U UNI V E RSI T Y O F C IN C INN A T I 08/10/2009 Date: Gokce K. USTUNISIK I, , hereby submit this original work as part of the requirements for the degree of: Doctor of Philosophy Arts&Sciences, Geology in It is entitled: Application of Magma Recharge, Plagioclase Zoning, and Crystal Size Distribution (CSD) Theory to Natural Solid-Liquid Equilibria Gokce Ustunisik Student Signature: This work and its defense approved by: Dr. Attila Kilinc Committee Chair: Dr. J. Barry Maynard Dr. Warren D. Huff Dr. David B. Nash Dr. A. Umran Dogan Approval of the electronic document: I have reviewed the Thesis/Dissertation in its final electronic format and certify that it is an accurate copy of the document reviewed and approved by the committee. Attila Kilinc Committee Chair signature: Application of Magma Recharge, Plagioclase Zoning, and Crystal Size Distribution (CSD) Theory to Natural Solid-Liquid Equilibria A dissertation submitted to the Graduate School of the University of Cincinnati in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Ph.D.) in the Department of Geology of the College of Arts and Sciences by Gokce K. Ustunisik B.S., Ankara University, 2001 M.S., Ankara University, 2004 August 2009 Committee Chair: Dr. Attila Kilinc OVERVIEW During the last decade, insightful applications of thermodynamic calculations embodied within the MELTS thermodynamic model of Ghiorso and Sacks (1995) have greatly enhanced our understanding of the evolution of magmas. MELTS does the isobaric and isothermal constrained calculations by minimizing the Gibbs free energy for the system; temperature, pressure, fO2 constrained calculations by minimization of the ⎛ ∂G ⎞ Korzhinski potential ( L = G − n O2 ⎜ ⎟ ) for systems open to oxygen transfer. The ⎝ ∂nO2⎠ adiabatic calculations are done by minimization of the enthalpy subject to fixed pressure and entropy (e.g. heat content). Applications of the MELTS calculations demonstrated in several publications have shown how complex petrologic hypotheses can be tested to yield quantitative and occasionally suprising results. This doctorate research focuses on developing quantitative models for crystal fractionation, magma recharge, magma mixing, compositional zoning in plagioclase by applying the MELTS algorithm to natural solid-liquid equilibria, and the application of Crystal Size Distribution (CSD) theory to compute crystal residence time in magma. This dissertation is composed of three chapters. In Chapter 1, the initial system parameters of parental magma (pressure, temperature, water content, and oxygen fugacity) of the Small Hasandag volcano in Central Turkey were constrained using the MELTS algorithm and then using the same algorithm, the feasibility of isobaric fractional crystallization, magma recharge, and isobaric-isenthalpic magma mixing was tested as the controlling process in the evolution of the parental magma. iii In Chapter 2, the consequences of two different physical models of magma dynamics on plagioclase zoning were determined. The consequences of magma pooling at several levels within the crust before eruption (decompressional and isobaric fractional crystallization) were explored. This process can produce normal, reverse and even oscillatory zoning in plagioclase. In another model, the effect of convection within a shallow magma chamber on plagioclase zoning was explored and was demonstrated that oscillatory zoning develops under these conditions. In Chapter 3, the CSD theory was used to calculate the crystal nucleation rate, crystal growth rate and crystal residence time using the plagioclase and clinopyroxene present in a single basaltic lava flow at the Small Hasandag volcano. Although crystal residence times using a single mineral in a single lava flow has been studied using the CSD theory, it is not clear if the same crystal residence times can be obtained if more than one mineral is used. The results of this part of my research show that residence times calculated from the CSD theory gives the same crystal residence times whether a single mineral or more than one mineral is used in the calculations. The results show that plagioclase and clinopyroxene residence times overlap within the limits of error implying that crystal residence times calculated by using the CSD theory either clinopyroxene or plagioclase of the Small Hasandag volcano can be used in the calculation of residence time. iv v ACKNOWLEDGEMENTS I would like to gratefully acknowledge the assistance of all people who made a significant contribution to this dissertation. I am deeply grateful to my advisor for his patience, enthusiasm, advice, encouragement, and continuous support over the past four years without which this dissertation would not have come alive. I have greatly benefited from his experience and knowledge which helped me to learn how to think, how to define the problems, and how to approach them along my work. I would like to thank to my committee members Dr. John Grover, Dr. David Nash, Dr. Warren D. Huff, Dr. J. Barry Maynard, and Dr. A. Umran Dogan for their constructive criticism and valuable advice that were always essential to the completion of this work. I especially thank to Dr. J. Barry Maynard and Dr. Tammie Gerke for their help during the XRF analysis. I should thank to the organizations that courteously provided funding for the research and travel. These include the Geological Society of America, the Department of Geology at the University of Cincinnati for “Wycoff” and “Geology Alumni Distinguished Doctoral” Fellowships, the University Research Council at the University of Cincinnati, and the Graduate Student Governance Association at the University of Cincinnati. I also thank to Dr. Arnie Miller and Dr. Lewis Owen for their encouragement and support during this work. A special thanks to Ana Cristina Londoño not only for her great friendship and continuous encouragement but also for her patience and invaluable time for her critics in my writings. I would also like to thank to Dr. Irem Yesilyurt, Murat Akkus, Onur Conger, Sebnem Tosun, and Yalin Senyurt for their help in the field. Many vi thanks to my friends who made an unforgettable and wonderful experience my life in Cincinnati. Last but not least, I am always thankful to my parents who continually encouraged and supported me with a great patience, unconditional love, and understanding of all times. vii TABLE OF CONTENTS OVERVIEW……………………………………………………………………………...iii ACKNOWLEDGEMENTS………………………………………………………………vi LIST OF FIGURES…………………………………………………………………….....x LIST OF TABLES………………………………………………………………………xvi CHAPTER I: ROLE OF FRACTIONAL CRYSTALLIZATION, MAGMA RECHARGE, AND MAGMA MIXING IN THE DIFFERENTIATION OF THE SMALL HASANDAG VOLCANO, CENTRAL ANATOLIA, TURKEY Abstract……………………………………………………………………………….1 1. Introduction………………………………………………………………………...3 2. Mineralogy of Small Hasandag Volcanic Rocks…………………………………...5 3. Chemical Composition of Small Hasandag Volcanic Rocks……………………….6 4. MELTS Calculations to Constrain the Initial System Parameters………………….7 4.1. Parental Magma Composition………………………………………………...8 4.2. Initial System Pressure………………………………………………………..9 4.3. Initial System Oxygen Fugacity………………………………………………9 4.4. Initial System Water Content………………………………………………..10 5. Testing Isobaric Fractional Crystallization Hypothesis…………………………...11 6. Testing Magma Recharge Hypothesis…………………………………………….13 7. Testing Isobaric-Isenthalpic Magma Mixing Hypothesis…………………………14 8. Conclusions………………………………………………………………………..16 9. Acknowledgements………………………………………………………………..18 viii References……………………………………………………………………………18 CHAPTER 2: NEW INSIGHTS INTO THE PROCESSES CONTROLLING PLAGIOCLASE ZONING Abstract………………………………………………………………………………33 1. Introduction………………………………………………………………………34 2. Materials and Methods…………………………………………………………...37 3. Discussion of Results…………………………………………………………….39 3.1.MELTS Simulations to Demonstrate the Role of Total Pressure (Ptotal) and Water Content of the Melt (wt% H2O) on Plagioclase Zoning……………………..39 3.2. Normal and Reverse Zoning under Decompressional Crystallization Followed by Isobaric Cooling Conditions…………………………………………………………40 3.3.Oscillatory Zoning under Polybaric and Isothermal Convection Conditions……43 3.4. Partial Molal Volume Effect on Plagioclase Zoning and Calculation of Partial Molal Volumes of Na2O and CaO……………………………………………….......44 4. Conclusions………………………………………………………………………..46 5. Acknowledgements………………………………………………………………..48 References……………………………………………………………………………48 CHAPTER 3: CRYSTAL SIZE DISTRIBUTIONS (CSDs) in a BASALTIC FLOW AT THE SMALL HASANDAG VOLCANO, CENTRAL TURKEY: COMPARISON OF CALCULATED RESIDENCE TIMES WITH PLAGIOCLASE AND CLINOPYROXENE CRYTALS Abstract………………………………………………………………………………74 1. Introduction………………………………………………………………………75 ix 2. The Small Hasandag Volcano……………………………………………………77 3. The Principles of the Crystal size Distribution (CSD) Theory…………………..78 3.1. Open System at Steady-State……………………………………………………79 4. Methods-Measuring the CSDs…………………………………………………...80 5. Results……………………………………………………………………………82 5.1. Plagioclase CSDs…………………………………………………………..........82 5.2. Clinopyroxene CSDs……………………………………………………………84 5.3. Comparison of Residence Times………………………………………………..85 6. Conclusions………………………………………………………………………86 References……………………………………………………………………………87 APPENDIX………………………………………………………………………….98 x LIST OF FIGURES Chapter 1 Figure 1 Total alkali (Na2O+K2O wt %) vs. SiO2 (wt %) diagram (Le Bas et al. 1986) for the Small Hasandag volcanic rocks. Figure 2 Major oxides variation diagrams for the Small