: n.l. 1.21 1.93 9.79 1.61 0.00 marsh zone night: n.l. 0.00 1.49 1.66 2.75 17.60 n.l. -1.84 * n.l. = flux rate was non-linear (R2 < 0.90) - black type indicates the soil was inundated during sampling
-2 -1 Table B.1. Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2005.
159
Feb. Mar. Apr Oct. 22 21-22 4-5 29 Wetland 1 a.m.: 0.00 -2.28 0.00 0.00 Inflow with noon: 0.00 2.60 0.46 macrophytes night: 0.00 0.00 0.00 Wetland 1 a.m.: 0.00 0.00 0.00 0.58 Outflow with noon: n.l. -3.47 0.00 0.80 macrophytes night: 0.00 0.00 -3.15 0.00 Wetland 2 a.m.: 0.00 -0.78 2.74 n.l. Inflow with noon: -1.01 -2.05 0.92 0.00 macrophytes night: 0.00 0.00 1.94 0.60 Wetland 2 a.m.: 0.00 0.00 0.00 0.75 Outflow with noon: -2.12 0.00 -5.94 0.74 macrophytes night: 0.00 0.00 0.00 n.l. Wetland 1 a.m.: 0.00 2.02 1.29 0.20 Inflow noon: 1.11 n.l. 0.00 -0.50 marsh zone night: 0.00 0.00 -0.29 Wetland 2 a.m.: 0.00 0.00 n.l. 0.00 Inflow noon: 0.00 0.00 1.21 0.00 marsh zone night: 0.00 0.00 -0.39 0.00 * n.l. = flux rate was non-linear (R2 < 0.90) - black type indicates the soil was inundated during sampling
-2 -1 Table B.2. Methane fluxes (mg CH4-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2005.
160
Apr. May May June June July Aug. Sept. 25-26 9-10 25-26 9 27-28 25-26 17 29-30 Wetland 1 a.m.: 0.00 0.31 1.00 n.l 2.41 n.l 0.67 Inflow without noon: 0.00 0.43 0.88 1.06 39.32 0.51 n.l macrophytes night: 0.00 0.99 1.72 2.54 0.96 1.28 0.00 n.l Wetland 1 a.m.: 0.00 0.21 0.74 3.48 2.56 5.46 0.00 Outflow without noon: 0.19 n.l 2.63 5.08 1.68 0.00 0.00 macrophytes night: 0.86 2.00 0.92 5.79 n.l 1.70 0.00 Wetland 2 a.m.: -0.20 -0.18 0.93 6.28 0.73 1.43 Inflow without noon: -0.15 0.00 n.l 10.14 10.32 0.54 n.l macrophytes night: 0.00 0.00 1.02 15.99 n.l n.l 0.00 Wetland 2 a.m.: 0.06 1.06 0.68 5.48 5.28 9.27 n.l Outflow without noon: 0.16 0.00 1.66 5.56 14.82 0.61 n.l macrophytes night: 0.45 0.83 0.74 4.32 7.14 43.18 0.04 0.22 * n.l. = flux rate was non-linear (R2 < 0.90) - black type indicates the soil was inundated during sampling
-2 -1 Table B.3. Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the growing season, 2005.
Feb. Mar. Apr Oct. 22 21-22 4-5 29 Wetland 1 a.m.: 0.00 0.00 0.00 0.52 Inflow without noon: 0.00 -0.56 0.00 0.00 macrophytes night: 0.00 0.00 0.00 -0.12 Wetland 1 a.m.: 0.00 0.00 0.00 0.00 Outflow without noon: 0.00 n.l 0.35 0.00 macrophytes night: 0.00 0.00 0.13 0.00 Wetland 2 a.m.: 0.00 -0.50 0.00 0.00 Inflow without noon: 0.32 -0.17 n.l n.l macrophytes night: 0.00 0.00 0.39 0.05 Wetland 2 a.m.: 0.00 n.l 0.00 0.20 Outflow without noon: 0.00 0.00 n.l n.l macrophytes night: 0.00 0.00 0.00 0.63 * n.l. = flux rate was non-linear (R2 < 0.90) - black type indicates the soil was inundated during sampling
-2 -1 Table B.4. Methane fluxes (mg CH4-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2005.
161
Apr. May May June June July Aug. Sept. 25-26 9-10 25-26 9 27-28 25-26 17 29-30 Wetland 1 a.m.: 0.25 0.32 1.00 0.21 11.20 54.21 1.60 Inflow C.I. noon: 0.41 0.14 2.42 0.51 3.75 0.70 night: 0.00 0.08 0.12 0.32 4.64 10.96 1.10 0.60 Wetland 1 a.m.: 0.80 5.86 2.65 1.63 3.00 10.10 Outflow C.I. noon: 0.55 5.30 6.74 11.28 14.38 1.13 night: 2.51 2.49 16.09 16.01 18.91 8.61 Wetland 2 a.m.: n.l. 0.00 2.32 5.01 4.29 27.80 5.19 Inflow C.I. noon: 2.11 1.00 3.26 7.26 19.00 22.98 7.60 night: 0.50 13.79 5.16 42.56 3.37 19.74 12.66 7.54 Wetland 2 a.m.: 1.66 3.35 7.32 4.25 5.04 Outflow C.I. noon: 0.89 1.80 3.72 5.75 28.88 40.01 3.04 night: 1.40 9.34 12.99 3.63 45.77 36.10 27.69 12.85 Wetland 2 a.m.: 1.22 1.71 19.54 39.43 Outflow noon: 4.98 3.63 8.67 63.60 147.49 8.52 4.59 marsh zone night: 0.00 2.45 2.03 42.26 41.75 0.13 6.34 Wetland 1 a.m.: 0.35 6.25 11.55 16.35 23.75 1.86 Outflow noon: 0.57 6.39 6.11 12.54 15.96 2.01 1.28 marsh zone night: n.l. 9.30 6.82 16.90 17.56 32.02 n.l. * n.l. = flux rate was non-linear (R2 < 0.90) - black type indicates the soil was inundated during sampling
-2 -1 Table B.5. Methane fluxes (mg CH4-C m h ) from continuously inundated zone chambers on each sampling date during the growing season, 2005.
162
Feb. Mar. Apr Oct. 22 21-22 4-5 29 Wetland 1 a.m.: 0.00 -0.59 0.51 1.05 Outflow noon: n.l. 0.00 0.39 0.46 marsh zone night: 0.00 0.52 Wetland 2 a.m.: 0.00 -0.50 n.l. 0.29 Outflow noon: -0.77 0.46 marsh zone night: 0.00 0.00 0.00 0.25 Wetland 1 a.m.: 0.00 n.l. 0.00 0.13 Inflow C.I. noon: 0.42 0.00 1.34 0.26 night: 0.00 0.00 0.37 0.67 Wetland 1 a.m.: 0.00 0.00 0.00 3.20 Outflow C.I. noon: 0.32 0.42 1.93 0.53 night: 0.00 0.00 0.67 0.83 Wetland 2 a.m.: 0.00 1.73 0.00 n.l. Inflow C.I. noon: n.l. 0.00 1.55 4.57 night: 0.00 0.00 -0.51 3.04 Wetland 2 a.m.: 0.00 1.63 0.00 -1.70 Outflow C.I. noon: n.l. n.l. 0.00 n.l. night: 0.00 0.00 0.39 n.l.
* n.l. = flux rate was non-linear (R2 < 0.90) - black type indicates the soil was inundated during sampling
-2 -1 Table B.6. Methane fluxes (mg CH4-C m h ) from continuously inundated zone chambers on each sampling date during the non-growing season, 2005.
163
Apr. May May June June July Aug. Sept. 25-26 9-10 25-26 9 27-28 25-26 17 29-30 Wetland 1 a.m.: n.l. n.l. -362 -496 -1343 -843 -164 Inflow with noon: n.l. -478 -496 n.l. -787 n.l. macrophytes night: 284 138 491 429 452 252 89 Wetland 1 a.m.: 291 -478 -730 -379 -215 -698 -232 Outflow with noon: n.l. -395 -246 -402 -379 -256 macrophytes night: 309 159 224 346 424 421 94 Wetland 2 a.m.: n.l. -546 -918 -903 -758 -910 -550 Inflow with noon: n.l. -629 -375 n.l. -224 -354 -337 macrophytes night: 243 195 254 506 n.l. 338 284 Wetland 2 a.m.: n.l. -726 -831 -339 -1814 -857 -550 Outflow with noon: -604 -599 -119 n.l. -151 -181 -270 macrophytes night: 219 290 -513 574 n.l. 194 160 Wetland 1 a.m.: n.l. -301 -560 -605 -806 -371 -360 Inflow noon: -222 n.l. -632 -692 -294 n.l. marsh zone night: n.l. 353 226 832 449 302 176 n.l. Wetland 2 a.m.: n.l. -569 -1001 -427 n.l. -868 Inflow noon: n.l. -384 -994 0 n.l. -376 n.l. marsh zone night: n.l. n.l. 491 683 868 282 n.l. - n.l. = flux rate was non-linear (R2 < 0.88) - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table B.7. Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the growing season, 2005.
164
Mar. Apr Oct. 21-22 4-5 29 Wetland 1 a.m.: 0 n.l. -130 Inflow with noon: 0 n.l. n.l. macrophytes night: 0 71 Wetland 1 a.m.: 0 291 151 Outflow with noon: 0 n.l. -101 macrophytes night: 0 309 22 Wetland 2 a.m.: 0 n.l. -213 Inflow with noon: 0 n.l. n.l. macrophytes night: 0 243 26 Wetland 2 a.m.: 0 n.l. n.l. Outflow with noon: 0 -604 n.l. macrophytes night: 0 219 n.l. Wetland 1 a.m.: 0 0 n.l. Inflow noon: 0 0 n.l. marsh zone night: 0 n.l. Wetland 2 a.m.: 0 0 -125 Inflow noon: 0 0 0 marsh zone night: 0 0 0
- n.l. = flux rate was non-linear (R2 < 0.88) - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table B.8. Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers with emergent macrophytes on each sampling date during the non-growing season, 2005.
165
Apr. May May June June July Aug. Sept. 25-26 9-10 25-26 9 27-28 25-26 17 29-30 Wetland 1 a.m.: n.l. 0 20 -99 n.l. Inflow without noon: n.l. n.l. -37 n.l. -204 n.l. -118 macrophytes night: 55 29 n.l. 70 91 35 n.l. Wetland 1 a.m.: n.l. 8 0 34 n.l. 106 n.l. Outflow without noon: -5 -33 n.l. -55 -34 macrophytes night: 42 29 57 n.l. 62 41 Wetland 2 a.m.: n.l. -39 -27 18 -22 -82 Inflow without noon: n.l. -22 n.l. -48 -31 -58 macrophytes night: n.l. 28 11 136 49 n.l. Wetland 2 a.m.: n.l. 0 -66 -52 -212 n.l. Outflow without noon: -53 -41 -21 n.l. -111 -48 -228 macrophytes night: n.l. 38 44 113 151 280 9 163 - n.l. = flux rate was non-linear (R2 < 0.88) - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table B.9. Carbon dioxide fluxes (mg CO2-C m h ) from edge/marsh zone chambers without emergent macrophytes on each sampling date during the growing season, 2005.
Mar. Apr Oct. 21-22 4-5 29 Wetland 1 a.m.: 0 0 n.l. Inflow without noon: 0 -24 macrophytes night: 0 7 Wetland 1 a.m.: 0 0 33 Outflow without noon: 0 0 1 macrophytes night: 0 0 7 Wetland 2 a.m.: 0 0 n.l. Inflow without noon: 0 0 -59 macrophytes night: 0 0 8 Wetland 2 a.m.: 0 0 -60 Outflow without noon: 0 0 n.l. macrophytes night: 0 0 71 * n.l. = flux rate was non-linear (R2 < 0.88) - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table B.10. Carbon dioxide fluxes (mg CO2-C m h ) from edge zone chambers without emergent macrophytes on each sampling date during the non-growing season, 2005.
166
Apr. May May June June July Aug. Sept. 25-26 9-10 25-26 9 27-28 25-26 17 29-30 Wetland 1 a.m.: 105 28 0 n.l. -59 -34 n.l. Inflow C.I. noon: n.l. -10 -17 n.l. n.l. -43 night: n.l. 17 0 n.l. 53 63 6 -13 Wetland 1 a.m.: n.l. n.l. n.l. n.l. -38 Outflow C.I. noon: n.l. -12 -53 -64 n.l. n.l. night: 37 n.l. n.l. n.l. n.l. -37 Wetland 2 a.m.: n.l. 33 -10 n.l. -5 -112 31 Inflow C.I. noon: n.l. n.l. 12 0 n.l. -119 4 night: n.l. 44 -50 n.l. 103 57 -12 n.l. Wetland 2 a.m.: n.l. -183 n.l. -115 16 -21 Outflow C.I. noon: -57 0 -27 n.l. -63 n.l. 0 night: 45 102 5 0 n.l. n.l. n.l. n.l. Wetland 2 a.m.: n.l. -744 -1337 -457 -580 Outflow noon: -303 -378 -419 -632 -408 -625 -277 marsh zone night: -490 109 176 353 455 435 71 Wetland 1 a.m.: -156 -158 n.l. -288 Outflow noon: -944 n.l. 149 n.l. -125 n.l. -120 marsh zone night: 217 119 159 150 193 n.l. n.l. - n.l. = flux rate was non-linear (R2 < 0.88) - C.I. = continuously inundated zone - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table B.11. Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the growing season, 2005.
167 Mar. Apr Oct. 21-22 4-5 29 Wetland 1 a.m.: 0 0 n.l. Inflow C.I. noon: 0 0 n.l. night: 0 8 Wetland 1 a.m.: 0 0 3 Outflow C.I. noon: 0 -11 night: 0 0 8 Wetland 2 a.m.: 0 0 n.l. Inflow C.I. noon: 0 0 0 night: 0 0 n.l. Wetland 2 a.m.: 0 0 n.l. Outflow C.I. noon: 0 0 n.l. night: 0 0 n.l. Wetland 2 a.m.: 0 0 -225 Outflow noon: 0 0 -4 marsh zone night: 0 0 26 Wetland 1 a.m.: 0 0 -465 Outflow noon: 0 0 n.l. marsh zone night: 0 0 0 - n.l. = flux rate was non-linear (R2 < 0.88) - C.I. = continuously inundated zone - black type indicates the soil was inundated in that chamber during sampling
-2 -1 Table B.12. Carbon dioxide fluxes (mg CO2-C m h ) from chambers in continuously inundated wetland zones on each sampling date during the non-growing season, 2005.
168
APPENDIX C
CH4 AND CO2 FLUXES FROM MESOCOSMS, 2005
169 Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. n.l. 0.00 0.00 0.00 0.00 0.00 TUB 2 noon 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 night 0.00 0.43 0.00 0.12 0.00 0.21 0.22 0.27 a.m. 0.14 0.00 0.00 0.00 0.00 0.00 TUB 4 noon 0.00 0.44 0.00 0.00 0.00 0.00 0.00 0.00 night 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.54 a.m. 0.00 0.00 0.00 0.00 0.38 0.58 TUB 16 noon 0.00 0.42 0.17 1.50 0.34 0.00 0.00 0.74 night 0.00 0.00 0.00 0.00 0.97 3.62 -2.08 a.m. 0.66 0.00 -0.56 0.00 0.21 0.00 TUB F1 noon 0.00 0.00 -0.13 n.l. 0.46 n.l. 0.00 n.l. night 0.00 0.00 0.14 0.11 0.00 0.00 -0.26 0.31 a.m. 0.00 0.00 0.00 0.00 0.00 0.65 TUB G1 noon 0.00 0.00 0.64 1.52 0.00 0.00 0.98 0.97 night 0.00 0.64 n.l. 0.14 0.00 0.92 2.91 0.68 a.m. 0.00 0.00 0.00 0.00 0.00 0.11 TUB H2 noon 0.00 0.00 0.00 0.40 0.00 0.00 0.24 0.00 night 0.00 0.00 0.00 0.21 0.73 0.00
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. 0.00 0.00 0.00 TUB 2 noon 0.37 0.00 0.00 0.00 0.68 n.l. 0.65 0.00 -1.00 night 0.75 -1.03 0.06 2.87 n.l. a.m. 0.00 -0.45 0.00 TUB 4 noon 0.18 -0.43 0.00 5.07 4.07 -0.93 0.81 -0.01 0.24 night -2.37 0.00 4.87 4.14 0.00 a.m. 0.65 0.87 0.00 TUB 16 noon 0.44 0.00 0.73 -0.17 -0.45 -0.16 0.00 0.00 night 0.56 -0.13 0.00 0.54 0.00 a.m. -0.53 0.52 TUB F1 noon 0.00 0.00 0.28 0.00 n.l. 0.24 n.l. 0.34 n.l. night 0.00 0.00 -1.03 n.l. 0.24 a.m. 3.03 0.85 TUB G1 noon 3.00 5.27 2.86 0.00 1.12 1.48 n.l. 1.10 -0.60 night 0.14 0.91 1.16 0.69 0.00 a.m. 0.47 n.l. 0.55 TUB H2 noon 0.74 0.00 0.89 0.97 1.58 0.29 0.92 -2.06 night 0.00 0.21 0.00 0.47 0.00
- n.l. = flux rate was non-linear (R2 < 0.90)
-2 -1 Table C.1. Methane fluxes (mg CH4-C m h ) from mesocosms with non-hydric soils and steady-flow hydrology on each sampling date (2005).
170
Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. 0.51 0.58 1.84 3.13 TUB3 noon 0.34 1.71 1.59 3.03 4.54 4.54 night 0.00 0.51 3.01 3.20 4.18 7.78 a.m. 0.00 0.28 0.00 0.00 0.25 0.69 TUB 7 noon 0.00 0.00 0.09 n.l. 0.00 0.00 1.05 1.36 night 0.00 0.00 0.00 0.23 0.80 0.96 1.31 a.m. -0.56 0.00 1.05 0.32 0.54 0.60 TUB 9 noon 0.00 0.00 0.33 1.45 0.41 0.97 1.52 n.l. night 0.00 0.00 0.00 0.38 n.l. 0.85 1.04 0.93 a.m. 0.00 0.00 n.l. 0.00 1.11 2.22 TUB 10 noon 0.80 0.62 0.64 0.65 0.98 2.84 1.81 n.l. night 0.00 0.00 -0.50 0.76 0.93 1.78 0.90 0.95 a.m. 0.00 0.42 n.l. 0.00 0.00 1.48 TUB 13 noon 0.00 0.00 0.00 0.53 0.29 1.15 1.14 0.00 night 0.00 0.55 0.00 0.15 0.50 1.36 0.00 n.l.
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. 4.31 7.09 3.31 TUB3 noon 7.47 5.17 4.80 0.00 0.00 5.29 1.74 2.94 1.09 night 20.99 10.52 0.00 -1.17 1.12 a.m. 0.62 1.52 0.61 TUB 7 noon 0.79 1.18 0.46 -1.05 1.04 0.10 0.59 0.00 night 1.20 0.83 0.60 0.83 0.32 a.m. 0.68 1.13 0.29 TUB 9 noon 0.88 1.00 1.34 1.42 2.04 0.94 1.25 0.74 0.00 night 0.71 0.82 1.28 0.96 0.22 a.m. 1.85 1.19 0.72 TUB 10 noon 1.49 1.69 1.80 1.30 2.04 1.07 0.77 n.l. 0.43 night 2.14 0.97 1.21 2.17 0.63 a.m. 3.80 0.39 4.18 TUB 13 noon 1.13 1.72 0.00 n.l. 1.19 -1.45 0.69 0.22 night 1.77 1.96 1.55 n.l. 0.50
- n.l. = flux rate was non-linear (R2 < 0.90)
-2 -1 Table C.2. Methane fluxes (mg CH4-C m h ) from mesocosms with hydric soils and steady-flow hydrology on each sampling date (2005).
171
Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. -0.29 0.00 0.00 0.00 0.00 0.00 TUB 12 noon 0.00 0.00 0.00 0.00 0.00 n.l. 0.00 0.00 night 0.00 0.84 0.00 0.19 0.00 0.00 2.87 0.78 a.m. 0.11 0.00 0.00 0.00 0.00 0.00 TUB 14 noon 0.00 -0.13 0.00 -0.65 0.26 0.00 0.59 -1.90 night 0.45 0.16 0.00 0.22 0.00 0.18 1.59 0.00 a.m. 0.00 n.l. n.l. 0.00 0.00 0.00 TUB 18 noon 0.00 0.50 0.00 0.00 -0.08 0.31 0.00 0.00 night 0.00 0.18 0.00 0.00 -0.15 0.00 2.49 0.00 a.m. 0.00 n.l. -2.12 0.00 0.29 0.94 TUB 19 noon 0.00 n.l. 0.00 1.82 0.00 0.00 0.57 0.00 night 0.00 0.36 0.25 0.14 0.15 0.96 0.00 0.00 a.m. 0.00 0.00 0.00 0.00 2.76 0.00 TUB 20 noon 0.00 0.00 0.00 -0.94 0.00 0.00 0.00 n.l. night 0.00 0.00 0.15 0.23 0.22 -0.92 0.00
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. 0.00 0.45 0.00 TUB 12 noon n.l. 0.00 0.00 0.00 0.00 -0.74 -0.60 -0.27 0.36 night 0.33 0.00 n.l. 0.28 0.00 a.m. n.l. 0.24 0.27 TUB 14 noon 0.43 0.00 0.66 2.17 0.67 n.l. n.l. -1.31 0.00 night 0.00 0.49 n.l. 1.12 0.17 a.m. 0.00 0.00 0.32 TUB 18 noon 1.30 0.00 0.00 0.00 -0.44 1.75 n.l. 0.00 n.l. night 0.00 0.29 0.00 n.l. 0.00 a.m. 0.31 0.37 0.00 TUB 19 noon 0.99 0.31 0.80 0.00 0.95 0.45 0.84 n.l. n.l. night 1.21 0.00 0.26 -0.27 a.m. 0.00 n.l. 0.00 TUB 20 noon n.l. 0.00 0.28 0.00 n.l. 0.00 n.l. 0.00 0.00 night 0.00 -0.49 0.89 0.10
- n.l. = flux rate was non-linear (R2 < 0.90)
-2 -1 Table C.3. Methane fluxes (mg CH4-C m h ) from mesocosms with non-hydric soils and flood pulse hydrology on each sampling date (2005).
172
Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. 0.00 0.00 0.12 0.96 2.53 6.20 TUB 5 noon 0.00 0.83 1.68 2.57 2.98 4.79 6.47 0.00 night 0.00 0.00 0.58 1.37 2.83 5.08 0.71 0.00 a.m. 0.00 0.00 0.00 0.95 1.57 3.34 TUB 6 noon 0.00 0.00 1.00 3.17 1.91 2.88 2.68 0.00 night 0.49 0.07 1.00 1.06 2.15 0.00 0.80 a.m. 0.00 0.00 -0.65 0.00 0.65 n.l. TUB 11 noon -0.60 -1.54 0.37 0.53 0.36 0.97 0.77 0.00 night n.l. 0.00 0.21 0.67 0.53 2.26 0.00 a.m. 0.29 0.00 0.00 0.00 0.00 0.00 TUB 15 noon 0.00 0.00 0.58 n.l. 0.20 0.00 0.00 0.00 night 0.00 0.00 0.45 0.29 0.00 0.00 0.00 1.06 a.m. 0.00 0.00 0.00 0.00 0.00 0.40 TUB 17 noon 0.00 0.00 0.68 -0.45 0.00 1.63 0.63 0.00 night 0.00 -0.89 0.36 0.00 0.00 0.00 1.17
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. 3.52 n.l. 1.85 TUB 5 noon 5.21 0.00 5.05 0.00 0.36 0.44 0.00 1.81 0.92 night 1.78 2.28 0.66 n.l. 0.90 a.m. 2.32 -0.33 0.68 TUB 6 noon 2.45 2.06 2.85 n.l. -0.53 0.66 0.71 night 0.00 1.33 0.37 n.l. 0.61 a.m. n.l. 0.00 0.00 TUB 11 noon 0.00 2.35 0.74 2.36 1.54 0.00 -3.09 0.00 -2.08 night 0.63 -0.27 1.93 -0.31 0.00 a.m. 0.18 0.00 0.00 TUB 15 noon 0.57 0.00 0.00 0.00 -2.73 -0.25 -0.49 0.00 -0.95 night 0.11 0.79 -2.04 1.27 -0.30 a.m. 0.81 0.00 0.21 TUB 17 noon n.l. 0.65 0.53 0.33 1.93 0.85 -0.18 0.59 0.00 night 0.33 0.00 0.00 n.l. 0.31
- n.l. = flux rate was non-linear (R2 < 0.90)
-2 -1 Table C.4. Methane fluxes (mg CH4-C m h ) from mesocosms with hydric soils and flood pulse hydrology on each sampling date (2005).
173 Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. -136 -478 n.l. -526 -417 TUB 2 noon n.l. -235 -530 -792 -333 -701 -806 -809 night n.l. 189 n.l. n.l. 156 739 n.l. n.l. a.m. -58 192 -479 -478 -220 -882 TUB 4 noon -99 n.l. -595 -326 -433 -661 -623 -588 night n.l. -449 194 100 163 399 419 208 a.m. n.l. n.l. -90 -324 -563 TUB 16 noon n.l. n.l. -468 -697 n.l. -507 n.l. -471 night 91 253 n.l. 150 184 382 n.l. a.m. 60 n.l. -433 -380 -480 TUB F1 noon n.l. -253 -659 -566 -516 -302 -638 -648 night -244 n.l. 17 155 n.l. 312 93 264 a.m. n.l. n.l. -143 -364 -312 -475 TUB G1 noon n.l. n.l. -420 -459 -352 -532 -637 night -72 267 n.l. 88 108 179 262 a.m. n.l. n.l. -271 -561 -720 TUB H2 noon n.l. n.l. -508 -336 n.l. -574 -931 night n.l. 67 82 394 130
June July July July July Aug Aug Aug Aug Sept
29 7 12 18 28 2 10 16 25 20 a.m. -425 -357 -324 TUB 2 noon -306 -585 -579 -963 -363 -394 -336 -304 n.l. night 378 n.l. 188 n.l. a.m. -436 -395 -312 TUB 4 noon -421 -578 -447 -663 -338 -1 -504 -405 n.l. night 174 320 311 n.l. 78 a.m. -364 -495 -246 TUB 16 noon -472 -474 -408 -441 -536 -427 -358 n.l. night 184 187 169 176 95 a.m. -470 -438 TUB F1 noon -834 -715 -228 -446 -82 -124 -523 -301 n.l. night 206 190 251 79 135 a.m. -679 -469 TUB G1 noon -822 -554 -210 -477 -425 -466 -350 -359 n.l. night 304 40 100 168 n.l. a.m. -516 -442 -377 TUB H2 noon -582 -898 -483 -598 -208 -461 -380 -65 night n.l. 194 312 n.l. - n.l. = flux rate was non-linear (R2 < 0.88)
-2 -1 Table C.5. Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with non-hydric soils and steady-flow hydrology on each sampling date (2005). 174
Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. -190 -261 n.l. -512 TUB3 noon -428 -352 -353 -515 -584 -583 night 85 n.l. 159 559 a.m. n.l. n.l. -386 -456 -646 TUB 7 noon n.l. 11 -439 -441 -492 -565 -55 -348 night 100 n.l. 344 116 296 308 139 a.m. n.l. n.l. -150 -302 -175 -602 TUB 9 noon n.l. n.l. -269 -337 -317 -383 -196 -293 night n.l. n.l. 71 n.l. 396 317 133 a.m. 34 n.l. n.l. -445 -347 -902 TUB 10 noon n.l. n.l. -380 -453 -530 -513 -180 -501 night n.l. n.l. -127 78 n.l. n.l. n.l. 93 a.m. n.l. n.l. n.l. -520 -530 -724 -634 TUB 13 noon n.l. n.l. -463 -456 -221 -568 -900 night n.l. 130 157 128 n.l. 253 n.l.
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. -508 -516 n.l. TUB3 noon -610 -469 -276 -399 n.l. -345 n.l. night n.l. 212 226 n.l. a.m. -471 -452 -259 TUB 7 noon -380 -543 -114 -427 n.l. -356 -281 n.l. night 230 192 233 181 172 a.m. -949 -409 -306 TUB 9 noon -390 -482 -380 -316 n.l. -251 -501 -786 -51 night 210 137 n.l. n.l. 90 a.m. -471 -836 -639 TUB 10 noon -223 -512 -411 -336 -288 -303 -340 -428 n.l. night 316 240 163 208 167 a.m. -810 -665 -346 TUB 13 noon -480 -481 -543 -218 -291 -387 -250 n.l. night 358 192 189 n.l. 106
- n.l. = flux rate was non-linear (R2 < 0.88)
-2 -1 Table C.6. Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with hydric soils and steady-flow hydrology on each sampling date (2005).
175
Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. n.l. -95 -329 -411 n.l. n.l. TUB 12 noon n.l. n.l. -557 -459 n.l. -1389 -679 n.l. night -63 n.l. n.l. 65 51 n.l. n.l. 98 a.m. n.l. -86 -306 -482 -634 TUB 14 noon n.l. 57 -718 -498 -243 -714 -655 -576 night 148 -165 n.l. 128 292 71 a.m. n.l. 470 n.l. -323 -295 -498 TUB 18 noon n.l. 121 -343 -273 -194 -328 -1141 -281 night 194 n.l. -248 79 212 508 397 93 a.m. n.l. -162 -157 -43 n.l. -215 TUB 19 noon n.l. n.l. -320 n.l. n.l. -339 night n.l. n.l. n.l. 100 n.l. n.l. a.m. n.l. -75 -105 n.l. -334 TUB 20 noon n.l. n.l. -446 -554 -338 -341 -645 -535 night -258 n.l. n.l. 125 215 534 409 159
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. -378 -579 -437 TUB 12 noon n.l. -224 -405 -238 -474 -588 -592 -137 night 50 127 227 232 214 a.m. -416 -736 -160 TUB 14 noon -488 -389 -595 -395 n.l. -208 -460 -407 n.l. night n.l. 164 160 264 90 a.m. -379 -570 n.l. TUB 18 noon -575 -517 -338 -171 -527 -365 -459 -279 -79 night 100 115 205 106 185 a.m. -635 -581 -376 TUB 19 noon -594 -189 -629 -316 -336 -242 -743 -340 -194 night 155 176 55 n.l. a.m. -389 n.l. -212 TUB 20 noon -522 n.l. -390 -510 -562 -177 -353 -327 n.l. night 278 203 95 147
- n.l. = flux rate was non-linear (R2 < 0.88)
-2 -1 Table C.7. Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with non-hydric soils and flood pulse hydrology on each sampling date (2005).
176 Apr Apr May May May June June June June June 6 18 4 17 31 2 8 14 21 23 a.m. 44 n.l. -163 -364 -97 -544 TUB 5 noon n.l. 230 -311 -385 -339 -547 -373 -378 night n.l. -166 n.l. 117 660 n.l. 380 a.m. n.l. -644 n.l. -428 -224 n.l. TUB 6 noon n.l. n.l. -392 -566 n.l. -520 -388 night 108 -92 218 74 526 n.l. 116 a.m. n.l. 136 -249 -322 -390 -469 TUB 11 noon n.l. n.l. -406 -721 -413 -728 n.l. -401 night n.l. n.l. n.l. 68 n.l. 1223 161 a.m. n.l. -154 -307 -309 -551 TUB 15 noon n.l. n.l. -520 -455 -236 n.l. -159 n.l. night n.l. n.l. n.l. 279 421 512 n.l. 162 a.m. n.l. n.l. -342 -358 -364 TUB 17 noon n.l. 158 -392 -508 -144 -337 -387 -284 night 20 -184 76 n.l. 374 60 n.l.
June July July July July Aug Aug Aug Aug Sept 29 7 12 18 28 2 10 16 25 20 a.m. -449 -559 -342 TUB 5 noon -549 -566 -392 -279 -231 -423 -444 n.l. night 468 377 184 220 100 a.m. -430 -543 -408 TUB 6 noon -481 -303 -487 -246 -645 -555 -63 night n.l. n.l. n.l. 182 a.m. -384 -669 -451 TUB 11 noon -393 n.l. -600 -224 -437 n.l. -433 -314 -137 night n.l. 150 171 n.l. n.l. a.m. -562 -571 -123 TUB 15 noon -369 -346 -496 -282 -654 -251 -408 -180 -73 night 399 215 418 184 130 a.m. -310 -547 -240 TUB 17 noon -341 -354 -346 -512 -368 -226 -472 -354 -99 night 242 182 181 n.l. n.l.
- n.l. = flux rate was non-linear (R2 < 0.88)
-2 -1 Table C.8. Carbon dioxide fluxes (mg CO2-C m h ) from mesocosms with hydric soils and flood pulse hydrology on each sampling date (2005).
177