The Integral Role of Phytoplankton Stoichiometry in Ocean Biogeochemical Dynamics
Total Page:16
File Type:pdf, Size:1020Kb
The integral role of phytoplankton stoichiometry in ocean biogeochemical dynamics A Dissertation SUBMITTED TO THE FACULTY OF THE UNIVERSITY OF MINNESOTA BY Tatsuro Tanioka IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Katsumi Matsumoto November, 2019 © Tatsuro Tanioka 2019 ALL RIGHTS RESERVED Acknowledgements I would like to thank my advisor Katsumi Matsumoto who has given me the best ad- vice and encouragement whenever I needed. I am also grateful to the members of my committee- Jake Bailey, Jim Cotner, and Bill Seyfried for their time and energy. I would also like to thank Kathy Tokos for helping me with MESMO. This work was supported by student fellowships from the Earth Sciences Department, Doctoral Dissertation Fel- lowship from the University of Minnesota Graduate School, and by NSF grant awarded to Dr. Katsumi Matsumoto. Thank you to all my friends and colleagues from the de- partment and outside the department; and to my teammates from the Australian Rules Football Team, Minnesota Freeze. Finally, thank you to my family for their support over the years. Tatsuro Tanioka November, 2019 i Dedication To my life-coach, my late grandfather Mitsugu. ii Abstract Photosynthesis by ocean algae (phytoplankton) contributes roughly half of the earth's net carbon production. Organic matter produced using carbon dioxide in the atmosphere not only supports marine food webs, but also acts as a climate stabilizer, because carbon is subsequently transported to the deep ocean and stored there for thousands of years. Attempts to model global marine biological production and its impacts on global biogeochemical cycles often assume a constant elemental stoichiometry of carbon, nitro- gen, and phosphorus in phytoplankton biomass. This ratio, known as the Redfield ratio, was determined on the basis of an analysis of many samples of marine plankton collected over 70 years ago. This notion is well established in the oceanographic community but there is no clear physiological justification for why the C:N:P ratios in phytoplankton should strictly follow the Redfield ratio. Many recent studies revealed that C:N:P ra- tio of particulate organic matter can deviate significantly from the Redfield Ratio with some noticeable spatial and temporal variability. Studies suggest that factors such as nutrient availability, light, and temperature play a crucial role in modifying C:N:P ratio of phytoplankton. In this dissertation, I investigate the roles of marine phytoplankton stoichiometry in the global marine biogeochemical dynamics by combining meta-analysis, numerical modeling, and remote sensing. I propose a mechanistic framework for predicting C:N:P in phytoplankton under different environmental conditions and I incorporate this frame- work into an Earth System Model to show their effects on global carbon cycle. I also present results on how the change in elemental composition of phytoplankton could af- fect the feeding behavior of zooplankton as well the ecosystem stoichiometry. Finally, I show that C:N:P is closely tied to the rate at which oxygen is consumed during organic matter remineralization and I propose that the change in phytoplankton stoichiometry could ameliorate the rate of marine deoxygenation. In summary, C:N:P of phytoplankton is flexible and will play key roles in future global ocean biogeochemical dynamics. iii Contents Acknowledgements i Dedication ii Abstract iii List of Tables viii List of Figures ix 1 Introduction 1 1.1 Overview of Chapters . 2 2 Phytoplankton stoichiometry and global export production 4 2.1 Synopsis . 4 2.2 Introduction . 5 2.3 New Stoichiometry Sensitivity Factor . 7 2.3.1 Derivation . 7 2.3.2 Estimation of Parameters . 10 2.3.2.1 Nonlinear Least Squares Regression . 11 2.3.2.2 Estimation of Parameters from First Principles . 12 2.4 First-Order Estimation of Global Stoichiometric Buffer Effect . 13 2.5 Flexible Stoichiometry in a Global Ocean Model . 16 2.5.1 Model Description . 17 2.5.2 Model Results . 22 iv 2.6 Conclusions . 25 2.7 Acknowledgments . 26 3 Environmental drivers of phytoplankton stoichiometry: a meta-analysis 27 3.1 Synopsis . 27 3.2 Introduction . 28 3.3 Materials and Methods . 30 3.3.1 Bibliographic search and screening . 30 3.3.2 Stoichiometry sensitivity factor as effect size . 32 3.3.3 Meta-analysis . 33 3.4 Results . 34 3.4.1 Effects of Phosphate . 34 3.4.2 Effects of Nitrate . 35 3.4.3 Effects of Nitrate:Phosphate supply ratio . 35 3.4.4 Effects of Irradiance . 35 3.4.5 Effects of Temperature . 37 3.4.6 Effects of Iron . 37 3.5 Discussion . 37 3.5.1 Basic framework . 37 3.5.2 Macronutrients (Phosphate and Nitrate) . 43 3.5.3 Irradiance . 45 3.5.4 Temperature . 47 3.5.5 Iron . 49 3.6 Implications for global biogeochemical cycles . 50 3.6.1 Conclusions . 53 3.6.2 Acknowledgements . 54 4 Phytoplankton stoichiometry and feeding behavior of zooplankton 55 4.1 Synopsis . 55 4.2 Introduction . 56 4.3 Method . 57 4.3.1 C:N:P as a function of age . 59 4.3.2 Maximum assimilation rate as a function of age . 60 v 4.3.3 Flexible grazing preference as a function of age . 62 4.3.4 Experimental setup . 65 4.4 Results . 66 4.5 Discussion . 70 4.6 Acknowledgments . 71 5 Phytoplankton stoichiometry and organic matter respiration 72 5.1 Synopsis . 72 5.2 Introduction . 73 5.3 Methods . 75 5.3.1 Estimating O2:C from satellite-dericed phytoplankton macromolec- ular composition . 75 5.3.2 Estimating O2:C from the vertical gradient method of dissolved nutrients and oxygen . 76 5.3.3 Estimating O2:C from laboratory and in situ measurements of phytoplankton macromolecules . 77 5.3.4 Monte Carlo simulation . 78 5.4 Results and Discussion . 79 5.5 Implications for the Future Marine Oxygen Cycle . 84 5.6 Acknowledgments . 86 6 Concluding Remarks 88 References 91 Appendix A. Supporting Information For Chapter 2 126 Appendix B. Supporting Information For Chapter 3 128 Appendix C. Supporting Information For Chapter 4 130 Appendix D. Supporting Information For Chapter 5 131 D.1 Uncertainties associated with O2:C remineralization ratio calculated from the satellite-derived estimate of macromolecules . 132 D.2 Derivation of equation(5.4) in the main text . 132 vi D.3 Assumptions and limitations of the vertical gradient method . 134 D.4 Validating the vertical gradient method . 134 vii List of Tables 2.1 Comparison of Flexible P:C Formulations . 9 2.2 Model Parameters for the Preindustrial and Transient Simulation . 18 2.3 Model Results for the Preindustrial and Transient Simulations . 23 3.1 Breakdown of the number of experimental units . 32 3.2 Summary of s-factors for P:C and N:C . 41 3.3 Projected change in C:P and C:N between 1981-2000 and 2081-2100 . 51 B.1 List of 64 studies used in the meta-analysis . 128 C.1 Default food preference of mesozooplankton in the original ERSEM model130 D.1 Assumed elemental composition of main phytoplankton macromolecules obtained from literatures . 145 D.2 Uncertainties in satellite-derived estimate of macromolecules . 145 D.3 Effects of changing percent fraction of nucleic acid in estimating O2 :Crem from the satellite-derived estimate of phytoplankton macromolecules . 146 D.4 Summary of macromolecular data from Roy (2018) . 147 D.5 O2 :Crem from 9 sensitivity analyses accounting for seasonal variability and depth choice . 148 viii List of Figures 2.1 Observed particulate P:C versus surface PO4 concentrations . 11 2.2 Stoichiometry sensitivity factor against PO4 . 14 2.3 Change in export production from 1990s as a function of the fractional change in PO4 from 1990s . 16 2.4 Modeled community stoichiometry sensitivity factor under preindustrial condition . 19 2.5 Modeled nutrient limitation under preindustrial condition in the surface layer...................................... 20 2.6 Modeled C:P and N:P of phytoplankton in the surface layer . 21 2.7 Simulated changes in surface PO4 and total export production in 2090s relative to 1990s . 22 2.8 Results from a global ocean model under IPCC RCP8.5 scenario . 25 3.1 low chart showing the preliminary selection criteria and the refined se- lection criteria used for determining s-factors . 31 3.2 S-factors for P:C and N:C with respect to changes in macronutrients . 36 3.3 S-factors for P:C and N:C with respect to changes in irradiance . 38 3.4 S-factors for P:C and N:C with respect to changes in temperature . 39 3.5 S-factors for P:C and N:C with respect to changes in iron . 40 3.6 Illustration of how the five environmental drivers under a typical future climate scenario affect the cellular allocation . 43 4.1 Biogeochemical model ERSEM used in simulations . 58 4.2 C:N:P ratios and maximum assimilation rate during copepod ontogeny . 61 4.3 Schematic diagram showing the effect of food quality on food preference of copepod (mesozooplankton) . 64 ix 4.4 Equilibrium biomass of plankton functional types as a function of meso- zooplankton N:P ratio . 67 4.5 Equilibrium food preference of mesozooplankton as a function of meso- zooplankton N:P ratio . 68 4.6 N:P ratio of released nutrient and large POM as a function of mesozoo- plankton N:P ratio . 69 5.1 Satellite-derived annually averaged phytoplankton macromolecular con- tent and O2:C remineralization ratio binned into 11 oceanographic regions 80 5.2 Export and remineralization ratios determined from the vertical gradient method . 81 5.3 O2 :Crem estimated from laboratory-based measurements of phytoplank- ton macromolecule content . 83 5.4 O2 :Crem in various marine ecosystems estimated from biochemical com- position of phytoplankton .