USF Libraries
USF Digital Collections

Using Keeling Plots to trace δ ¹³C and δ ¹³O through processes of heterotrophic respiration, diffusion and soil water eq...

MISSING IMAGE

Material Information

Title:
Using Keeling Plots to trace δ ¹³C and δ ¹³O through processes of heterotrophic respiration, diffusion and soil water equilibration in artificial C3- and C4-grassland soils
Alternate title:
Using Keeling Plots to trace deltasuperscript 13C and deltasuperscript13 of COsubscript 2 through processes of heterotrophic respiration, diffusion and soil equilibration in artificial C3- and C4- grassland soils
Physical Description:
Book
Language:
English
Creator:
Chelladurai, Jennifer
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Stable isotopes
Carbon
Oxygen
Carbonates
Fractionation
Dissertations, Academic -- Geology -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Global carbon cycle dynamics and fluxes of CO₂ between biosphere and atmosphere have been progressed through the use of Keeling Plots. Processes that control and effect the isotopic composition of soil-respired CO₂, soil CO₂, and equilibrated soil carbonate are specifically addressed in this study through the use of Keeling Plots. Replicate grassland soil profiles containing either C3 or C4 homogenized organic matter were constructed and maintained under controlled settings to encourage the production of soil-respired CO₂ and the precipitation of pedogenic carbonate. Soil CO₂ was sampled over five months and analyzed with IRMS. Keeling Plots illustrated source CO₂ affected by mixing with atmospheric CO₂ near the surface and equilibration with ¹³C-depleted CO₂ at depth in the zone of likely carbonate precipitation. The δ ¹³C Keeling Plot intercepts for the surface horizons (~ -24.7 per mil for C3 profiles and ~ -11.1 per mil for C4 profiles) follow the diffusion-production model when corrected with a constant 4.4 per mil diffusional fractionation, but the Keeling Plot intercepts for developing Bk horizons were curved towards depleted values (~ -36.2 per mil for C3 profiles and ~ -18.4 per mil for C4 profiles). This change in isotopic composition with depth deviates from the usual interpretations of Keeling Plots (steady-state, source to background diffusional mixing). δ ¹³C Keeling Plot intercepts indicated evaporative enrichment in the surface horizons of C3 and C4 profiles). This study uses Keeling Plots as a measure of mixing to assess the efficacy of steady-state diffusion-production models of soil CO₂ equilibration with soil carbonate.
Thesis:
Thesis (M.S.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Jennifer Chelladurai.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 101 pages.
General Note:
Non-Latin script record

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002029607
oclc - 436988846
usfldc doi - E14-SFE0002946
usfldc handle - e14.2946
System ID:
SFS0027263:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Using Keeling Plots to Trace 13C and 18O of CO2 Through Processes of Heterotrophic Respiration, Diffusion and Soil Water Equilibra tion in Artificial C3and C4-Grassland Soils by Jennifer Chelladurai A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology College of Arts and Sciences University of South Florida Major Professor: Jonathan Wynn, Ph.D. Mark Rains, Ph.D. Mark Stewart, Ph.D. Date of Approval: April 8, 2009 Keywords: stable isotopes, carbon, oxygen, carbonates, fractionation Copyright 2009, Jennifer Chelladurai

PAGE 2

Dedication I dedicate this thesis to my daught er, Anoushka Chelladurai, whose future I hope will be brighter on account of my successful completion of this Master of Science degree.

PAGE 3

Acknowledgements I would first and foremost like to acknowledge my husband, for his patience and his hard work that allowed me the time to complete this degree; my mother-in-law for assisting me with ch ildcare and household chores while I was busy, and my daughter for tolerating my fr equent absence from her life during my work on this degree. I woul d also like to acknowledge my parents for listening to my frustrations, and my aunt for inspir ing me to complete a Masters degree. I would like to acknowledge my advi sor, Jonathan G. Wynn, Ph.D. for answering my questions and for providi ng help and encouragement when needed throughout my completion of this project and degree. I would also like to acknowledge my committee members, who provided valuable feedback on my thesis and for providing valuable instructi on in the writing of theses and papers.

PAGE 4

i Table of Contents List of Tables iii List of Figures iv Abstract vi Introduction 1 Soil Processes and the Global Carbon Cycle 2 Pedogenic Carbonates 3 Controls on Soil Respiration 5 Stable Isotopes as a Tracer of the Carbon Cycle 7 Keeling Plots 11 Cerling Model of Soil CO2 Diffusion-Production 12 Expectations and Hypotheses 17 Materials and Methods 18 Apparatus 19 Soil Profiles 24 Water Addition 26 Gas Sampling 28 IRMS Analysis 29 Plots 30 Results CO2 Concentration with Depth 31 13C Keeling Plots 36 18O Keeling Plots 46 13C and 18O with Depth 57 13C versus 18O 61 Discussion Concentration versus Depth 64 13C Keeling Plots 68 18O Keeling Plots 70 Cross plots of 13C versus 18O 72 Possible Issues 73

PAGE 5

Conclusions 75 Future Thoughts 76 References 78 Bibliography 82 Appendices 84 Appendix A: Normalized Values from Utilized Samples of Soil CO2 85 Appendix B: Data from Early Sampling Dates 89 Appendix C: Effluent pH 93 Appendix D: Watering and Gas Sampling Schedule 95 Appendix E: Additional Keeli ng Plot Line Equations and R2 Values 96 Appendix F: Column T3 Problems 101 ii

PAGE 6

List of Tables Table 1. Apparatus Measurements 24 Table 2. Soil Profile Parameters 26 Table 3. 13C Keeling Plot Line Equations and R2 Values 39 Table 4. 13C Keeling Plot y-intercepts for Varying Soil Moisture 46 Table 5. 18O Keeling Plot Line Equations and R2 Values 50 Table 6. 18O Keeling Plot y-intercepts for Varying Soil Moisture 57 iii

PAGE 7

List of Figures Figure 1. Soil Carbonate Fractionation Diagram 10 Figure 2. Cerling Model CO2 versus Depth 14 Figure 3. Cerling Model 13C versus Depth 15 Figure 4. Cerling Model Keeling Plot 16 Figure 5. Apparatus 21 Figure 6. Gas Collection Tube 22 Figure 7. Empty Soil Column 23 Figure 8. Soil Profile 25 Figure 9. CO2 Concentration versus Depth 32 Figure 10. CO2 Concentration versus Depth in Wet Soil 33 Figure 11. CO2 Concentration versus Depth in Moist Soil 34 Figure 12. CO2 Concentration versus Depth in Dry Soil 35 Figure 13. C3 13C Keeling Plots 37 Figure 14. C4 13C Keeling Plots 38 Figure 15. C3 13C Keeling Plots for Wet Soil Conditions 40 Figure 16. C3 13C Keeling Plots for Moist Soil Conditions 41 Figure 17. C3 13C Keeling Plots for Dry Soil Conditions 42 Figure 18. C4 13C Keeling Plots for Wet Soil Conditions 43 Figure 19. C4 13C Keeling Plots for Moist Soil Conditions 44 iv

PAGE 8

Figure 20. C4 13C Keeling Plots for Dry Soil Conditions 45 Figure 21. C3 18O Keeling Plots 48 Figure 22. C4 18O Keeling Plots 49 Figure 23. C3 18O Keeling Plots for Wet Soil Conditions 51 Figure 24. C3 18O Keeling Plots for Moist Soil Conditions 52 Figure 25. C3 18O Keeling Plots for Dry Soil Conditions 53 Figure 26. C4 18O Keeling Plots for Wet Soil Conditions 54 Figure 27. C4 18O Keeling Plots for Moist Soil Conditions 55 Figure 28. C4 18O Keeling Plots for Dry Soil Conditions 56 Figure 29. 13C versus Depth 59 Figure 30. 18O versus Depth 60 Figure 31. C3 13C versus 18O 62 Figure 32. C4 13C versus 18O 63 Figure 33. 18O versus Depth from Sternberge, et al. (1998) 71 v

PAGE 9

Using Keeling Plots to Trace 13C and 18O of CO2 Through Processes of Heterotrophic Respiration, Diffusion and Soil Water Equilibra tion in Artificial C3and C4-Grassland Soils Jennifer Chelladurai ABSTRACT Global carbon cycle dynamics and fluxes of CO2 between biosphere and atmosphere have been progressed through the use of Keeling Plots. Processes that control and effect the isotop ic composition of soil-respired CO2, soil CO2, and equilibrated soil carbonate are specifically addressed in this study through the use of Keeling Plots. Replicate grassland soil profiles containing either C3 or C4 homogenized organic matter were constr ucted and maintained under controlled settings to encourage the pr oduction of soil-respired CO2 and the precipitation of pedogenic carbonate. Soil CO2 was sampled over five months and analyzed with IRMS. Keeling Plots illustrated source CO2 affected by mixing with atmospheric CO2 near the surface and equilibration with 13C-depleted CO2 at depth in the zone of likely carbonate precipitation. The 13C Keeling Plot intercepts for the surface horizons (~ -24.7 for C3 profiles and ~ -11.1 for C4 profiles) follow the vi

PAGE 10

diffusion-production model when corrected with a constant 4.4 diffusional fractionation, but the Keeling Plot inte rcepts for developing Bk horizons were curved towards depleted values (~ -3 6.2 for C3 profiles and ~ -18.4 for C4 profiles). This change in isotopic compos ition with depth deviates from the usual interpretations of Keeling Plots (stea dy-state, source to background diffusional mixing). 18O Keeling Plot intercepts indicated evaporative enrichment in the surface horizons of C3 and C4 replicate so il profiles. This study uses Keeling Plots as a measure of mixing to assess the efficacy of steady-state diffusionproduction models of soil CO2 equilibration with soil carbonate. vii

PAGE 11

1 Introduction This study specifically addresses the need for increased knowledge of the processes that control the isotopic compositi on of soil-respired CO2 and its effect on the 13C and 18O values of soil CO2 and equilibrated so il carbonate. Much work has focused on the isotopic composition of total soil-respired CO2 although much of what has been done was carried ou t in field soils, in which the ability to separate causative factors ma y be limited. There has also been some work using the isotopic composition of soil-respired CO2 in pedogenic carbonates to determine paleoenvironmental conditions. In neither of these situations have laboratory-built soils been utilized to measure the isotopic composition of soilrespired CO2 and to precipitate pedogenic carbon ates to test the effect of their precipitation on the isotopic values of carbon and oxygen in the soil CO2. It is important to fully understand the factors which influence soil CO2 isotopic composition and its role in the global car bon cycle; and the sources and sinks of atmospheric CO2 are often determined by stable isotope studies. Knowledge gained on the isotopic processes during soil respiration will ultimately be of use in biogeochemistry, ecosystem studies, a nd paleoenvironmental reconstruction. The purpose of this thesis is to iden tify the factors that may influence the carbon and oxygen isotopic composition of soil respired CO2, and to validate the use of Keeling Plots as a measure of mixing soil respired CO2 with atmospheric

PAGE 12

CO2, through the use of artificial soil profiles. This study will also compare the results from collected data to t hose predicted by the Cerling (1984) CO2 diffusion production model. This introductory section will cover so il processes and their role in the Global Carbon Cycle, the formation and utilization of pedogenic carbonates in terms of soil respiration, the controls on soil respiration, how stable isotopes can be used to trace the Carbon Cycle, Keeling plots and their use in soil studies, and the use of the Cerling model of CO2 diffusion-production. Soil Processes and the Global Carbon Cycle Carbon isotope ratios were used to determine that fossil fuel burning increased atmospheric CO2 by Keeling et al, (1979). Now that this process is known it is important to study the potential carbon sinks. The realization that there was a large terrestrial carbon sink was brought about by studies in the early 1990s (Tans et al., 1990; Sternberge 1998). A signifi cant component of the carbon cycle is the terrestrial biosphere -atmosphere flux through soil systems, with 68 80 PgC/year soil respiration gl obally; soil respiration is the second largest flux of carbon between land and atmosphere (Reichstein et al, 2003). Among the annual total ecosystem respira tion, the soil respiration in forests makes up 30 80 % (Davidson et al., 2006). Soil systems globally contain approximately 1200 PgC in the first meter, in which the carbon is contained in the 2

PAGE 13

form of soil organic matter (SOM) (Amundson, 2004). Soils also generally contain between 380 and 60,000 ppm of CO2, depending on various soil properties and the amount of respiration (Amundson, 2004). The primary fluxes involved in the cycling of carbon in soils involve exchange with the atmosphere and with organic matter in th e surface litter. This study will isolate the flux of atmospheric exchange. The global flux of carbon into soils from surface litter is estimated at a rate of 4 PgC/year, the fl ux going from the soils to the atmosphere due to heterotrophic respiration is estimated at a rate of 3.5 Pg C/year; and there is an additional flux of carbon going from soils to the atmosphere due to anthropogenic land use, it is estimated at 0.4 PgC/year. This leaves 0.1 PgC/year accumulating in soils in the form of pedogenic carbonates (Amundson, 2004). This study is set up with th e intention of growing pedogenic carbonates with the ability to measure the isotopic values a nd concentrations of the involved soil CO2. Pedogenic Carbonates Pedogenic carbonates have a number of applications in paleoenvironmental studies, in a ddition to their relevance to soil CO2. Pedogenic carbonates typically form in arid to semi -humid environments where the soil pH is 7 or above (Cerling, 1984), and can be found in various forms. They begin as small crystals, and then form small nodul es; they can eventually form massive indurated horizons, if conditions perm it (Buck, 2005). It is important to 3

PAGE 14

remember that significant horizons of pe dogenic carbonates gene rally take greater than 100 years to accumulate, and that the carbonate precipitation may be seasonal (Blisnuk, 2005). Because pedogenic mineral accumulations occur over longer time spans they represent a timeaveraged isotopic composition (Blisniuk, 2005; Deutz, 2002), but that still may only represent a fraction of the time that it took for the host soil to develop (Deutz 2002). The isotopic compositions of pedogenic carbonates (CaCO3) in paleosols have been us ed in interpretations of paleoclimate and paleoelevation. Carbonate minerals can yield isotopic values for 13C and 18O, which can be used to interpret the relative proportions of C3 and C4 flora, and to determine the isotopic composition of meteoric water. This has been applied to study paleoclimate us ing carbonates from paleosols. The controls on oxygen isotopes in soil systems are less well understood than carbon, but it is known that oxygen isotope compositions of CO2, DIC, and pedogenic carbonate are related to me teoric water (Dworkin, 2005). The 18O of pedogenic carbonates will be affected by the 18O of rainwater; which itself varies with mean annual air temperature, such that as the temperature increases, so do the 18O values (Cerling, 1984; Schmid, 2006). Given that it does take a signifi cant amount of time to accumulate significantly observable pedogeni c carbonates nodules in soils, it is not expected that the soils in this study will produce pedogenic carbonate nodules within the time of this study, although it is expected that fine car bonate crystals will form, 4

PAGE 15

bridging gaps between soil particles. Even this small amount of carbonate precipitation is expected to affect the isotopic values of the soil CO2. Controls on Soil Respiration Soil respiration is the primary source of soil CO2. In natural soils, both root, or autotrophic respirat ion and heterotrophic respir ation (respiration from soil microbes, and other heterotrophs) are th e important components of total soil respiration. Soil respirati on rate varies considerably depending on a number of outside factors. Seasonal changes, the local climate, and the substrate can all affect soil respiration. Sp ecifically, the factors that influence soil respiration include CO2 concentration of the surrounding a tmosphere, and temperature, soil water availability, and plant growth rate s (Davidson et al., 2006). For example, Wan, S. et al. (2007) completed tests of so il respiration in old agricultural fields under atmospheric and enriched CO2 and found elevated soil respiration in the presence of elevated atmospheric CO2. This was believed to be fueled by increased plant growth. The study also found that there was no effect on soil respiration in the presence of elevated temp erature. However; that is not always the case, and this is a subject of much debate in the global change community. A study by Yutse et al. (2007) found that elevated temperature increased soil respiration in ponderosa pine forests. So the different ecosystems responded differently to the same stimulus in this case. 5

PAGE 16

Soil respiration can be affected by changing moisture content and decomposition rates as well (Kirschbaum, 2004). The study by Wan, et al (2007) also tested changing moisture levels in the old fields and found no correlation between moisture levels and soil respira tion. Once again, the Yutse et al. (2007) study showed differently. The ponderosa pine forests demonstrated no effect of moisture on soil respiration; however, they also analyzed an oak savanna ecosystem, and did find an increase in soil respiration with an increase in moisture, although the oak understory a nd the open savanna portions responded slightly differently. Adding to this a study by Pendall et al. (2004) found a decrease in soil respiration in the pr esence of the combination of increased moisture and increased atmospheric CO2 concentrations. Because soil respiration involves both plant roots and microbial co mmunities, it varies by ecosystem. By design of this study, many of these envir onmental controls on re spiration rate and processes can be eliminated: root respira tion, temperature, substrate and moisture content. By keeping these factors cons tant in laboratory conditions, it will simplify some of these contradictory issues and allow specific details of heterotrophic soil respiration to be an alyzed. Heterotrophic respiration is generally a significant compone nt of total soil respiration and has been reported from numerous field sites to comprise approximately 70 % of the total soil respiration (Buchmann, 2000). 6

PAGE 17

Stable Isotopes as a Tra cer of the Carbon Cycle As carbon moves between different Ea rth system reservoirs it undergoes isotopic fractionation. Thus, the ratio of 13C to 12C will be different in the different reservoirs. As photosynthesis moves carbon from the atmosphere into organic matter there is a very large fr actionation, the degree of which will differ according to the photosynthetic pathway. The most common photosynthetic pathways are C3 and C4. CAM (crassulace an acid metabolism) photosynthesis is used only by arid-adapted succulents and is not considered here because a comparatively small percentage of plants use this pathway (only about 10%), and because these plants utilize both C3 and C4 pathways, switching as needed (Clark and Fritz, 1997). C4 plants can acco unt for 16 30 % of total terrestrial photosynthesis (Ehleringer, 2002), and this fraction varies systemically with environmental conditions, and temporal ly on several time scales. This difference in the fractionation factor between photosynthetic pathways gives the SOM different starting isotopic values. The isotopic values are written in delta notation: 1000 CC CC-CC Cstandard 1213 standard 1213 sample 1213 13 (1) 7

PAGE 18

Soil with pure C3 biomass will have a 13C value around -25 whereas pure C4-derived SOM has 13C values of ~ -12 If the organic matter is of mixed composition, containing both C3and C4-derived organic matter, the 13C value varies proportionally, between -25 and 12 There is some variability in carbon isotope discrimination of C3 a nd C4 plants that corresponds to physiological and environmental differences (Ehleringer, 2002); but this is not significant enough to disrupt the non-ove rlapping trend between C3 and C4 plants. These differences in 13C values between photosynthetic pathways are reflected in the isotopic composition of CO2 respired by heterotrophic consumers of biomass. Both carbon and oxygen isotopic values will be measured on the respired soil CO2 in this study, each of which responds to different processes. The carbon isotopic signature derives prim arily from the decomposing organic matter, i.e. the source of respired C, while the oxygen isotopic signature derives from equilibration of CO2 with the local meteoric water. When rainwater infiltrates the so il it becomes soil water, and as such may undergo some fractionation, particularly as surface soil water evaporates and leaves behind 18Oenriched water in the soil. The soil wa ter will also equilibrate with the soil CO2, imparting a 18O signature to the soil CO2. CO2 that is dissolved in the soil water is the source of C for the precipitation of pedogenic carbonates. Some soil CO2 that remains in a gaseous form diffuses through the soil. During th is diffusion process the lighter isotopes 8

PAGE 19

move quicker, causing an enrichment of h eavier isotopes in th e soil (Tans, 1998). Diffusional fractionation in soils causes a 4.4 enrichment of 13C values Cerling, 1991). Equations 2 and 3 il lustrate the process by which soil CO2 is dissolved and is incorporat ed into pedogenic carbonate s (from Schlesinger, 1997). 2CO2 + 2H2O 2H+ + 2HCO3 (2) Ca2+ + 2HCO3 CaCO3 + H2O + CO2 (3) It can be seen from these reactions that there are a number of influential factors that may affect isotopic values of oxygen and car bon in soil systems. In this study soils constructed to replicate semi-arid grassland soils are used, and they were built with the intention of developing a Bk horizon (a subsurface illuvial horizon with an accumulation of carbonates). Soils in which carbon is accumulating in a developing Bk horizon may deplete soil CO2 of 13C during this process, because of the fr actionation factor between CO2 and carbonate. The fractionation factor ( ) is the change in 13C value that occurs with each step in the processes leading to the fractionation. As pedogenic carbonates are precipitated in soils from dissolved inorganic carbon (DIC ), the ultimate source of their carbon is the soil CO2, which is then dissolved in the soil water prior to utilization in the ca rbonate mineral (Figure 1). This occurs by the soil CO2 first dissolving from a gas in the soil to aqueous CO2, which entails a 1.1 fractionation factor. Aqueous CO2 is in equilibrium with bicarbonate (HCO3 -) with a 9.0 fractionation fact or between these two phase s. Equilibration with CO3 2has a -0.4 fractionation factor. The final step of precipitating CO3 2into 9

PAGE 20

CaCO3 comes with a 0.9 fractionation factor. Overall the equilibrium fractionation factor between soil CO2 gas and CaCO3 is 10.6 and varies slightly with temperature effects (Clark and Fritz, 1997, p. 120). Figure 1. Soil Carbonate Fractionation Diagram. This diagram from Clark and Fritz, 1997 (p. 120), illustrates the process of carbon from soil-respired CO2 becoming incorporated in pedogenic carbonate (CaCO3). The 13C of the carbon in each intermediate step is show n, as is the fractionation factor ( ). 10

PAGE 21

Keeling Plots Keeling Plots are essentially a twocomponent mixing model that allows the isotopic composition of source CO2 to be distinguished from soil CO2 that has undergone mixing with b ackground atmospheric CO2. The use of Keeling plots began in the late 1950s a nd early 1960s by C. D. Kee ling to interpret carbon isotope ratios of ambient CO2 and to identify the s ources that contribute to atmospheric CO2 concentrations on a regional basis. (Yakir, 2000). The work by Keeling hypothesized that carbon isotope ratios and CO2 concentration vary proportionally, and with a plot of 13C versus 1/[CO2] it was possible to see a process of the mixing of atmospheric CO2 and CO2 from soil and plant respiration (Keeling, 1958 and 1961). The y-intercept of a Keeling Plot represents the 13C of the source CO2. It was only in the late 1980s and 1990 s, that the use of Keeling plots for ecosystem respiration became widely known and was used to analyze the isotopic composition of atmospheric CO2 to determine carbon sources and sinks (Sternberge et al., 1998 ). Work that has been done using Keeling plots for soil respiration has generally been done in fiel d conditions, not in ar tificial soils in a laboratory. An example of this can be seen in Mortazavi et al., 2004; in which they utilized soil probes and sampling chambers to collect soil CO2 samples from 11

PAGE 22

natural soils. These were used to determine the 13C and 18O of soil efflux via the intercepts of the Keeli ng plots. There have been other works pertaining to soils, such as work by Liu et al., 2006, who studied the 13C and 14C in natural soilsand used the Keeling plot to de termine at what depths in the soil atmospheric CO2 had an effect and what depths it was no longer present. Cerling Model of Soil CO2 diffusion-production Cerlings diffusion-production model is utilized in studies of pedogenic carbonates and their ability to infer past C3/C4 flora compositions. The model utilizes a number of assumed input parameters (including atmospheric CO2 concentration, diffusion coefficient for CO2 in the air, free-air porosity, and a tortuosity factor). The diffusion-producti on equation is solved for steady state conditions, which allows the model calculation of CO2 concentration profiles in soils (equation 4). The model goes further to predict the 13CCO2 throughout the soil profile, but it assumes the same soil diffusion coefficient throughout the profile. The model solves the diffu sion-production equation separately for isotopically substituted molecules of CO2, and then predicts the ratio of 13C/12C in soil CO2 which is converted to in equation 5. 2* *0 2C z Lz D Cs s (4) 12

PAGE 23

10001 1* 1 2* * 2* 10 2 0 2 a s s s a s s s PDB sC D D z Lz D C D D z Lz D R (5) Cs* is the concentration of soil CO2, s* is the permil value for soil air, RPDB is the ratio of 13C/12C in PDB standard, Ds* is the diffusion coefficient for CO2 in soil, is the production rate of CO2, L is the depth at the base of the soil profile, z is the depth within the soil profile, is the permil value for respired CO2, is the permil value for atmosphere, Ds is the diffusion coefficient of 13CO2 in soil, and C0* is the atmospheric concentration of CO2. aResults of this model from Cerling ar e shown in Figures 2 4, with some standard assumptions. The diffusionproduction model demonstrates that atmospheric CO2 concentration is important at sh allow depth (the first 10 cm in the soil from the surface) and when soil respiration value is low because it mixes significantly with soil-respired CO2 and imparts an isotopic signature on the soil CO2 indicative of this mixing ratio (Cerling, 1984). Predicted CO2 concentration with depth is shown in Figure 2; predicted 13C with depth is illustrated in Figure 3, and a Keeling Plot made using these predicted values, is shown in Figure 4. 13

PAGE 24

Cerling Model CO2 vs. Depth0 10 20 30 40 50 60 70 80 90 100 05000100001500020000 [CO2], ppm VDepth, cm Figure 2. CO2 versus depth using the Cerling model for CO2 diffusion production. Model parameters of the ar tificial soils in this study and the atmospheric CO2 value for the lab atmosphere at the time of the study were input into Cerlings model (using the methods of Amundson, 2004). 14

PAGE 25

Cerling Model 13C vs. Depth0 10 20 30 40 50 60 70 80 90 100 -35.0-25.0-15.0 13C CO2, Depth, cm Figure 3. 13C versus depth using th e Cerling model for CO2 diffusion production. Model parameters of the ar tificial soils in this study and the atmospheric CO2 value for the lab atmosphere at the time of the study were input into Cerlings model (using the methods of Amundson, 2004). 15

PAGE 26

Cerling Model Keeling Plot -25.0 -23.0 -21.0 -19.0 -17.0 -15.0 -13.0 -11.0 -9.0 -7.0 -5.0 0.00000.00050.00100.00150.00200.00250.0030 1/[CO2]13C Figure 4. Keeling Plot constructed with the values predicted by the Cerling CO2 diffusion production model. The yintercept represents the source 13C, which is 23.4 in this model. 16

PAGE 27

Expectations and Hypotheses Because soil CO2 equilibrates with soil solution to form pedogenic carbonates and is the source of ecosystem respiration, it is important to fully understand the effect of mixing on the isotopic values of soil CO2, and how it is influenced by carbonate precipita tion and background atmospheric CO2. This study uses laboratory analysis of artificial soils to test th e isotopic values in soils with either C3 or C4 organic matter, at various depths to determine the influence on the isotopic values of soil CO2 by background atmospheric mixing and carbonate precipitation. It is expected th at there will be 4.4 fractionation factor for the CO2 diffusion through air, from a source of soil-respired CO2. The soil CO2 concentration is expected to incr ease with depth, and the carbon isotopic signatures are expected to reflect the va lues of the starting soil organic matter mixed with atmospheric CO2, whereas the oxygen isotopic signatures are expected to reflect equilibration with the water used to irrigate the columns. The diffusivity of soil CO2 will be dependent on the connectivity of pore space in the soil, and is thus affected by so il type and by the presence of water. If the pore space in a soil is saturated, the diffusion of gas through the soil may be halted (Weerst, 2001). This study will also test the effect of soil moisture on the diffusivity of CO2 under variable degrees of saturation. 17

PAGE 28

The Keeling Plots are expected to re present the process of mixing between a homogenous source of respired CO2 and atmospheric CO2. If this is the only process involved, the Keeling pl ots should be linear for both 13C and 18O; any deviation from a linear plot will illuminate processes that fractionate 13C/12C or 18O/16O during the production, diffusion, and equilibration. Materials and Methods Replicate soil columns were built with two types of artificial SOM profiles composed of C3 and C4 biomass in sterile mineral soil. By using artificial soils in laboratory conditions, the effects of root respiration have been removed, leaving microbial decomposition of organic ma tter and evaporation as the primary variables that control th e isotopic composition of soil respiration and any precipitation of pedogenic carbonate. Three re plicates were built of two types of soil profiles each containing a homogenized grass litter (from C3 or C4 grasses) in a matrix of clean organic-free sand. Profiles were built with fresh grass clippings as the starting organic matt er to simulate decomposition and CO2 concentration that would be typical of a natural grassland soil. In a typical grassland soil with a mean respiration flux rate (Fresp) = 442 g C m-2 yr-1, the CO2 concentration increases from atmospheric CO2 concentration at the surface (currently 384 ppm) to about 7,000 ppm at 100 cm depth (Amundson, 2005). 18

PAGE 29

Periodically, an isotopically-homogenous source of water with 200 mg/L Ca2+ was added to each profile in order to s timulate both decomposition of the organic matter and precipitation of pedogenic carbona te as the water was evaporated and calcite saturation was reached. The conc entration and isotopic composition of CO2 within the soil profiles were then monitored over ~6 months. Apparatus The soil profiles were contained in si x columns that were built using 2.5inch clear PVC pipe, each pipe measuring 101 cm in length. The columns were measured and marked for the horizons as follows: From the top, 0 2.5 cm was to be left empty for watering space; 2 .5 12.5 cm was marked for the A1 horizon; 12.5 42.5 cm for the A2 horizon (A horizons are surface horizons that contain humified organic matter and minerals); 42.5 62.5 for the E horizon (E horizons are eluvial horizons th at have been leached of organi cs and minerals such as clay, iron, and aluminum); 62.5 92.5 cm for the Bk horizon (Bk horizons are subsurface soil horizons which accumulate leached materials from the surface horizons and feature the preci pitation of carbonate s within them); with the bottom remaining space for the C horizon (unweathered material beneath the soilin this case it is used to assist in draining th e columns). Figure 5 illustrates the soil column apparatus, with horizons and th e horizon boundary depth in cm. Table 1 contains the measurements for soil horizons in the columns. 19

PAGE 30

After marking the horizons, holes were drilled (very slowly with a 13/16 Speedbor woodboring bit) in the columns to allow for the installation of the gas sampling tubes. A1, A2, and E horizons each had one gas sampling tube installed in the center of the horizon, and the Bk horizon had two gas sampling tubes (Figure 5). 20

PAGE 31

Figure 5. Apparatus. Diagram of soil column appa ratus with measurements. The horizons are labeled. Columns are made from clear PVC with PVC and CPVC fittings. Gas collection tubes are made with perforated stainless steel tubes. 21

PAGE 32

No gas sampling tubes were installed in the C horizon. The gas sampling tubes were installed to provi de a volume within the soil horizon for collection of CO2 that is larger than the volume of ga s removed during sampling. This is to avoid creating a pressure gradient that may pull CO2 through the soil during sampling, which could possibly cause a slight isotopic change as the 12C will diffuse through the soil slightly faster than the 13C. A Dremel tool with drum sanding bit was used to smooth the e dges of the holes that were cut. Figure 6. Gas Collection Tube. Thirty of these tubes were constructed from stainless steel mesh and perforated stainless steel sheet metal. Their purpose was to allow CO2 to collect in the space in a greater volume than would be pulled from the collection tube during sampling. The gas sampling tubes were made usi ng perforated stai nless steel sheet (5/16 inch diameter holes, centers 3/8 inches apart; st aggered array of holes) and using a hammer to form it around an oak dowel (1/2 inch diameter). Then a fine stainless steel mesh (sst 316, 70 x 70 0.0037 inches) was cut into slightly smaller rectangles, formed into tubes around an i nk pen, and inserted into the perforated 22

PAGE 33

tubes (this is to keep sand out of the t ubes, but to allow gas to exchange between the sampling tubes and pore spaces in the soil). All of the tubes were washed with Nalgene detergent to remove greas e and then they were fitted into the columns with CPVC fittings to hold them in place. After removing the black rubber rings from the FIP adapters, septa were added (headspace septa, Teflon/rubber; C4020-34, www.nationalscienti fic.com) to all of the adapters, which were then screwed on the CPVC 1/ 2 MIP adapters. The CPVC fittings have rubber septa added to seal the colu mns from atmospheric gas and to allow the gas sampling needle to remove gas samples from the tubes. 2.5 inch slip cap PVC endcaps we re attached to the bottom of each column. In the bottom of each endcap a 3/16 inch hole was drilled. Each of these holes th en had a 1.5 inch length of 3/16 thinwall rigid airline tubing a dded for drainage of excess pore water. A length of flexible airline tubing was then added to each to allow drainage to specific containers. The completed columns are held upright on a monkey lattice and rest on cut pieces of 3 inch PVC pipes. An empty completed column is shown in Figure 7. Figure 7. Completed soil column. Perforated stainless t ubes used to collect and sample soil CO2, via syringe and rubber septa, are visible through the clear PVC. 23

PAGE 34

Horizon Top Depth (cm) Bottom Depth (cm) Thickness (cm) Depth of Sampling Tubes (cm) A1 2.5 12.5 10 7.5 A2 12.5 42.5 30 27.5 E 42.5 62.5 20 52.5 Bk 62.5 92.5 30 72.5 (Bk1) 82.5 (Bk2) Table 1. Apparatus Measurements. This table contains the measurements for soil horizons and gas sampling points within the soil columns. Soil Profiles Soil horizons were built within the columns to simulate a simplified grassland soil profile in which th e dominant processes were aerobic decomposition of the organic matter, eva potranspiration of the pore water, and precipitation of soil carbonate. The C horiz ons were filled to the base of the Bk horizons with quartz sand (Quartz play sand from Home Depot) and crushed scoria to improve drainage from the base of the profile. The Bk horizons were then added, each one has 750g Ca-montmorilinite (smectite) clay and 1187g quartz sand. The Bk horizon contained 38.7% smectitic clay to retain Ca2+concentrated water during ev aporation from the surface. Next the E horizons were added, each one has pure quartz sand, enough was added to fill that segment; its mass was not measured, but a smaller porti on of the sand was measured to obtain the bulk density. This was done by measuring 550 cm3 of the quartz and obtaining the mass for that volume (the mass was 906.6g). Each A2 horizon was 24

PAGE 35

filled with a mixture of gr ass and quartz sand, resulting in 5.0% organic matter. Each A1 horizon was filled with a mixture of grass and quart z sand resulting in 6.2% organic matter. The organic matter in three of the columns consisted of a homogenous C4 grass biomass, collected fr om grass clippings from the greens of a golf course in Tampa, Florida (Temple Terrace Golf & Country Club); and three of the columns used a relatively homogenous C3 grass biomass, collected from a golf course in St. Andrews, Scotland (the Old Course). Grass was ground in a coffee grinder prior to mixing with th e sand to obtain a relatively homogenous consistency and partic le size, equivalent for both profiles. The grass biomass had starting isotopic values of 29.09 0.31 for the 13C of the C3 grass, and -13.77 0.19 for the 13C of the C4 grass. Predicted 13CCO2(VPDB) of soil CO2 respired from these substrates are: -24.7 (C3) and -9.4. Two of the filled columns are shown in Figure 8. The bulk density of each horizon is isplayed in Table 2. The centralized light-colored horizon is the E horizon. The developing Bk horizon is below the E horizon, and the A horizons are above the E horizon. dFigure 8. Soil Profile. Two of the completed columns with the soil profiles 25

PAGE 36

Horizon Volume (cm3) Mass (g) Bulk Density (g/cm3) Sand (%) Organic Matter (%) Clay (%) A1 301.9 395.6 1.31 93.8 6.2 0 A2 905.7 1186.9 1.31 95.0 5.0 0 E 603.8 N/A 1.65 100 0 0 Bk 905.7 1937.0 2.14 61.3 0 38.7 Table 2. Soil Profile Parameters. Bulk density and other parameters of the soil horizons. Column volumes and the mass of added materials for each horizon are given in the table. Water Addition The water used for watering the columns was prepared with the intention of inducing carbonate precipi tation within the Bk horizon and with providing a distinct isotopic signature from the wate r that could be visi ble in any resulting carbonates. Two clean jugs, one 20 L and one 50 L were filled with 12 M deionized water, and placed open under a fu me hood to allow some evaporation (to obtain slightly 18O-enriched water). After about three months of evaporation time the water was all (about 60 L) placed in a large Rubbermaid bin and 22.2 g of calcium hydroxide was added in or der to charge the water with Ca2+. Because the resulting pH was higher than typical for grasslands, HCl was added to bring down the pH to 6.48, to approximate the pH of slightly buffered rainwater. This water 26

PAGE 37

was placed back in the jugs and sealed to prevent further evaporation. The starting isotopic value for the water was 18O = -1.97 0.07 Predicted 18OCO2(VSMOW) of CO2 in oxygen isotopic equilibrium with this water at 25C is +28.8. Soil columns were watered re gularly with either 175 ml or 650 ml volumes of water; the light watering was intended to wet the profile, and the large watering to flush (see Appendix D for the wa tering chart). To assist in drainage of the excess pore water, column drainage tubes were attached to a peristaltic pump the day after watering for about 1 hour. Airline tube check valves from Tetra are used to prevent atmospheric gas from entering the columns via the drainage tubes during this pumping. 40 W heating lamps were placed on a 12-hr timer above each profile to encourage ev aporation of pore water from the surface of the profile. After approximately six months of use it was determined that the peristaltic pump drainage system was not sufficient to drain the columns. To remedy this situation 1 inch diameter holes were drilled with a hole saw and 1 inch CPVC slip threaded bushings were added. These were sealed with 1 inch PVC threaded caps. Thereafter, subsequent to each watering a ceramic pore water sampler (SG25 porous borosilicate from UMS, Munich, Germany) was inserted with a tube extending from it to a sealed vacuum flask. The flask was kept under vacuum to pull water from the soil columns for approximately 1 hour the day after watering. 27

PAGE 38

Gas Sampling Gas samples were collected for isotope ratio mass spectrometer (IRMS) analysis by removing a small volume of gas from the gas sampling tube with a gas-tight syringe (Hamilton, or Bee-Stinger). The gas-tight syringe was inserted through the rubber septa into the perforate d steel gas collection tubes within the columns. Gas was always sampled from the top to the bottom of each column for consistency in the following pattern: A1 first, A2 second, E third, Bk1 fourth and Bk2 last. The columns were always sampled in the following order: StA1 first, StA2 second, StA3 third, T1 fourth, T2 fi fth, and finally T3 (StA referring to C3 biomass columns and T referring to C4 biomass columns). From each column, 7.5 ml of gas was taken from the A1 horiz ons, and all other horizons had 2 ml of gas taken. The gas samples from the soil columns were then injected from the Gas-tight syringe into Helium-flushed 10 mL Exetainers (Labco, Hertfordshire, UK) in preparation for IRMS analysis. Three samples of ambient lab atmosphere were collected during every sampling session. These were collected by leaving three Exetainers open for several minutes and then sealing them. Samples were analyzed within two hours to avoid leak age from the Exetainer septa. Gas samples for the three IRMS runs used in this study were taken at different time intervals after the columns had been wa tered. One sampling date was 2 days 28

PAGE 39

post-watering, one sampling run was 25 da ys post-watering, and one sampling run was 64 days post-watering. This provided the opportunity to measure the soil respiration in the columns when the soil was wet, moist, and dry. IRMS Analysis IRMS runs were set up as follows: 30 vials were flushed with Helium and were used for the samples from the columns; 6 vials flushed with CO2/He were run as standards with the samples. 12 vials were flushed with Helium and had varying increments (4 with 7.5 ml, 4 with 5 ml, and 4 with 2 ml) of a custom CO2/He gas mixture (2987 ppm CO2 in balance of He). The latter samples were added in order to correct sample analysis (which vary signific antly in amount of CO2) for linearity effects of the mass spect rometer. Samples of lab atmosphere were added at the end of a run. [CO2], 13CCO2 and 18OCO2 were measured on a Finnigan Delta V IRMS and corrected with respect to a CO2/He standard which was calibrated to the VPDB scale via the international NBS-18 and NBS-19 reference materials (limestones). Normalization was as follows: CO2 samples were measured with respect to replicates of a CO2/He mixture, which was calibrated to the VPDB scale via NB S-18 and NBS-19. Concentration of CO2 was measured for each gas sample by comparison of an equivalent sample of the CO2/He gas mixture (2967 ppm CO2 in balance He). For comparability of water 29

PAGE 40

and CO2 data 18O (VPDB) values were converted to 18O (VSMOW) via the equation of Coplen, et al. (1983): 18OVSMOW = 1.03091 18OPDB + 30.91 (6) A predicted 18O value for CO2 in equilibrium with the soil water was determined by the equilibrium fractionation factor in ONeil & Adami (1969). Modeled 13CCO2 values follow the diffusion-producti on model of Cerling (1984). Plots Keeling Plots and all other plots were created in Microsoft Excel XP Pro. The Keeling plots were created by plotting the corrected IRMS results versus the reciprocal of the CO2 concentration, and then appl ying Excels linear regression function to create a trend line which repres ents the Keeling li ne. This was done for each horizon, C3 columns together and C4 columns togetherresulting in 4 Keeling Plots (two for 13C and two for 18O). Additional plots were made for each sampling day, to illustrate any differences that different soil moisture conditions may have on CO2 concentration or on the isotopic values of the soil CO2. 30

PAGE 41

Results CO2 Concentration with Depth The CO2 concentration was measured for each horizon, and the average for each horizon plotted versus the column depth (Figure 9). Soil CO2 concentration was greater than atmospheric CO2 concentration (380 ppm) in all horizons during all sampling intervals. Average CO2 concentration increased from near atmospheric values up to nearly 12000 ppm between 70 and 80 cm depth (note this is only an average for replicate profiles and multiple sampling intervalsactual values taken at time of sampling were higher/lower depending on the soil moisture). Figures 11 12 contain additional CO2 versus depth plots separated according to one of three moisture regimes, which the sampling occurred in wet, moist, or dry soil. Average CO2 concentration increased from the A1 horizon to the A2 horizon, and remained fairly steady through the E horizon for the C4 columns, and continued to increase in the C3 columns. C3 columns had a continued increase in CO2 concentration through the Bk1 sampling point, and then decreased at the Bk2 sampling point. C4 columns decreased at the Bk1 sampling point. The C4 columns only had one sample from the Bk2 sampling points, which showed a significantly greater CO2 concentration. 31

PAGE 42

32 CO2 Concentration vs. Depth0 10 20 30 40 50 60 70 80 90 02000400060008000100001200014000[CO2], ( ppm)Depth (cm) A1 A2 E Bk1 Bk1 Bk2 Figure 9. CO2 Concentration versus Depth. Average of [CO2] depth profiles. Sampling points were in the center of each horizon for A1, A2, and E horizons. The Bk horizon contained two sampling points (note that the lo wer Bk horizon sample was often problematic with extremely high [CO2], likely due to low porosity. The gray triangle is the single sample point that was obtained for the Bk2 horizon in the C4 columns. Catm = 380 ppm

PAGE 43

CO2 Concentration vs. Depth 08 09 2008 Wet Soil0 10 20 30 40 50 60 70 80 90 02000400060008 000100001200014000 [CO2] (ppm)Depth (cm) C3 C4A1 A2 E Bk1 Figure 10. Concentration versus Depth in Wet Soil. [CO2] depth profiles for a single sampling time, 2 days after the co lumns were irrigated with 650 ml of water per column. Sampling points were in the cente r of each horizon for A1, A2, and E horizons. The Bk horizon contained two sampling points (note that the lower Bk horizon sample was often pr oblematic with extremely high [CO2], likely due to low porosity). 33

PAGE 44

CO2 Concentration vs. Depth 08 01 2008 Moist Soil0 10 20 30 40 50 60 70 80 90 02000400060008000100001200014000 [CO2] (ppm)Depth (cm) C3 C4A1 A2 E Bk1 Figure 11. Concentration versus Depth in Moist Soil. [CO2] depth profiles for a single sampling time, 25 days after the columns were irrigated with 650 ml of water per column. Sampling points were in the cente r of each horizon for A1, A2, and E horizons. The Bk horizon contained two sampling points (note that the lower Bk horizon sample was often pr oblematic with extremely high [CO2], likely due to low porosity). 34

PAGE 45

CO2 Concentration vs. Depth 10 12 2008 Dry Soil0 10 20 30 40 50 60 70 80 90 02000400060008000100001200014000 [CO2] (ppm)Depth (cm) C3 C4A1 A2 E Bk1 Bk1 Bk2 Figure 12. Concentration versus Depth in Dry Soil. [CO2] depth profiles for a single sampling time, 64 days after the co lumns were irrigated with 650 ml of water per column. Sampling points were in the cen ter of each horizon for A1, A2, and E horizons. The Bk horizon contained two sampling points (note that the lower Bk horizon sample was often pr oblematic with extremely high [CO2], likely due to low porosity). 35

PAGE 46

13C Keeling Plots Keeling Plots separated for each horizon type but averaged for all sampling periods are shown in Figure 13 and 14, which highlights changes in slope and y-intercept between horizons. Figure 13 shows the 13C Keeling Plots for the C3 columns and Figure 14 shows the Keeling Plots for the C4 columns. In both cases it is clear that the 13C values determined by the y-intercepts on the Keeling Plots ( 13C(KP)) decrease with depth. In all of the 13C Keeling Plots the A and E horizons fit a line well, but th e B horizons do not. The y-intercepts become more 13C-depleted in the B horizons. Keeling Plots shown here consist of all of the data combined from the three gas-sampling days, while Figures 15 20 contain Keeling Plots for each sampling period. These are organized into one of the three moisture regimes during which the sampling occurred: wet, moist, and dry. Variation in the different moisture regimes produced negligible difference in the 13C(KP), for both the C3 and C4 soil profiles. The Equations and R2 values for the average Ke eling Plot lines are listed in Table 3. The y-intercepts for individua l Keeling Plots (in Figures 15 20) that were plotted according to the approximate soil moisture regime are in Table 4. It is important to notice that there is not significant change in y-intercept between the different moisture regimes. 36

PAGE 47

C3 13C Keeling Plots-45.0 -40.0 -35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.00301/[CO2]13C (o/oo) Lab Atmosphere Catm = 380 ppmPredicted 13C of soil CO2 is -24.7o/oo. Bk 2 Horizons Bk 1 Horizons E Horizons A2 Horizons A1 Horizons Figure 13. C3 13C Keeling Plots. Keeling Plots of 13C by horizon for C3 columns (plots summarize 5 months of sampling; variation between sampling interval s is evident). The values for all of the replicate columns are plotted as point s and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Notice that the Keeling plot intercepts become more negative for deeper soil horiz ons (Bk horizon). Predicted 13C of soil CO2 from this C3 biomass is -24.7 The starting 13C of the C3 biomass was -29.09 0.31 37

PAGE 48

C4 13C Keeling Plots-35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.00301/[CO2]13C (o/oo) Lab AtmosphereCatm = 380 ppmPredicted 13C of soil CO2 is -9.4o/oo. Bk 1 Horizons E Horizons A2 Horizons A1 Horizons Figure 14. C4 13C Keeling Plots. Keeling Plots of 13C by horizon for C4 columns (plots summarize 5 months of sampling; variation between samp ling intervals is evident). The values for all of the repl icate columns are plotted as points and the Keeling Plot lines were plotted using linear regression for all samp ling intervals. Notice that the Keeling plot intercepts become more negative for the Bk horizon. The predicted 13C of soil CO2 from the C4 biomass is -9.4 The starting 13C of the C4 biomass was -13.77 0.19 38

PAGE 49

C3 Keeling Plot Line Equations Horizon Equation R2 Value A1 y = 5690x 24.3 0.9 A2 y = 5914x 24.7 1.0 E y = 6059x 25.1 1.0 Bk1 y = 8602x 31.7 1.0 Bk2 y = 11902x 40.6 0.9 C4 Keeling Plot Line Equations Horizon Equation R2 Value A1 y = 709x 11.2 0.3 A2 y = 615x 10.8 0.4 E y = 751x 11.2 0.5 Bk1 y = 3473x 18.4 0.9 Bk2 N/A N/A Table 3. 13C Keeling Plot Line Equations and R2 Values Equations and R2 values for the Keeling plot lines in Figures 9 and 10. 39

PAGE 50

C3 Keeling Plots 08 09 2008-35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 00.00050.0010.00150.0020.00250.003 1/[CO2]13C (VPDB, o/oo ) A1 A2 E Bk1 Lab Atmosphere Figure 15. C3 13C Keeling Plots for Wet Soil Conditions. Keeling Plots of 13C by horizon for C3 columns; sampling was done 2 days after watering the columns with 650 ml of prepared water per column. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Predicted 13C of soil CO2 from this C3 biomass is -24.7 The starting 13C of the C3 biomass was -29.09 0.31 40

PAGE 51

C3 Keeling Plots 08 01 2008-35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 00.00050.0010.00150.0020.00250.003 1/[CO2]13C (VPDB, o/oo) A1 A2 E Bk1 Lab Atmosphere Figure 16. C3 13C Keeling Plots for Moist Soil Conditions. Keeling Plots of 13C by horizon for C3 columns; sampling was done 25 days after watering the columns with 650 ml of prepared water per column. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Predicted 13C of soil CO2 from this C3 biomass is -24.7 The starting 13C of the C3 biomass was -29.09 0.31 41

PAGE 52

C3 Keeling Plots 10 12 2008-45.0 -40.0 -35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 0.00000.00050.00100.00150.00200.00250.0030 1/[CO2]13C (VPDB, o/oo) A1 A2 E Bk1 Bk2 Lab Atmosphere Figure 17. C3 13C Keeling Plots in Dry Soil Conditions. Keeling Plots of 13C by horizon for C3 columns; sampling was done 64 days after watering the columns with 650 ml of prepared water per column. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Predicted 13C of soil CO2 from this C3 biomass is -24.7 The starting 13C of the C3 biomass was -29.09 0.31 42

PAGE 53

C4 Keeling Plots 08 09 2008-35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 00.00050.0010.00150.0020.00250.003 1/[CO2]13C (VPDB, o/oo) A1 A2 E Bk1 Lab Atmosphere Figure 18. C4 13C Keeling Plots in Wet Soil Conditions. Keeling Plots of 13C by horizon for C4 columns from sampling 2 days post watering. Each column was irrigated with 650 ml of prepared water. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Notice that the Keeling plot intercepts become more negative for the Bk horizon. The predicted 13C of soil CO2 from the C4 biomass is -9.4 The starting 13C of the C4 biomass was -13.77 0.19 43

PAGE 54

C4 Keeling Plots 08 01 2008-35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 00.00050.0010.00150.0020.00250.003 1/[CO2]13C (VPDB, o/oo) A1 A2 E Bk1 Bk2 Lab Atmosphere Figure 19. C4 13C Keeling Plots in Moist Soil Conditions. Keeling Plots of 13C by horizon for C4 columns from sampling 25 days post watering. Each column was irrigated with 650 ml of prepared water. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Notice that the Keeling plot intercepts become more negative for the Bk horizon. The predicted 13C of soil CO2 from the C4 biomass is -9.4 The starting 13C of the C4 biomass was -13.77 0.19 44

PAGE 55

C4 Keeling Plots 10 12 2008-35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.0 0.00000.00050.00100.00150.00200.00250.0030 1/[CO2]13C (VPDB, o/oo) A1 A2 E Bk1 Lab Atmosphere Figure 20. C4 13C Keeling Plots in Dry Soil Conditions. Keeling Plots of 13C by horizon for C4 columns from sampling 64 days post watering. Each column was irrigated with 650 ml of prepared water. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sampling intervals. Notice that the Keeling plot intercepts become more negative for the Bk horizon. The predicted 13C of soil CO2 from the C4 biomass is -9.4 The starting 13C of the C4 biomass was -13.77 0.19 45

PAGE 56

C3 13C y-intercepts for Different Soil Moisture Conditions. Horizon Wet Moist Dry A1 -25.4 -23.2 -26.8 A2 -25.0 -24.3 -25.6 E -25.4 -24.7 -25.8 Bk1 -33.9 -30.4 -32.0 Bk2 N/A N/A -40.0 C4 13C y-intercepts for Different Soil Moisture Conditions. Horizon Wet Moist Dry A1 -10.6 -10.7 -13.1 A2 -10.9 -10.5 -11.6 E -11.3 -10.6 -12.0 Bk1 -17.1 -18.3 -19.8 Bk2 N/A -14.2 N/A Table 4. 13C Keeling Plot y-Intercepts. 13C(KP) under different soil moisture conditions. 18O Keeling Plots 18O Keeling Plots are shown in Figure 21 and Figure 22 separated by horizon type but arranged over three samp ling periods. Both the C3 and C4 columns show the surface (A1 hor izon) to have a more positive 18O(KP) value than the rest of the horizons for all runs combined. The plots here contain all data points from the sampling during the three different moisture regimes, and thus represent an average value for the soil profiles over time (which in the natural world consists of dry periods and interspers ed rain events that temporarily wet the soil). Individual plots separated by each mo isture regime are in Figures 23 28. 46

PAGE 57

In these plots it is evident that the A1 horizon 18O composition is strongly influenced by the soil moisture. 18O(KP) in A1 horizons in wet soil was enriched compared to other horizons; whereas in dry conditions the A1 horizons showed little to no distinction from the other horiz ons. This is evident in both C3 and C4 profiles, but is most prom inent in the C3 profiles. 47

PAGE 58

Figure 21. C3 18O Keeling Plots Keeling Plots of 18O by horizon for C3 columns (plots summarize 5 months of sampling as a bove). The values for all of the replicate columns are plotted as points and the Keeling Plot lines were plotted using linear regression for all sample intervals. Noti ce that the Keeling Plot intercept for the A1 horizon is more positive than the others. 48 C3 18O Keeling Plots10.0 15.0 20.0 25.0 30.0 35.0 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.00301/[CO2]18O (VSMOW, o/oo) Lab Atmosphere Catm = 380 ppm Bk 2 Horizons Bk 1 Horizons E Horizons A2 Horizons A1 Horizons

PAGE 59

Figure 22. C4 18O Keeling Plots Keeling Plots of 18O by horizon for C4 columns (plots summarize 5 months of sampling as a bove). The values for all of the replicate columns are plotted as points and the Keeling Plot lines were plotted using linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. C4 18O Keeling Plots10.0 15.0 20.0 25.0 30.0 35.0 40.0 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.00301/[CO2]18O (VSMOW, o/oo) Lab AtmosphereCatm = 380 ppm Bk 1 Horizons E Horizons A2 Horizons A1 Horizons 49

PAGE 60

C3 Keeling Plot Line Equations Horizon Equation R2 Value A1 y = -3758x + 34.1 0.7 A2 y = -1643x + 28.6 0.9 E y = -1508x + 28.2 0.8 Bk1 y = -1526x + 28.2 0.8 Bk2 y = -797x + 26.3 0.4 C4 Keeling Plot Line Equations Horizon Equation R2 Value A1 y = -4536x + 36.4 0.7 A2 y = -1535x + 28.3 0.8 E y = -1431x + 28.0 0.8 Bk1 y = -830x + 26.4 0.5 Bk2 N/A N/A Table 5. 18O Keeling Plot Line Equations and R2 Values. Equations and R2 values for the Keeling Plot lines in Figures 11 and 12. 50

PAGE 61

C3 Keeling Plots 08 09 200820.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 38.0 00.00050.0010.00150.0020.00250.003 1/[CO2]18O (VSMOW, o/oo) A1 A2 E Bk1 Lab Atmosphere Figure 23. C3 18O Keeling Plots in Wet Soil Conditions. Keeling Plots of 18O by horizon for C3 columns. Samples were taken 2 days after irrigating the columns with 650 ml of prepared water per column. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. 51

PAGE 62

C3 Keeling Plots 08 01 200820.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 00.00050.0010.00150.0020.00250.003 1/[CO2]18O (VSMOW, o/oo) A1 A2 E Bk1 Bk2 Lab Atmosphere Figure 24. C3 18O Keeling Plots in Moist Soil Conditions. Keeling Plots of 18O by horizon for C3 columns. Samples were taken 25 days after irrigating the column s with 650 ml of prepared water per column. The values for all of the replicate columns are plotted as points and the Keeling Plot lines were plotted us ing linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. 52

PAGE 63

C3 Keeling Plots 10 12 200820.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 0.00000.00050.00100.00150.00200.00250.0030 1/[CO2]18O (VSMOW, o/oo) A1 A2 E Bk1 Bk2 Lab Atmosphere Figure 25. C3 18O Keeling Plots in Dry Soil Conditions. Keeling Plots of 18O by horizon for C3 columns. Samples were taken 64 days after irrigating the column s with 650 ml of prepared water per column. The values for all of the replicate columns are plotted as points and the Keeling Plot lines were plotted us ing linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. 53

PAGE 64

C4 Keeling Plots 08 09 200822.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0 00.00050.0010.00150.0020.00250.003 1/[CO2]18O (VSMOW, o/oo) A1 A2 E Bk1 Lab Atmosphere Figure 26. C4 18O Keeling Plots in Wet Soil Conditions. Keeling Plots of 18O by horizon for C4 columns. Samples were taken 2 days after irrigating the columns with 650 ml of prepared water per column. The values for all of the replicate column s are plotted as points and the Keeling Plot lines were plotted using linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. 54

PAGE 65

C4 Keeling Plots 08 01 200820.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 38.0 00.00050.0010.00150.0020.00250.003 1/[CO2]18O (VSMOW, o/oo) A1 A2 E Bk1 Bk2 Lab Atmosphere Figure 27. C4 18O Keeling Plots in Moist Soil Conditions. Keeling Plots of 18O by horizon for C4 columns. Samples were taken 25 days after irrigating the column s with 650 ml of prepared water per column. The values for all of the replicate columns are plotted as points and the Keeling Plot lines were plotted us ing linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. 55

PAGE 66

C4 Keeling Plots 10 12 200820.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 0.00000.00050.00100.00150.00200.00250.0030 1/[CO2]18O (VSMOW, o/oo) A1 A2 E Bk1 Lab Atmosphere Figure 28. C4 18O Keeling Plots in Dry Soil Conditions. Keeling Plots of 18O by horizon for C4 columns. Samples were taken 64 days after irrigating the column s with 650 ml of prepared water per column. The values for all of the replicate columns are plotted as points and the Keeling Plot lines were plotted us ing linear regression for all sample intervals. Notice that the Keeling Plot intercept for the A1 horizon is more positive than the others. 56

PAGE 67

C3 18O y-intercepts for Different Soil Moisture Conditions. Horizon Wet Moist Dry A1 +37.3 +35.3 +30.0 A2 +28.0 +28.7 +29.1 E +28.1 +28.7 +27.7 Bk1 +28.8 +28.1 +28.1 Bk2 N/A +26.0 +26.7 C4 18O y-intercepts for Different Soil Moisture Conditions. Horizon Wet Moist Dry A1 +42.8 +37.0 +32.1 A2 +27.5 +28.6 +28.6 E +27.5 +28.9 +27.5 Bk1 +27.0 +26.7 +25.9 Bk2 N/A +29.3 N/A Table 6. 18O Keeling Plot y-Intercepts. 18O(KP) under different moisture conditions. 13C and 18O with Depth To better illustrate the change in isot opic values seen in the Keeling Plots, the raw 13C and 18O data was plotted versus depth. The C3 columns have 13C values that deviate significantly from the predicted equilibrium soil-respired CO2 values in both the A1 horizons and the Bk horizons. The C4 columns have 13C values that deviate from the predicte d equilibrium (determined by adding 4.4 diffusional fractionation to the 13C values of the starting materials) only at the 57

PAGE 68

Bk horizons, but the A1 horizons remain close to the predicted equilibrium values. For the 18O versus depth (Figure 30), both the C3 and C4 columns deviate from the predicted equilibrium values in both the A1 and Bk horizons. The results in the isotopic compositi on versus depth plots (Figures 29 and 30) and the results in the Keeling Plots (Figures 13, 14, 21, and 22) illustrate the same trends, simply in different ways. The difference mainly being that the Keeling Plots allow the isotopic value that would be present with no atmospheric mixing to become known; the isotopic composition versus depth plots do not allow such resolution, and so only as sume a 4.4 fractionation between soilrespired CO2 and soil CO2. 58

PAGE 69

Figure 29. 13C versus Depth Variance in values for all columns (small points) is summarized in average values for each horizon (larger points and a line); C3 organic matter is represented with black and C4 w ith gray. The small gray tria ngle is the single sample point that was obtained for the Bk2 horizon in the C4 columns. 13C vs Depth0 10 20 30 40 50 60 70 80 90 -35.00-30.00-25.00-20.00-15.00-10.00-5.0013C (o/oo)Depth (cm) Predicted Equilibrium value Predicted Equilibrium value C4 C3 59

PAGE 70

Figure 30. 18O versus Depth Variance in values for all columns (small points) is summarized in average values for each horizon (larger points and a line); C3 organic matter is represented with black a nd C4 with gray. The small gray triangle is the single sample point that was obtain ed for the Bk2 horizon in the C4 columns 18O vs Depth0 10 20 30 40 50 60 70 80 90 101214161820222426283032343618O (SMOW, o/oo)Depth (cm) Predicted Equilibrium value C4 C3 60

PAGE 71

13C versus 18O Figures 31 and 32 show cross plots of 13C versus 18O for both C3 and C4 columns, respectively. Lines representing trends expected for processes responsible for the isotopic values are illustrated the plot s, with the intersection of two lines representing the predicted equili brium values. Above the horizontal line is a trend of 18Oenrichment at the surface and below it is a trend of mixing with 18O-depleted atmospheric CO2 in deeper horizons. To the right of the vertical line is a trend of mixing with 13C-enriched atmospheric CO2, and to the left is a trend of 13C being preferentially taken up in the precipitation of CaCO3. It is interesting to note that the A2 and E horizons, in both C3 and C4 profiles, fall nearest to the predicted equilibrium value. The A1 horizons and the Bk horizons are the most influenced by additional factors. 61

PAGE 72

C3 13C vs 18O10.0 15.0 20.0 25.0 30.0 35.0 -35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.013C (o/oo)18O (VSMOW, o/oo) A1 Horizons A2 Horizons E Horizons Bk 1 Horizons Bk 2 Horizons Lab AtmospherePredicted Equilibrium value CaCO3 sink p referential13C-uptake M ixing with13C-enriched atmospheric CO2 Evaporative 18Oenrichment at surface Mixing with 18Odepleted atmospheric CO2 Figure 31. C3 13C versus 18O 13C vs. 18O for C3 columns with a model of deviation from equilibrium predicted values. 62

PAGE 73

C4 13C vs 18O10.0 15.0 20.0 25.0 30.0 35.0 -35.0 -30.0 -25.0 -20.0 -15.0 -10.0 -5.013C (o/oo)18O (VSMOW, o/oo) A1 Horizons A2 Horizons E Horizons Bk 1 Horizons Lab AtmospherePredicted Equilibrium value E vaporative 18Oenrichment at surface Mixing with 18O-depleted atmospheric CO2 CaCO3 sink preferential13Cuptake Mixing with13Cenriched atmospheric CO2 Figure 32. C4 13C versus 18O 13C vs. 18O for C4 columns with a model of deviation from equilibrium predicted values. 63

PAGE 74

Discussion Concentration Versus Depth Plots Trends of CO2 versus depth demonstrate that the CO2 being produced in the soil is diffusing up through the columns to mix this respired CO2 with atmospheric CO2, as predicted by the Cerling diffusion-production model. The E horizon showed little change in the C4 columns, and had a slight increase in concentration in the C3 columns. The variation in CO2 concentration (and isotopic compositio n) in the lab atmosphere samples is likely due to the building air handling system recycling a portion of used indoor air, and due to the outdoor air being from an urban environment, which typically has above average CO2 concentrations (Grimmond et al., 2002; Clark-Thorne and Yapp, 2003; Pataki and Bowling, 2003; Pataki et al., 2007). The CO2 diffusing through the soil pore spaces passes through variable media, thus the diffusion rate will vary from horizon to horizon. For example, the Bk horizon contains a large amount of clay mixed in with the sand, which impedes and slows the diffusion of respired CO2 through it due to reduction of pore space and permeability from the expansion of clay when water is present. The montmorillonite clay used in this e xperiment is 2:1 phyllosilicate clay, which expands in the presence of water, and may seal off the subsurface from the atmosphere. The E horizon, consisting of quartz sand, will have the greatest CO2 64

PAGE 75

diffusivity among all of the horizons beca use it will have the greatest amount of unsaturated pore space and permeability. This variability in diffusivity explains trends of high CO2 concentration visible in the Bk horizons. Perhaps the distinct difference between the sets of C3 and C4 columns is held due the respired CO2 having decreased diffusivity through the Bk horizon. It could be that they would have appeared more consistent with diffusion-production if gases could more readily mix and move through them. Although the Bk horizons in both C3 and C4 columns contain the same amount of clay, sand, and organic matter, the C4 columns appeared to retain more moisture perhaps due to a difference in moisture retention between the types of organic ma tter. If the Bk horizons of the C4 profiles retained more moisture than C3 profiles, the expansio n of clay due to moisture in the C4 columns could impact diffusivity. Respiration rate (the flux at the t op of the soil profiles) can be estimated using the soil CO2 concentration profiles. To accomplish this, the soil CO2 concentration must be converted from ppm to moles cm-3, using: ppm 1000000Lcm1000 moleL4.22 ppm 1 cm moles3 3 (7) Then the following equation adapte d from the Cerling (1984) diffusionproduction model is used to calculate respiration rate: 22z Lz CC Dos s (8) 65

PAGE 76

Where is the production in moles cm-3 s-1, Ds is the diffusion coefficient for CO2 in the soil in cm2 s-1, Cs is the concentration of soil CO2 in moles cm-3, Co is the atmospheric CO2 concentration in moles cm-3, L is the column depth in cm, and z is the sample depth in cm (Cerling, 1984) To convert from production rate to flux, was multiplied by 100 cm (the column depth). This method resulted in an average respiration rate for al l runs combined of 5.3 x 10-11 moles cm-2 s-1. For wet conditions it was 5.1 x 10-11 moles cm-2 s-1, for moist conditions it was 7.4 x 10-11 moles cm-2 s-1, and for dry conditions it was 3.3 x 10-11 moles cm-2 s-1. These rates are comparable to the estimated typical grassland soil respiration rates for th e non-growing season from Cerling (1984), which are up to 2.5 x 10-10 moles cm-2 s-1 in the growing season to 2.8 x 10-11 moles cm-2 s-1 in the non-growing season. The variation among these respiration rates with respect to soil moisture is likely due the response of the specific microbial community in these soil columns. The respiration rates were highest in moist conditions, when the soil was neith er saturated, which would likely shut down aerobic respiration, or dry which would limit the availability of water to the microorganisms. An interesting trend that is visible in the CO2 concentration versus depth plots that are separated by moisture regime is that the soil moisture status has a strong influence on soil respiration. In the wet conditions the CO2 concentration increased to 13431 ppm in th e Bk1 horizons of the C3 profiles as compared to 2361 ppm in dry conditions. In this wet soil situation the C4 66

PAGE 77

profiles did not provide data for the Bk horizons, likely because the pore spaces were saturated to the point that aerobic soil respiration shut down or was severely diminished. The Bk2 horizons of the C3 profiles may have similarly shut down during this saturated condition. In the moist soil conditions CO2 depth profiles showed some different trends. The Bk horizons in C3 pr ofiles only reached 10617 ppm and 10725 ppm for the Bk1 and Bk2 horizons, respectivel y. The C4 profiles showed greatly increased soil respiration rates during this moisture regime, reaching 12054 ppm in the Bk1 horizon and 12423 ppm in the E horizon. The high CO2 concentration in the E horizon was likely due to diffusion of CO2 from the Bk1 horizon. The C4 profile Bk2 horizons ha d a relatively low CO2 concentration of 6830 ppm, likely due to wet conditions remaining at the bottom of the columns. In the dry conditions, the C4 profiles did not produce CO2 concentrations greater th an 3780 ppm in the E horizon and 2370 ppm in the Bk horizon. The C3 profiles were more productive in these conditions, producing a CO2 concentration of 5039 ppm in the E horizon and a much higher 9731 ppm in the Bk1 horizon. The C3 Bk2 horizons (at 2361 ppm) produced similar concentrations to the C4 Bk1 horizons. As discussed previously, some studi es have found that soil moisture content influences soil resp iration rate while others ha ve found that it does not. In this situation it does, and there is some variation between the two profiles. The difference between CO2 concentration depth profiles in this study may be due to 67

PAGE 78

different saturation conditions and the microbial response to those conditions, which may depend on the quality of the C3 and C4 biomass. Both profiles contained the same amount of all starting materials, but for one having C3 biomass and the other having C4 biomass. It is possible that there is enough difference in the water holding capacity of the organic matter between these two that one profile retains water more than the other. This could explain why the C4 profiles seemed to retain water to a greater degree than the C3 profiles. 13C Keeling Plots Over entire columns, 13C values reflect the source of respired CO2 corrected for a 4.4 diffusional fractionati on, but values change with depth in a way that deviates somewhat from the diffusion-production model of soil CO2 (Cerling, 1984). The Keeling Pl ots (Figures 13 and 14) and 13C vs. depth plots (Figure 29) show changes in 13C(KP) ( 13C values obtained from Keeling Plots) and 13C(raw) values (respectively), in the developing Bk horizon that are significantly more 13C-depleted than in upper horizons Keeling plot intercepts for the Bk horizons are much more 13C-depleted than pred icted by diffusion of CO2 from the source organic matter (reaching values of -40 for the C3 soil and -18 for the C4 soil). This may be due to the precipitation of pedogenic carbonate in the developing Bk horizons which would preferentially uptake 13C (according to the fractionation factor between calcite and CO2 ~ 10), leaving 68

PAGE 79

the remaining soil CO2 13C-depleted. This is supported by mass balance calculations of Ca2+-additions suggesti ng that some CaCO3 may be precipitating. In each column, 0.625 g Ca2+ has been added, which could result in a maximum potential precipitati on of about 1.56 g CaCO3. Such a CaCO3 sink would amount to ~22.4 g C m-2 yr-1, roughly 5-10% of the total estimated soil CO2 production rate. However, it is unknown what amount of the added Ca2+ was incorporated in carbonate precipitation and how much drained through the profile. The pH of soil column effluent may hint at relative amounts of Ca2+ either draining through the profile or precipitating in CaCO3. At the start of the experiments the column effluent pH was approximately 3, likely due to organic acids leaching from the large amount of freshly added organic matter, and/or oxidation of ammonia released by decomposition. As the organic matter in the columns was decomposed by the soil microbial commun ity the pH gradually began to rise, gradually approaching the pH value of the water that entered the columns (6.48). Because carbonates do not precip itate in acidic conditions it is likely that much of the Ca2+ in the beginning of the experiment was drained through the columns. Once the column effluent reached pH levels consistent with the column irrigation water it is likely that much of the Ca2+ was being incorporated into CaCO3. For both the C3 and C4 profiles, varying moisture conditions had no consistent effect on the trends of the 13C depth profiles of soil CO2; nor the 13C composition of the soil-respired CO2. Although the moisture regimes affected the 69

PAGE 80

CO2 concentration profiles, this only affected the rate of soil respiration without any changes in fractionation. 18O Keeling Plots 18O Keeling Plots of both C3 and C4 soils (Figures 21 and 22) show a mean value similar to that predicted by equilibration with soil water, reflecting the equilibrium fractionation between CO2 and H2O at ambient lab temperature. 18O(KP) becomes significantly more 18O-enriched in the surface horizons (A1) due to evaporative 18Oenrichment of water near th e surface, and equilibration of CO2 with this water (surface intercepts reach +34-36 compared to a predicted value of ~29 for equilibrat ion with soil water). This effect may contribute to enrichment of 18O in ecosystem respiration in warm climates with high soil moisture, and should be accounted for in global 18O budgets (Ciais, et al, 2005). The soils in this study may mimic a wa rm moist climate during the periods of wet and moist soil conditions in the study. The heat lamps suspended above the columns in this study likely contribut ed to increased evaporation from the surface, which may or may not reflect c onditions in natural soils, depending on environment. The 18O(KP) values in the A2 and E horizons remain steady and similar to the predicted equilibrium value with depth, but are more variable in the deepest (Bk) horizons. Because a CO2 molecule can travel ~12 cm in the soil before it takes the isotopic signature of the soil wate r (Ciais et al., 2005). Soil 70

PAGE 81

CO2 diffusing through the soil may not be fully equilibrated with soil water until it reaches the E and A2 horizons, and may equilibrate with 18O-enriched water near the soil surface. These results of 18O versus depth are similar to the 18O versus depth plotted in Ster nberg et al (1998) from an Amazonian rainforest soil (Figure 33). The differences between th ese two depth trends are attributed to seasonal differences in soil respiratio n as a consequence of variation in temperature and moisture. In the lab soils in this study the temperature was held at a constant diurnal cycle for the en tire duration of the study; however, the moisture content of the soils in this study varied during the three sampling periods. It is therefore inte resting that the profiles in this study show a consistent pattern of variability in the 18O of the A1 horizons with the induced moisture changes. This illustrates how evaporation at the surface affects the 18O composition of the soil CO2 in the surface horizon and the resulting CO2 respired into the atmosphere. Figure 33. 18O versus Depth from Sternberg, et al. (1998). Note the similarity between these plots and Figure 14. It especially has similarity to the wet and moist soil plots in for 18O versus depth in Appendix E. 71

PAGE 82

The plots of 18O versus depth using Amazonian rainforest soils by Sternberg et al (1998) show similarity to the C3 and C4 profiles in this study during wet and moist conditions, but th e dry conditions did not show any similarity. That wet conditions show an increase in 18O at the surface is likely due to evaporative enrichment because there is enough water available to evaporate, and to equilibrate with CO2 as it exits the soil. In dry conditions there is not enough water present in the surface hor izons to evaporate, nor to equilibrate with CO2, so evaporative enrichment cannot take place. Cross Plots of 13C versus 18O The 13C versus 18O cross-plots for both C3 and C4 grasses show the deviations of the 13C and 18O values from the predicted equilibrium values in ways that are both consistent and in consistent with the diffusion-production model of soil CO2 (Cerling, 1984). The middle-range horizons (A2 and E) of the C3 columns plotted close to the predicted equilibrium values, with only a slight effect from mixing with 13C-enrcihed atmospheric CO2. The effects of mixing of atmospheric CO2 increase in the surface horizons, as predicted by the diffusionproduction model. In the C4 columns they also plotted close to the predicted equilibrium value, but had less effe ct of mixing with atmospheric CO2 and rather had a slight effect of mixing with 13C-depleted CO2 evident in the deeper horizons 72

PAGE 83

(where the suspected CaCO3 sink is). The Bk horizons of both the C3 and C4 columns are 13C-depleted, which is indicative of a preferential uptake of 13C in a CaCO3 sink. Although the columns were built, and the experiment designed, with the intention of precipitating carbonates in the Bk horizons, it was not expected that the developing Bk hori zons would exhibit such 13C-depletion as per the Cerling diffusion-production model. The A2 and E horizons of both C3 and C4 columns are not 18O-enriched, indicating mixing with 18O-depleted atmospheric CO2. Both C3 and C4 columns show the surface horizons (A1) displaying 18O-enriched values, indicating effects of equilibration with surface soil waters that have been evaporatively 18Oenriched. These cross-plots support the results of the Keeling Plots as both indicate two deviations from the Keeling Plot mixing model: (1) evaporative enrichment of 18O in the surface horizons (A1) and (2) the depletion of 13C in the deep horizons (developing Bk horizons) due to uptake in a carbonate sink. Possible Issues Overall the results of the isotopic an alyses illustrate the respired soil CO2 trends. This provides an ex cellent analog for what may occur in natural soils, but under controlled environmental conditions. It should be noted however, that some differences will exist between an arti ficial soil and a natural soilbesides the 73

PAGE 84

absence of root respiration in the artificial soil. One factor to consider is that the low level of litter diversity in this study (only one type of grass per column and a single stage of decomposition, i.e. grass cl ippings). This may or may not have a negative impact on the microbial commun ities of the soil (Bardgett and Shine, 1999). One reason why it may not have a negative impact is: despite microbial communities differing between soils (between soils of different locations and possibly between real and arti ficial soils as well), it is generally assumed that the soil CO2 that is heterotrophically-respired will be similar in terms of isotopic composition provided the starting mate rials are equivalent in isotopic composition. It is possible that the actual concentration of CO2 in the columns may be slightly different than that exp ected in a natural semi-arid grassland soil; simply because the limited supply and variety of organic matter for the microbes to decompose is different than the real soil situation, in which there is a continuous addition of new organic matte r as plant life grows and dies on a regular basis. Despite this potential issue, the CO2 concentration with depth appears to have the same overall trend that is expected for a semi-arid grassland soil. Also in relation to natural soils, the artificial soils used in this study will experience a gradual reduction in soil respiration with timebecause as the organic matter decomposed and CO2 respired, there was no fresh input of organic matter to sustain the microbial communities In natural grassland soil there will be a more consistent rate of input of new organic matter to the soil, predominately 74

PAGE 85

after the growing season (in the fall). A large die-off of grass and other herbaceous plant matter in natural conditi ons creates a great input of organic matter (Kirschbaum, 1995). Because the soil columns in this study were only monitored for about six months, similar to an annual growing season, the time limiting scope of this is not a problem. If it were to be continued for a much longer time period, the addition of fresh or ganic matter may be required to sustain decomposition. Conclusions Replicate C3 and C4 grassland soil prof iles were constructed for this study and regularly irrigated such that soil respiration would produce CO2 which could be extracted and analyzed. 13C Keeling Plot intercepts track the 13C of source CO2 when corrected with 4.4 diffusional fractionation for entire columns. However, the 13Cdepleted values in the developing Bk horizon may be related to pedogenic carbonate precipitation. 18O Keeling Plot intercepts track equilibration with soil water. 18O values from CO2 near the surface shows a pull towards 18O-enriched values, which is the likely result of preferential H2 16O evaporation from the surface during times of hi gh soil moisture. During dry conditions the surface horizons do not illustrate any evaporativ e effects, rather maintain similar 18O compositions to the rest of the profile. 75

PAGE 86

When 13C versus 18O is plotted, deviations from equilibrium are visible. The deviation of 13C from equilibrium may be controlled by mixing with atmospheric CO2 and by the preferential uptake of 13C in a CaCO3 sink, whereas the deviation of 18O from equilibrium may be controlled by equilibration with evaporatively 18Oenriched soil water near the surface in moist and wet soil conditions. The change in soil moisture did not affect the 13C composition within the soil profiles likely because no change in source of carbon, nor fractionation by the soil microbial community occurred. The primary effect that variation in soil moisture had was on soil respiration rate s. Moist conditions were favorable and produced high CO2 concentrations, but in satura ted conditions respiration was inhibited and CO2 concentration was reduced. Future Thoughts Future interesting work that coul d be done with these or similar soil columns includes, of course, growing pedogenic carbonates in the developing Bk horizon and analyzing the 13C and 18O of the carbonates to compare to the respired CO2 and DIC data. Other interesting plans would be to include the measurement and analysis of soil-respired CO2 and its possible variability under strict control of such factors as temperat ure, soil moisture, and substrate type and quality. It would be interesting to test these factors individually and in different 76

PAGE 87

combinations to compare to field studies and to help work out the subtleties in the process of heterotrophic soil respiration. 77

PAGE 88

References Amundson, R. 2004. Soil Formation. In: Treatise on Geochemistry, H.D. Holland and K.K. Turekian (Eds.). El sevier, Amsterdam, 1 35. Beheydt, D., P. Boeckx, T. J. Clough, J. Vermeulen, R. R. Sherlock, and O. Van Cleemput, 2005, Methods to adjust for the interference of N2O on 13C and 18O measurements of CO2 from soil mineralization. Rapid Communications in Mass Spectrometry; Vol. 19: 1365 1372. Blisniuk, P. M., and Stern, L. A., 2005, Stable isotope paleoaltimetry: A critical review. American Journal of Science. ; Vol. 305: 1033 1074. Cerling, T. E., 1984, The stable isotopic composition of modern soil carbonate and its relationship to climate. Earth and Planetary Science Letters; Vol. 71: 229 240. Cerling, T. E., D. K. Solomon, J. Quade, and J. R. Bowman, 1991, On the isotopic composition of carbon in soil carbon-dioxide. Geochimica Cosmochimica Acta; Vol. 55: 3403 3405. Ciais P., Cuntz M., Scholze M., Mouillot F., Peylin P. and Gitz V., 2005. Remarks on the use of 13C and 18O isotopes in atmospheric CO2 to quantify biospheric carbon fluxes, In Fl anagan L.B., Ehleringer J.R. and Pataki D.E. (Eds.) Stable Isotopes and Biosphere-Atmosphere Interactions, Academic Press, San Diego, 235 267. Clark, I. D., and P. Fritz, 1997. E nvironmental Isotopes in Hydrology. Contributer: P. Fritz; Publisher: CRC Press. 328 pages. Clark-Thorne, S. T., and C. J. Yapp, 2003, Stable carbon isotope constraints on mixing and mass balance of CO2 in an urban atmosphere: Dallas metropolitan area, Texas, USA. Applied Geochemistry; Vol. 18: 75 95. Coplen, T. B., C. Kendall, and J. Hopple, 1983, Comparison of stable isotope samples. Nature; Vol. 302: 236 238. 78

PAGE 89

Davidson, E. A., A. D. Richardson, K. E. Savage, and D. Y. Hollinger, 2006, A distinct seasonal pattern of the ratio of soil respiration to total ecosystem respiration in a spruce-dominated forest. Global Change Biology; Vol. 12: 230 239. Deutz, P., I. P. Montaez, and H. C. Monger, 2002, Morphology and stable and radiogenic isotope composition of pedogenic carbonates in Late Quaternary relict soils, New Mexico, U.S.A.: An integrated record of pedogenic overprinting. Journal of Sedimentary Research; Vol. 72, No. 6: 809 822. Dworkin, S. I., L. Nordt, and S. Atchley, 2005, Determining terrestrial paleotemperatures using the oxygen isotopic composition of pedogenic carbonate. Earth and Planetary Science Letters; Vol. 237: 56 68. Ehleringer, J. R., D. R. Bowling, L. B. Flanagan, J. Fessenden, B. Helliker, L. A. Martinelli, and J. P. Ometto, 2002, Stable isotopes and carbon cycle processes in forests and grasslands. Plant Biology; Vol. 4: 181 189. Ghosh, P., J. Adkins, H. Affek, B. Balta, W. Guo, E. A. Schauble, D. Schrag, and J. M. Eiler, 2006, 13C-18O bonds in carbonate minerals: A new kind of paleothermometer. Geochimica et Cosmochimica Acta; Vol. 70: 1439 1456. Grimmond, C. S. B., T. S. Ki ng, F. D. Cropley, D. J. Nowak, and C. Souch, 2002, Localscale fluxes of carbon dioxide in urba n environments:methodological challenges and results from Chicago. Environmental Pollution; Vol. 116: S243 S254. Keeling, C. D., 1958, The concentration and isotopic abundances of atmospheric carbon dioxide in rural areas. Geochimica et Cosmochimica Acta; Vol.13: 322 334. Keeling, C. D., 1961, The concentration and isotopic abundances of carbon dioxide in rural and marine air. Geochimica et Cosmochimica Acta; Vol. 24: 277 298. Kirschbaum, M. U. F., 1995, The temperature dependence of soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biology and Biochemistry; Vol. 27, No. 6: 753 760. Kirschbaum, M. U. F., 2004, Soil respira tion under prolonged soil warming: are rate reductions caused by acclimation or substrate loss? Global Change Biology; Vol. 10: 1870 1877. 79

PAGE 90

Kovda, I., C. I. Mora, and L. P. Wilding, 2006, Stable isotope compositions of pedogenic carbonates and soil organic matter in a temperate climate Vertisol with gilgai, southern Russia. Geoderma; Vol. 136: 423 435. Liu, W., J. Moriizumi, H. Yamazawa, and T. Iida, 2006, Depth profiles of radiocarbon and carbon isotopic compos itions of organic matter and CO2 in a forest soil. Journal of Environmental Radioactivity; Vol. 90: 210 223. Mortazavi, B., J. P. Prater, and J. P. Chanton, 2004, A field-based method for simultaneous measurements of the 18O and 13C of soil CO2 efflux. Biogeosciences; Vol. 1: 1 9. ONeil, J.R., and L.H. Adami, (1969). The oxygen isotope partition function ratio of water and the structure of liquid water, Journal of Physical Chemistry Vol. 73:1553 1558. Pataki, D. E., D. R. Bowling, and J. R. Ehleringer, 2003, Seasonal cycle of carbon dioxide and its isotopic compos ition in an urban atmosphere: Anthropogenic and biogenic effects. Journal of Geophysical researchatmospheres; Vol. 108, No. D23: Article Number 4735. Pataki, D. E., J. R. Ehleringer, L. B. Flanagan, D. Yakir, D. R. Bowling, C. J. Still, N. Buchmann, J. O. Kaplan, and J. A. Berry, 2003, The application and interpretation of Keeling plots in terrestrial carbon cycle research. Global Biogeochemical Cycles; Vol. 17 No. 1: 1022. Pataki, D. E., T. Xu, Y. Q. Luo, and J. R. Ehleringer, 2007, Inferring biogenic and anthropogenic carbon dioxide sources acr oss an urban to rural gradient. Ecosystem Ecology; Vol. 152: 307 322. Pendall, E. S. Bridgham, P. J. Hanson, B. Hungate, D. W. Kicklighter, D. W. Johnson, B. E. Law, Y. Luo, J. P. Megonigal, M. Olsrud, M. G. Ryan, and S. Wan, 2004, Below-ground process responses to elevated CO2 and temperature: a discussion of obs ervations, measurement methods, and models. New Phytologist; Vol. 162: 311 322. Schlesinger, W. H., 1997, Biogeochemistry An analysis of global change. 2nd Ed. Academic Press, San Diego, CA. 588 pages. Schmid, S., R. H. Worden, and Q. J. Fisher, 2006, Variations of stable isotopes with depth in regolith calcite cements in the Broken Hill region, Australia: Palaeoclimate evolution signal? Journal of Geochemical Exploration; 89: 355 358. 80

PAGE 91

Sternberg, Leonel Da S. L., M. Z. Moreira, L. A. Martinelli, R. L. Victoria, E. M. Barbosa, L. C. M. Bonates, and D. Nepstad, 1998, The relationship between 18O/16O and 13C/12C ratios of ambient CO2 in two Amazonian tropical forests. Tellus; Vol. 50B: 366 376. Tans, P. P., 1998, Oxygen isotopic equilib rium between carbon dioxide and water in soils. Tellus; Vol. 50B: 163 178. Van Groenigen, J. W., K. B. Zwart, D. Harris, and C. van Kessel, 2005, Vertical gradients of 15N and 18O in soil atmospheric N2Otemporal dynamics in a sandy soil. Rapid Communications in Mass Spectrometry; Vol. 19: 1289 1295. Wan, S., R. J. Norby, J. Ledford, a nd J. F. Weltzin, 2007, Responses of soil respiration to elevated CO2, air warming, and changing soil water availability in a model old-field grassland. Global Change Biology; Vol. 13: 2411 2424. Weerts, A. H.; Kandhai, D.; Bouten, W.; and Sloot, P. M. A. 2001, Tortuosity of an Unsaturated Sandy Soil Estimated using Gas Diffusion and Bulk Soil Electrical conductivity: Comparing Analogy-based Models and Lattice Boltzmann Simulations. Soil Science Society of America Journal, Division S-1Soil Physics; Vol. 65, No. 6: 1577 1584. Yakir, D., 2000, The use of stable isot opes to study ecosystem gas exchange. Oecologia; Vol. 123: 297 311. Yutse, J. C., D. D. Baldocchi, A. Gershenson, A. Goldstein, L. Misson, and S. Wong, 2007, Microbial soil respiration and its dependency on carbon inputs, soil temperature and moisture. Global Change Biology; Vol. 13: 2018 2035. 81

PAGE 92

Bibliography Bigeleisen, J., 1965, Chemistry of Isotopes. Science; Vol. 147, No. 3657: 463 471. Boone, R. D., K. J. Nadelhoffer, J. D. Canary, and J. P. Kaye, 1998, Roots Exert a Strong Influence on the Temperature Sensitivity of Soil Respiration. Nature; Vol. 396: 570 572. Cerling, T. E., 1991, Carbon Dioxide in the Atmosphere: Evidence from Cenozoic and Mesozoic Paleosols. American Journal of Science; Vol. 291: 377 400. Certini, G., 2005, Effects of Fire on Properties of Forest Soils: A Review. Oecologia; Vol. 143: 1 10. Davidson, E. A., and I. A. Janssens, 2006, Temperature Sensitivity of Soil carbon Decomposition and Feedbacks to Climate Change. Nature; Vol. 440: 165 173. Fang, C., P. Smith, J. B. Moncrieff, and J. U. Smith, 2005, Similar Response of Labile and Resistant Soil Organic Matter Pools to Changes in Temperature. Nature; Vol. 433: 57 59. Foley, J. A., 1995, An Equilibrium Model of the Terrestrial Carbon Budget. Tellus; Vol. 47B: 310 319. Ghosh, P., J. Eiler, S. E. Campana, and R. F. Feeney, 2007, Calibration of the carbonate clumped isotope paleothermometer for otoliths. Geochimica et Cosmochimica Acta; Vol. 71: 2736 2744. Gu, L., W. M. Post, and A. W. Ki ng, 2004, Fast Labile Carbon Turnover Obscures Sensistivity of Hetero trophic Respiration From Soil to Temperature: A Model Analysis. Global Biogeochemical Cycles; Vol. 18, GB1022: 1 11. 82

PAGE 93

Lin, G., J. R. Ehleringer, P. T. Rygiew icz, M. G. Johnson, and D. T. Tingey, 1999, Elevated CO2 and Temperature Impacts on Different Components of Soil CO2 Efflux in Douglas-fir Terracosms. Global Change Biology; Vol. 5: 157 168. Lloyd, J., and J. A. Taylor, 1994, On the Temperature Dependence of Soil Respiration. Functional Ecology; Vol. 8, No. 3: 315 323. Raich, J. W., and W. H. Schlesinger, 1992, The Global Carbon Dioxide Flux in Soil Respiration and its Relationship to Vegetation and Climate. Tellus; Vol. 44B: 81 99. Raich, J. W., and C. S. Potter, 1995, Global Patterns of Carbon Dioxide Emissions From Soils. Global Biogeochemical Cycles; Vol. 9, No. 1: 23 36. Raich, J. W., and A. Tufekcioglu, 2000, Vegetation and Soil Respiration: Correlations and Controls. Biogeochemistry; Vol. 48: 71 90. Rustad, L. E., T. G. Huntington, and R. D. Boone, 2000, Controls on Soil Respiration: Implications for Climate Change. Biogeochemistry; Vol. 48, No. 1: 1 6. Schauble, E. A., P. Ghosh, J. M. Eiler, 2006, Preferential formation of 13C-18O bonds in carbonate minerals, estimate d using first-principles lattice dynamics. Geochimica et Cosmochimica Acta; Vol. 70: 2510 2529. Schlesinger, W. H., and J. A. Andrew s, 2000, Soil Respiration and the Global Carbon Cycle. Biogeochemistry; Vol. 48, No. 1: 7 20. Sowerby, A., H. Blum, T. R.G. Gray, and A. S. Ball, 2000, The Decomposition of Lolium perenne in Soils Exposed to Elevated CO2: Comparisons of mass Loss of Litter With Soil Respiration and Soil Microbial Biomass. Soil Biology & Biochemistry; Vol. 32: 1359 1366. Ufnar, D. F., D. R. Grcke, and P. A. Beddows, 2008, Assessing Pedogenic Calcite Stable-Isotope Values: Can Positive Linear Covariant Trends be Used to Quantify Palaeo-Evaporation Rates? Chemical Geology; Vol. 256: 46 51. 83

PAGE 94

Appendices 84

PAGE 95

Appendix A Normalized Values from Utilized Samples of Soil CO2 This appendix includes three tables. Each table contains the normalized data from one sampling date. Samples th at were discarded in the normalization procedure are not included here. The samples shown in gray were accepted by the correction procedures but were not include d in the data analyzed in this thesis due to problems with column T3. See Appendix F for more information on column T3. St.A refers to columns with C3 organic matter and T refers to columns with C4 organic matter. The number immediately following the letters refers to the column number, and the s econd numeral refers to the horizon. For example, St.A1 1 is the A1 horizon of the first of the replicate C3 profiles. St.A1 2 St.A1 3, St.A1 4, and St.A1 5 refer to the A2, E, Bk1, and Bk2 horizons, respectively of the first of the replicate C3 profiles. 85

PAGE 96

08-01-2008 Identifier 13C 18O [CO2] Lab Atm 1 -9.52 23.44 366 Lab Atm 2 -9.56 23.55 371 Lab Atm 3 -9.83 23.38 380 St.A 1 1 -15.62 29.99 739 St.A 1 2 -21.65 28.08 2307 St.A 1 3 -22.86 27.95 3473 St.A 1 4 -25.75 28.03 6215 St.A 1 5 -32.73 24.99 928 St.A 2 1 -19.54 32.03 1274 St.A 2 2 -24.12 28.25 13160 St.A 2 3 -24.53 28.72 17163 St.A 2 4 -32.51 29.09 20080 St.A 3 1 -18.10 30.04 949 St.A 3 2 -23.68 28.49 8343 St.A 3 3 -24.31 28.61 11215 St.A 3 4 -30.03 26.37 5880 T1 1 -10.74 33.05 1153 T1 2 -11.07 28.40 14450 T1 3 -11.38 28.64 11164 T1 4 -18.37 26.03 10420 T2 1 -10.08 33.13 1428 T2 2 -9.79 28.44 10396 T2 3 -9.80 28.76 12944 T2 4 -16.97 26.92 3240 T2 5 -14.10 29.09 11675 T3 1 -8.74 32.42 3906 T3 2 -3.77 30.88 227121 T3 3 -3.54 31.32 221830 T3 4 -6.56 31.26 216053 T3 5 -7.58 30.75 207163 Table A1. Normalized values from samples taken on August 1, 2008. 86

PAGE 97

08-09-2008 Identifier 13C 18O [CO2] Lab Atm 1 -11.27 24.64 364 Lab Atm 2 -9.05 26.16 362 StA 2-1 -19.36 32.78 1061 StA 2-2 -24.20 27.78 8133 StA 2-3 -24.49 27.99 12100 StA 2-4 -34.31 29.28 20837 StA 3-1 -17.87 31.04 595 StA 3-2 -23.94 27.86 4855 StA 3-3 -24.89 27.96 6112 StA 3-4 -31.59 28.11 6026 TPA 1-1 -10.38 33.05 625 TPA 1-2 -10.86 27.47 3647 TPA 1-3 -11.25 27.37 4424 TPA 2-1 -10.33 33.03 666 TPA 2-2 -10.75 27.44 4818 TPA 2-3 -11.26 27.37 5816 TPA 2-4 -16.47 26.82 3906 TPA 3-1 -10.23 34.38 900 TPA 3-2 -10.69 29.86 54501 TPA 3-3 -10.53 29.91 62055 TPA 3-4 -7.55 30.14 53466 TPA 3-5 -7.21 29.79 64950 Table A2. Normalized values from samples taken on August 9, 2008. 87

PAGE 98

10-12-2008 Identifier 13C 18O [CO2] Lab Atm 1 -8.17 25.19 404.94 Lab Atm 2 -8.07 24.50 408.81 Lab Atm 3 -7.85 23.20 411.33 StA1 1 -13.64 24.82 565.12 StA1 2 -19.63 26.93 1130.20 StA1 3 -21.02 25.87 1509.00 StA1 4 -26.13 26.85 2667.59 StA1 5 -32.12 25.01 1287.22 StA2 1 -18.29 29.17 913.93 StA2 2 -24.17 28.75 5245.59 StA2 3 -24.84 27.88 7785.43 StA2 4 -30.41 29.20 22413.28 StA2 5 -34.48 27.06 3434.96 StA3 1 -17.53 25.64 820.84 StA3 2 -23.61 28.88 4064.82 StA3 3 -24.53 27.85 5822.22 StA3 4 -32.94 26.93 4112.72 T1 1 -10.96 27.93 811.10 T1 2 -11.72 28.17 2992.28 T1 3 -11.98 27.23 3701.10 T1 4 -19.05 24.98 1679.33 T2 1 -10.45 28.77 887.43 T2 2 -10.56 27.99 3305.44 T2 3 -11.17 27.15 3858.12 T2 4 -16.39 26.10 3059.70 T3 1 -10.38 31.62 1696.72 T3 2 -10.32 30.88 43528.70 T3 3 -10.29 30.00 51013.39 T3 4 -10.52 30.71 58130.82 T3 5 -10.39 31.02 59557.32 Table A3. Normalized values from samples taken on October 12, 2008. 88

PAGE 99

Appendix B. Data from sampling dates which were not included in this thesis. This appendix consists of tables that contain data from early samples that were not utilized in this thesis. These da ta were not utilized because processes of respiration in the columns were not yet stabilized. It also includes data from sampling dates that did not provide enough data to make use of due to insufficient volume of sampled CO2 for IRMS measurement. St.A refers to columns with C3 organic matter and T refers to columns with C4 organic matter. The numeral immediately following the letters refers to the column number, and the second numeral refers to the horizon. For example, St.A1 1 is the A1 horizon of the first of the replicate C3 profiles. St.A1 2 St.A1 3, St.A1 4, and St.A1 5 refer to the A2, E, Bk1, and Bk2 horizons, respectively of the first of the replicate C3 profiles. 89

PAGE 100

2-22-2008 Identifier 13C 18O [CO2] StA1 1 -20.24 48.83 4365 StA1 2 -24.01 39.17 50734 StA1 3 -25.93 37.49 83253 StA1 4 -21.69 38.15 77603 StA1 5 -18.51 36.93 92267 StA2 1 -20.15 50.55 2957 StA2 2 -25.86 38.43 66423 StA2 3 -8.34 35.95 103607 StA2 4 -9.06 33.60 132637 StA2 5 -10.93 34.66 118413 StA3 1 -20.22 50.46 3632 StA3 2 -25.41 38.60 60551 StA3 4 -27.90 35.44 110190 StA3 5 -20.48 35.82 107192 T1 1 -4.54 50.70 2531 T1 2 -8.31 38.61 68185 T1 3 -7.45 39.31 63205 T1 4 -6.98 39.31 64731 T1 5 -6.42 39.23 65248 T2 1 -3.72 49.41 2769 T2 2 -6.50 39.34 58266 T2 3 -5.81 39.51 60809 T2 4 -5.24 38.48 71644 T2 5 -5.20 38.58 67031 T3 1 -5.67 48.16 3931 T3 2 -9.86 39.55 56836 T3 3 -10.29 39.64 56394 T3 4 -9.33 39.10 65274 T3 5 -8.74 38.76 70511 Table A4. Data from early samples taken on February 22, 2008. 90

PAGE 101

2-28-2008 Identifier 13C 18O [CO2] St. A 1 1 -23.80 49.14 9691 St. A 1 2 -24.16 41.79 48488 St. A 1 3 -23.48 43.27 69148 St. A 1 4 -20.48 44.74 90854 St. A 1 5 -15.60 45.47 106084 St. A 2 1 -25.28 49.43 2275 St. A 2 2 -25.99 41.50 36938 St. A 2 3 -7.12 40.94 13493 St. A 2 4 1.77 49.95 200877 St. A 2 5 -1.08 48.90 184636 St. A 3 1 -25.57 48.78 3205 St. A 3 2 -25.56 41.70 49244 St. A 3 3 -25.26 43.06 64976 St. A 3 4 -24.47 43.62 72156 St. A 3 5 -13.34 49.88 199689 T 1 1 -8.88 48.15 2316 T 1 2 -8.29 43.09 65957 T 1 3 -7.93 43.63 66376 T 1 4 -7.21 43.82 69093 T 1 5 -6.43 43.92 71070 T 2 1 -8.71 47.88 2437 T 2 2 -8.17 43.16 67146 T 2 3 -8.25 43.63 68156 T 2 4 -1.22 44.43 88016 T 2 5 -2.33 44.07 85458 T 3 1 -9.83 47.41 2567 T 3 2 -9.29 43.11 65032 T 3 3 -9.71 42.73 58355 T 3 4 -8.25 43.97 76823 Table A5. Data from early samples taken on February 28, 2008. 91

PAGE 102

9-9-2008 Identifier 13C 18O [CO2] Lab Atm -11.27 24.64 364 Lab Atm -9.05 26.16 362 StA 2-1 -19.36 32.78 1061 StA 2-2 -24.20 27.78 8133 StA 2-3 -24.49 27.99 12100 StA 2-4 -34.31 29.28 20837 StA 3-1 -17.87 31.04 595 StA 3-2 -23.94 27.86 4855 StA 3-3 -24.89 27.96 6112 StA 3-4 -31.59 28.11 6026 TPA 1-1 -10.38 33.05 625 TPA 1-2 -10.86 27.47 3647 TPA 1-3 -11.25 27.37 4424 TPA 2-1 -10.33 33.03 666 TPA 2-2 -10.75 27.44 4818 TPA 2-3 -11.26 27.37 5816 TPA 2-4 -16.47 26.82 3906 TPA 3-1 -10.23 34.38 900 TPA 3-2 -10.69 29.86 54501 TPA 3-3 -10.53 29.91 62055 TPA 3-4 -7.55 30.14 53466 TPA 3-5 -7.21 29.79 64950 Table A6. Corrected values from samples taken on September 9, 2008. This was not used because CO2 samples taken on this date were accidentally not a sufficient volume for use in the mass spectrometer. 92

PAGE 103

Appendix C Effluent pH. This appendix contains a small table of the pH of the column effluent. The second test of pH does not list a precise value. Testing of the pH of the column effluent was halted after the e ffluent pH was observed to be reaching normal values. The pH of the water that entered the columns was 6.48. When the columns were relatively new and the pH of the effluent was low, a yellow and white material was being deposited wherev er the effluent evaporated. A sample of the effluent was evaporated over a burner, the pH reduced as the water evaporated and the precipitate became more concentrated. The starting pH was 3.07, and after only 25% of the liquid rema ined the pH had reduced to 1.67. A sample of the material that was prec ipitating out of solution was dried and prepared to run through an XRD to determine what the material was. Unfortunately, no meaningful results we re obtained due to XRD malfunction. 93

PAGE 104

Sampling Date Column Effluent pH 2-11-2008 3.07 4-1-2008 Close to 5.00 5-9-08 5.12 (St.A1), 5.38 (St.A3), 5.17 (T3) Table A7. pH of soil column effluent. Columns combined unless specified otherwise. 94

PAGE 105

Appendix D Watering and Gas sampling Schedule This appendix contains a table of all of the watering dates and sampling dates from the start of the project to the end of data collection for this thesis. Any watering and sampling that occurred after that point is not included. The sampling days that were utilized in this thesis were: 08 01 2008, 08 09 2008, and 10 12 2008. The data from the other sampling days can be viewed in Appendix B. Indicator W = watering S = gas sampling Date Watering Volume (mL) Gas sampling: days after watering W 02 08 -2008 175 ------S 02 28 2008 ------20 W 03 04 2008 650 ------W 03 24 2008 175 ------W 04 01 2008 650 ------S 04 19 2008 ------18 W 05 08 2008 175 ------S 05 21 2008 ------13 W 07 07 2008 650 ------S 08 01 2008 ------25 W 08 07 2008 650 ------S 08 09 2008 ------2 S 10 12 2008 ------64 W 10 23 2008 650 ------Table A8. Watering and Gas Sampling Schedule. 95

PAGE 106

Appendix E Additional Keeling Plot Line Equations and R2 Values This appendix contains additional Keeling Plot line equations and R2 values that correspond to the Keeling Plots in the Results secti on of this thesis 96

PAGE 107

C3 13C Keeling Plot Equations and R2 Values. 8-9-2008 (wet soil conditions) A1 y = 5425.8x 25.4 0.9 A2 y = 5375.9x 25.0 1.0 E y = 5524.2x 25.4 1.0 Bk1 y = 8631.2x 33.9 1.0 8-1-2008 (moist soil conditions) A1 y = 5071.9x 23.2 1.0 A2 y = 5464.5x 24.3 1.0 E y = 5614.9x 24.7 1.0 Bk1 y = 7755.7x 30.5 1.0 10-12-2008 (dry soil conditions) A1 y = 7657.5x 26.8 1.0 A2 y = 7149.6x 25.6 1.0 E y = 7249.7x 25.8 1.0 Bk1 y = 9788.0x 32.0 1.0 Bk2 y = 12976.0x 40.0 1.0 Table A9. Equations and R2 values for the Keeling Plot lines in Figures 15 17. 97

PAGE 108

C4 13C Keeling Plot Equations and R2 Values. 8-9-2008 (wet soil conditions) A1 y = 173.2x 10.6 0.0 A2 y = 258.2x 10.9 0.1 E y = 430.9x 11.3 0.3 Bk1 y = 2530.4x 17.1 0.9 8-1-2008 (moist soil conditions) A1 y = 396.5x 10.7 0.7 A2 y = 308.2x 10.5 0.5 E y = 367.3x 10.6 0.5 Bk1 y = 3244.1x 18.3 1.0 Bk2 y = 1716.2x 14.3 1.0 10-12-2008 (dry soil conditions) A1 y = 2084.2x 13.2 1.0 A2 y = 1454.2x 11.6 0.9 E y = 1622.2x 12.0 1.0 Bk1 y = 4761.3x 19.8 0.9 Table A10. Equations and R2 values for the Keeling Plot lines in Figures 18 20. 98

PAGE 109

C3 18O Keeling Plot Equations and R2 Values. 8-9-2008 (wet soil conditions) A1 y = -4259.4x + 37.3 0.9 A2 y = -931.0x + 28.0 0.8 E y = -977.2x + 28.1 0.8 Bk1 y = -1252.3x + 28.8 0.9 8-1-2008 (moist soil conditions) A1 y = -4410.6x + 35.3 1.0 A2 y = -1933.3x + 28.7 1.0 E y = -1958.8x + 28.7 1.0 Bk1 y = -1720.9x + 28.1 0.9 Bk2 y = -946.4x + 26.0 1.0 10-12-2008 (dry soil conditions) A1 y = -2409.6x + 30.0 0.6 A2 y = -1973.7x + 29.1 0.9 E y = -1423.5x + 27.7 0.8 Bk1 y = -1551.3x + 28.1 0.8 Bk2 y = -989.7x + 26.7 0.6 Table A11. Equations and R2 values for the Keeling Plot lines in Figures 23 25. 99

PAGE 110

C4 18O Keeling Plot Equations and R2 Values. 8-9-2008 (wet soil conditions) A1 y = -6327.7x + 42.8 1.0 A2 y = -815.3x + 27.6 0.8 E y = -769.2x + 27.5 0.8 Bk1 y = -565.7x + 27.0 0.5 8-1-2008 (moist soil conditions) A1 y = -5042.3x + 37.0 1.0 A2 y = -1905.4x + 28.6 1.0 E y = -2012.8x + 28.9 1.0 Bk1 y = -1198.2x + 26.7 0.9 Bk2 y = -2164.0x + 29.3 1.0 10-12-2008 (dry soil conditions) A1 y = -3189.2x + 32.1 0.9 A2 y = -1769.2x + 28.6 0.9 E y = -1318.8x + 27.5 0.8 Bk1 y = -646.5x + 25.9 0.4 Table A12. Equations and R2 values for the Keeling Plot lines in Figures 26 28. 100

PAGE 111

101 Appendix F Column T3 Problems Column T3 developed a dr ainage problem either due to or resulting in the growth of a fungus in the A horizons of the column. The fungus appeared to be hydrophobic, as it became impossible to water the column, the water would not infiltrate. The column also would no t dry, perhaps also due to the fungus blocking evaporation through the surface. The Bk horizon became anoxic and turned black. These problems caused the column to produce anomalously high concentration and 13C results of the soil CO2 (which was collected and analyzed despite the apparent problems). T3 data was excluded from all plots, but can be viewed in the tables in Appendix A and Appendix B.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200409Ka 4500
controlfield tag 001 002029607
005 20090918110355.0
007 cr bnu|||uuuuu
008 090918s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002946
035
(OCoLC)436988846
040
FHM
c FHM
049
FHMM
066
(S
090
QE26.2 (Online)
1 100
Chelladurai, Jennifer.
0 245
Using Keeling Plots to trace C and O through processes of heterotrophic respiration, diffusion and soil water equilibration in artificial C3- and C4-grassland soils
h [electronic resource] /
by Jennifer Chelladurai.
3 246
Using Keeling Plots to trace delta[superscript 13]C and delta[superscript13] of CO[subscript 2] through processes of heterotrophic respiration, diffusion and soil equilibration in artificial C3- and C4- grassland soils
260
[Tampa, Fla] :
b University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 101 pages.
502
Thesis (M.S.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Non-Latin script record
590
Advisor: Jonathan Wynn, Ph.D.
520
ABSTRACT: Global carbon cycle dynamics and fluxes of CO between biosphere and atmosphere have been progressed through the use of Keeling Plots. Processes that control and effect the isotopic composition of soil-respired CO, soil CO, and equilibrated soil carbonate are specifically addressed in this study through the use of Keeling Plots. Replicate grassland soil profiles containing either C3 or C4 homogenized organic matter were constructed and maintained under controlled settings to encourage the production of soil-respired CO and the precipitation of pedogenic carbonate. Soil CO was sampled over five months and analyzed with IRMS. Keeling Plots illustrated source CO affected by mixing with atmospheric CO near the surface and equilibration with C-depleted CO at depth in the zone of likely carbonate precipitation. The C Keeling Plot intercepts for the surface horizons (~ -24.7 per mil for C3 profiles and ~ -11.1 per mil for C4 profiles) follow the diffusion-production model when corrected with a constant 4.4 per mil diffusional fractionation, but the Keeling Plot intercepts for developing Bk horizons were curved towards depleted values (~ -36.2 per mil for C3 profiles and ~ -18.4 per mil for C4 profiles). This change in isotopic composition with depth deviates from the usual interpretations of Keeling Plots (steady-state, source to background diffusional mixing). C Keeling Plot intercepts indicated evaporative enrichment in the surface horizons of C3 and C4 profiles). This study uses Keeling Plots as a measure of mixing to assess the efficacy of steady-state diffusion-production models of soil CO equilibration with soil carbonate.
653
Stable isotopes
Carbon
Oxygen
Carbonates
Fractionation
690
Dissertations, Academic
z USF
x Geology
Masters.
773
t USF Electronic Theses and Dissertations.
880
Using Keeling Plots to trace C and O of CO through processes of heterotrophic respiration, diffusion and soil water equilibration in artificial C3- and C4-grassland soils
[electronic resource] /
by Jennifer Chelladurai.
4 856
u http://digital.lib.usf.edu/?e14.2946