Biogeochemistry of Marine Dissolved Organic Matter

Biogeochemistry of Marine Dissolved Organic Matter

Second Edition

Dennis A. Hansell

Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida

Craig A. Carlson

Department of Ecology Evolution And Marine Biology, University of California, Santa Barbara, California

Table of Contents

Cover image

Title page

Copyright

Dedication

List of Contributors

Foreword

Preface

Chapter 1: Why Dissolved Organics Matter: DOC in Ancient Oceans and Past Climate Change

Abstract

Acknowledgements

I Overview

II Marine Carbon Cycling

III Interpreting the Geological Past

IV Implications for Future Global Change?

Chapter 2: Chemical Characterization and Cycling of Dissolved Organic Matter

Abstract

Acknowledgments

I Introduction

II Isolation of DOM from Seawater

III Chemical Characterization of DOM

IV Links Between DOM Composition and Cycling

V Future Research

Chapter 3: DOM Sources, Sinks, Reactivity, and Budgets

Abstract

Acknowledgments

I Introduction

II DOM Production Processes

III DOM Removal Processes

IV DOM Accumulation

V DOM Reactivity

VI The Priming Effect

VII Microbial Community Structure and DOM Utilization

VIII DOC in the Ocean Carbon Budget

IX Summary

Chapter 4: Dynamics of Dissolved Organic Nitrogen

Abstract

Acknowledgments

I Introduction

II DON Concentrations in Aquatic Environments

III Composition of the DON Pool

IV Sources of DON to the Water Column

V Sinks for DON

VI Summary

Chapter 5: Dynamics of Dissolved Organic Phosphorus

Abstract

Acknowledgments

I Introduction

II Terms, Definitions, and Concentration Units

III The Early Years of Pelagic Marine P-Cycle Research (1884-1955)

IV The Pelagic Marine P-Cycle: Key Pools and Processes

V Sampling, Incubation, Storage, and Analytical Considerations

VI DOP in the Sea: Variations in Space

VII DOP in the Sea: Variations in Time

VIII DOP Pool Characterization

IX DOP Production, Utilization, and Remineralization

X Conclusions and Prospectus

Chapter 6: The Carbon Isotopic Composition of Marine DOC

Abstract

Acknowledgments

I Introduction

II Carbon Isotope Geochemistry Primer

III DOC Isotope Ratio Methods

IV Isotopic Composition of Bulk Marine DOC

V Isotopic Composition of DOM Constituents

VI Summary and Conclusions

Chapter 7: Reasons Behind the Long-Term Stability of Dissolved Organic Matter

Abstract

Acknowledgments

I Introduction: The Paradox of DOM Persistence

II The Environment Hypothesis

III The Intrinsic Stability Hypothesis

IV The Molecular Diversity Hypothesis

V Concluding Remarks

Chapter 8: Marine Photochemistry of Organic Matter: Processes and Impacts

Abstract

Acknowledgments

I Introduction

II Impact of Photochemistry on Elemental Cycles

III DOM Photolability Spectrum and Fate of Terrestrial DOM in the Sea

IV Impact of Photochemistry on Other Marine Processes

V Modeling Photochemical Rates and Impact on Marine Carbon Cycling

VI Future Directions

Chapter 9: Marine Microgels

Abstract

Acknowledgments

I Introduction

II What Are Polymer Gels?

III Structure, Properties, and Dynamics of Marine Polymer Gels

IV Phase Transition

V Marine Gels in the Atmosphere and Their Relevance for Cloud Formation

Chapter 10: The Optical Properties of DOM in the Ocean

Abstract

Acknowledgments

I Introduction

II UV-Visible Spectroscopy of DOM

III Sources of CDOM to the Marine Environment

IV Removal of CDOM in the Marine Environment

V Distribution

VI Conclusions and Future Research Needs

Chapter 11: Riverine DOM

Abstract

Acknowledgments

I Introduction

II Land Transport

III Riverine DOM Composition

IV Anthropogenic Influences

Chapter 12: Sediment Pore Waters

Abstract

Acknowledgments

I Preface

II Introduction

III Composition and Dynamics of Bulk Pore Water DOM

IV Composition and Dynamics of DOM at the Compound and Compound-Class Levels

V Modeling DOC Cycling in Marine Sediments

VI Controls on DOC Concentrations in Sediments

VII The Role of Benthic DOM Fluxes in the Ocean Carbon and Nitrogen Cycles

VIII Concluding Thoughts

Chapter 13: DOC in the Mediterranean Sea

Abstract

Acknowledgments

I Introduction

II DOC Distribution at Basin Scale

III The Role of DOC in Carbon Export

IV DOC Inventory and Fluxes

V DOM Stoichiometry

VI DOC Dynamics in the Med Sea, a Comparison with the Oceans

VII Summary

VIII Open Questions

Chapter 14: DOM in the Arctic Ocean

Abstract

Acknowledgments

I Introduction

II Composition of DOC Within the Arctic Ocean

III Distribution and Mass Balance of DOM

Chapter 15: Modeling DOM Biogeochemistry

Abstract

Acknowledgments

I Introduction

II Modeling Approaches

III Modeling the Role of DOM in Ocean Biogeochemistry

IV Lability in Focus: Concepts and Definitions

V Discussion

Index

Copyright

Academic Press is an imprint of Elsevier

32 Jamestown Road, London NW1 7BY, UK

225 Wyman Street, Waltham, MA 02451, USA

525 B Street, Suite 1800, San Diego, CA 92101-4495, USA

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK

Copyright © 2015, 2002 Elsevier Inc. All rights reserved

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher

Permissions may be sought directly from Elsevier's Science & Technology Rights Department in Oxford, UK: phone (+ 44) (0) 1865 843830; fax (+ 44) (0) 1865 853333; email: permissions@elsevier.com. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material

Notice

No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein

Library of Congress Cataloging-in-Publication Data

Biogeochemistry of marine dissolved organic matter/edited by Dennis A. Hansell, Craig A. Carlson. – Second edition.

pages cm

ISBN 978-0-12-405940-5 (hardback)

1. Seawater–Organic compound content. 2. Chemical oceanography. 3. Biogeochemistry. I. Hansell, Dennis A., editor. II. Carlson, Craig A., editor.

GC118.B56 2014

551.46'6–dc23

2014033432

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-0-12-405940-5

For information on all Academic Press publications visit our web site at http://store.elsevier.com

Dedication

For the support and balance that only family can provide we dedicate this book to our parents Paul and Rose Marie Hansell and David and Paula Carlson, our beloved spouses Paula and Alison and our children Allison and Rachel, and Matthew, Hayden and Sydney. The foundation they have provided us is, through this book, extended to coming generations of marine scientists.

List of Contributors

Rainer M.W. Amon     Department of Marine Sciences and Oceanography, Texas A&M University at Galveston, Galveston, Texas, USA

Thomas R. Anderson     National Oceanography Centre Southampton, Southampton, UK

Leif G. Anderson     Department of Chemistry and Molecular Biology, University of Gothenburg, Gothenburg, Sweden

Sandra Arndt     School of Geographical Sciences, University of Bristol, Bristol, UK

Steven R. Beaupré     Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA

Karin M. Björkman     Daniel K. Inouye Center for Microbial Oceanography: Research and Education, Department of Oceanography, School of Ocean and Earth Science and Technology, University of Hawaii, Honolulu, Hawaii, USA

Deborah A. Bronk     Virginia Institute of Marine Science, College of William & Mary, Virginia, USA

David J. Burdige     Department of Ocean, Earth and Atmospheric Sciences, Old Dominion University, Noroflk, Virginia, USA

Craig A. Carlson     Department of Ecology, Evolution and Marine Biology, University of California, Santa Barbara, California, USA

James R. Christian     Fisheries and Oceans Canada, Canadian Centre for Climate Modelling and Analysis, Victoria, British Columbia, Canada

Thorsten Dittmar     Research Group for Marine Geochemistry (ICBM-MPI Bridging Group), Institute for Chemistry and Biology of the Marine Environment (ICBM), University of Oldenburg, Oldenburg, Germany

Kevin J. Flynn     Centre for Sustainable Aquatic Research, Swansea University, Swansea, UK

Dennis A. Hansell     Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida, USA

David M. Karl     Daniel K. Inouye Center for Microbial Oceanography: Research and Education, Department of Oceanography, School of Ocean and Earth Science and Technology, University of Hawaii, Honolulu, Hawaii, USA

David J. Kieber     College of Environmental Science and Forestry, Department of Chemistry, State University of New York, Syracuse, New York, USA

Tomoko Komada     Romberg Tiburon Center, San Francisco State University, Tiburon, California, USA

Caroline Leck     Department of Meteorology, University of Stockholm, Stockholm, Sweden

Kenneth Mopper     Department of Chemistry and Biochemistry, Old Dominion University, Norfolk, Virginia, USA

Norman B. Nelson     Earth Research Institute, University of California Santa Barbara, California, USA

Mónica V. Orellana     Polar Science Center, University of Washington/Institute for Systems Biology, Seattle, Washington, USA

Peter A. Raymond     Yale School of Forestry and Environmental Studies, New Haven, Connecticut, USA

Daniel J. Repeta     Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA

Andy Ridgwell     School of Geographical Sciences, University of Bristol, Bristol, UK

Chiara Santinelli     Istituto di Biofisica, Pisa, Italy

Rachel E. Sipler     Virginia Institute of Marine Science, College of William & Mary, Virginia, USA

Robert G.M. Spencer     Department of Earth, Ocean and Atmospheric Science, Florida State University, Tallahassee, Florida, USA

Colin A. Stedmon     National Institute of Aquatic Resources, Technical University of Denmark, Charlottenlund, Denmark

Aron Stubbins     Skidaway Institute of Oceanography, Department of Marine Sciences, University of Georgia, Savannah, Georgia, USA

Foreword

As one of Earth’s largest exchangeable carbon reservoirs, similar in scale to atmospheric CO2, the biogeochemical behavior of marine dissolved organic matter (DOM) has major significance for the carbon cycle, climate, and global habitability. As Ducklow (2002) wrote in the first edition of this book: Oceanic DOM is now recognized as an important component of the biogeochemical system and possibly a barometer of global change. Accordingly, marine DOM research has undergone a renaissance, moving rapidly beyond issues of measurement methodology to critically important spatial-temporal mapping of DOC distribution in the global ocean (Hansell et al., 2009), and now poised to address the underlying regulatory mechanisms. Despite important strides, our understanding of the biogeochemical behavior of marine DOM is still in its exciting early exponential growth phase. Fundamental questions of the nature and sources of DOM and mechanisms of its production, transformation, and respiration remain unanswered. So, there is much to do, and there are fresh ideas and powerful new study tools, as reflected in this book. Below, I formulate some problems that I suggest must be solved to understand DOM behavior and to predict the future biogeochemical state of the ocean. I hope that some young—and not so young—scientists will take on the challenge to solve them.

The lifetimes of DOM constituents range from minutes to millennia: some are mineralized rapidly by heterotrophic microbes (labile; LDOM); others less readily (semi-labile; SLDOM); while an incredible diversity of molecules, 10¹²-10¹⁵ (Hedges, 2002) have accumulated over time to comprise the huge (~ 642 PgC) but intriguing refractory DOM pool. A challenge is to understand the biological and physicochemical forces that mediate and regulate the biogeochemical behavior of various DOM components.

Most oceanic DOM is ultimately derived from primary production, and owes some of its chemical complexity to it, as phytoplankton generate enormous molecular diversity at the expense of CO2 and just a few inorganic and trace nutrients in order to serve their diverse adaptive needs. Biochemical processing of primary production by genetically diverse bacteria further adds to the chemical complexity of DOM. Intriguingly, most of the biomass generated by primary production is particulate (POM) yet on average about one-half—but a variable fraction—of primary production becomes DOM within the upper mixed layer, assessed conservatively as bacterial carbon demand. The POM-DOM transition is a critical step in the flow of reduced carbon in the global ocean and the capacity of the ecosystem to retain elements in the upper ocean for air-sea exchange; yet we currently lack knowledge of the underlying mechanisms or the means of direct quantification. There is extensive literature showing that multiple physiological, biochemical, and trophic interactions cause the release of DOM from the particulate phase—including living organisms. Sloppy feeding and exudation were long believed to be the major mechanisms, but with discoveries of new links in the food web there has also been recognition of additional mechanisms of DOM production. The list now includes microbial ectohydrolase activity, viral lysis of phytoplankton and heterotrophic microbes, cellular release of transparent exopolymer particles and other gel particles, programmed cell death, and microbe- microbe antagonism. It is probable that all mechanisms—both those listed and unlisted—cause the POM-DOM transition, but their relative quantitative significance varies in time and location. In view of the critical importance of the POM-DOM transition for predictive models of carbon flow and sequestration in the ocean, I stress the need to develop quantitative methods and a better mechanistic understanding of the production of marine DOM.

The utilization side of the marine DOM dynamics is deceptively simple, since essentially all DOM uptake is due to bacteria and Archaea, the dominant osmotrophic heterotrophs. The strength of this coupling is a critical variable in the regulation of DOM utilization and respiration, and subsequent air-sea exchange of CO2. Tight coupling also prevents excursions in DOM concentration, so this regulation has major ecological and biogeochemical implications. How do microbes manage to biochemically couple so tightly with primary production; and what biochemical and behavioral (e.g., chemokinesis) mechanisms regulate the nature and strength of bacteria-organic matter coupling? What is the role of microbial genetic diversity and biogeochemical expressions in maintaining the strength of coupling?

Genomic predictions have provided powerful constraints. They tell us the molecular interactions among DOM molecules and microbes that are possible. However, predicting the biogeochemical dynamics of the complex DOM pool also requires ecophysiological and biochemical studies of DOM-bacteria interactions to determine: what DOM transformations do take place, at what rates, by what biochemical mechanisms, subject to what regulatory forces and in what ecosystem context. This is indeed a tall order; but the problem is critical to solve because of the central role of DOM-bacteria interactions in predicting the carbon cycle of the future ocean. Ocean acidification and warming are likely to affect the nature and rates of microbial production and transformation of DOM with potential influence on the carbon balance between the ocean and the atmosphere. Understanding what renders some of the DOM semi-labile or refractory will also require such mechanistic studies. This important research on dissolved phase carbon cycling and sequestration requires new methods, model systems, and concepts (e.g., Microbial Carbon Pump; Jiao et al., 2010) addressing in situ dynamics and interactions among microbes and (DOM) molecules.

Method refinement has been an important goal in marine DOM research. The field was transformed in the 1990s by the fundamental discovery that DOM was measurably dynamic, contrary to earlier thinking that the DOM pool was inert. This discovery changed everything. Interestingly, chemists and microbiologists reached this conclusion by different paths. Chemists worked diligently to refine the DOC method (in spite of or perhaps because of initial setbacks; Hedges, 2002; Sugimura and Suzuki, 1988), achieving ~ 1 μM precision, sufficient to show DOC gradients over days to seasons and across locations and depths (Hansell et al., 2009). Marine microbiologists had been finding as early as the 1960s that ¹⁴C labeled amino acids and sugars added to seawater as metabolic tracers were readily assimilated and respired with lifetimes of hours to days. Clearly, a DOM fraction represented by the radiotracers was highly dynamic. Remarkably, the seemingly opposite views of DOM lability coexisted for a decade. As it turned out, the microbiologists had been observing the behavior of a tiny but dynamic labile DOM (~ 1 μM) embedded within an ~ 40 to 70-fold larger recalcitrant DOM pool. Thus, the divergent impressions of the shape of the elephant were due to the enormous range of lifetimes of DOM components. This problem of the biogeochemical behaviors of labile and recalcitrant DOM, and implications for dissolved phase carbon sequestration, is an active research area, formalized as the Microbial Carbon Pump (Jiao et al., 2010).

While high precision measurements of DOC concentrations (Sharp, 2002) has transformed marine DOM research, further method refinement and new method development is a high priority. First, achieving 1-2 μM precision still requires a magic touch; we need a plug and play method that even a microbial oceanographer could use! Second, experimental studies of DOM uptake, respiration, and sequestration require yet higher analytical precision. An order of magnitude improvement may even enable measurements of bacterial utilization of semi-labile DOC (lifetime 1.5 years; Hansell, 2013) albeit requiring long incubations. By analogy, consider if the scientists studying ocean acidification could only measure seawater pH with precision of 0.1 units! (While Dennis Hansell was waiting for me to finally finish this Foreword, X-Prize worth $2 M was announced for precise and user-friendly pH instrument: http://www.xprize.org/prize/wendy-schmidt-ocean-health-xprize.) Ultra-precise DOC, DON, and DOP methods will transform marine DOM biogeochemistry and climate predictions. Finally, we need a standardized method to measure microheterotrophic (bacteria + Archaea) respiration that does not significantly perturb the process being observed. These prokaryotes essentially monopolize DOM and their carbon growth efficiency is low, typically 10-30% (i.e., 70-90% of the labile DOM-C is respired by heterotrophic microbes). It is currently debated whether or not respiration and primary production are in balance or instead display spatial patterns of imbalance related to oligotrophic versus meso-/eutrophic systems (Ducklow and Doney, 2013).

The DOM problem has been studied for the better part of a century against significant methodological odds, yet significant advances have been made in the last 10-20 years that have enriched the field. Today, there is strong interdisciplinary convergence and integration, as marine chemists, organic geochemists, analytical chemists, microbiologists and molecular biologists, and modelers join to address big questions of carbon cycling and sequestration, climate predictions, and the biogeochemical state of the future ocean. This multifaceted pursuit has also led to the emergence of modern biogeochemistry as a distinct and dynamic field that is maturing rapidly and attracting students and postdocs as well as accomplished scientists from other fields. This advance of marine biogeochemistry as a discipline responsive to challenges posed by climate change may well be the most important development that has been stimulated by research on marine DOM.

Farooq Azam, Scripps Institution of Oceanography, University of California, San Diego

References

Ducklow HW. Foreword. In: Hansell DA, Carlson CA, eds. Biogeochemistry of Marine Dissolved Organic Matter. San Diego: Academic Press; 2002:xv–xix.

Ducklow HW, Doney SC. What is the metabolic state of the oligotrophic ocean? A debate. Ann. Rev. Mar. Sci. 2013;5:525–533.

Hansell DA. Recalcitrant dissolved organic carbon fractions. Ann. Rev. Mar. Sci. 2013;5:421–445.

Hansell DA, Carlson CA, Repeta DJ, Schlitze R. Dissolved organic matter in the ocean: new insights stimulated by a controversy. Oceanography. 2009;22:52–61.

Hedges J. Why dissolved organics matter. In: Hansell DA, Carlson CA, eds. Biogeochemistry of Marine Dissolved Organic Matter. San Diego: Academic Press; 2002:1–33.

Jiao N, Hernd GJ, Hansell DA, Benne R, Kattner G, Wilhelm SW, et al. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nat. Rev. Microbiol. 2010;8:593–599.

Sharp JH. Analytical methods for total DOM pools. In: Hansell DA, Carlson CA, eds. Biogeochemistry of Marine Dissolved Organic Matter. San Diego: Academic Press; 2002:35–58.

Sugimura Y, Suzuk Y. A high-temperature catalytic oxidation method for the determination of non-volatile dissolved organic carbon in seawater by direct injection of a liquid sample. Mar. Chem. 1988;24:105–131.

Preface

Dennis A. Hansell; Craig A. Carlson

Efforts by the ocean science community to understand cycling of the major bioactive elements (C, N, P) in the ocean have experienced great success in the past two decades and continues today. Intensive focus on elemental cycling resulted from society’s need to determine the role of the ocean in climate change, in turn requiring an understanding of its essential functions today, in the past, and in the future. The fundamentals of the biological processes involved in the transformations of the major elements have been identified. The next phase of research requires linking the biological processes to the very large oceanic reservoirs of the major elements, and identifying the sensitivities of those links. Establishing a linkage between processes and reservoirs falls into the discipline of biogeochemistry.

One of Earth’s largest exchangeable reservoirs of carbon is dissolved organic matter (DOM) in the ocean. With a stock of 662 PgC, the pool approximates the amount of carbon resident in atmospheric CO2. Prior to the 1990s, this major pool of carbon was primarily evaluated from a geochemical viewpoint; resolving the biochemical and isotopic composition of the pool was a central goal. With an enhanced biogeochemical perspective on DOM, the scientific questions began to rest broadly on the role of DOM in the oceanic C, N, and P cycles. Central questions were: can we accurately measure the concentrations of DOM in the ocean; what are the distributions of the dissolved organic C/N/P pools and what processes controls these distributions; what are the rates, biogeographical locations, and controls on elemental cycling through the pools; what are the biological and physicochemical sources and sinks; what is the composition of DOM and how does that illuminate elemental cycling? Finally, do we understand DOM in elemental cycling well enough to accurately represent the processes in models of the modern, past, and future oceans?

In this edition, the progress of the last decade in answering these questions is reported and synthesized by key contributors to those advances. The book opens with a chapter by A. Ridgwell and S. Arndt, setting the paleoceanographic context for marine DOM. Study of the chemical and isotopic compositions of DOM has provided unique information on elemental cycling; both subjects have seen great growth over the past decade. These works are reviewed in chapters by D. Repeta and S. Beaupre, respectively. The biological cycling of the major elements (C, N, P) through DOM is reviewed in chapters by C. Carlson and D. Hansell (focusing on biological processes and carbon budgets), D. Karl and K. Björkman (focusing on P), and R. Sipler and D. Bronk (focusing on N). Particular emphasis is placed in these chapters on marine microbes as active agents in the processing of DOM. Photochemical reactivity of DOM, and implications for elemental cycling, is presented by K. Mopper, D. Kieber, and A. Stubbins. The contributions of optically active (chromophoric) DOM are covered by C. Stedmon and N. Nelson. The book continues with chapters covering DOM at ocean interfaces and in marginal seas (i.e., the rivers draining into the ocean, the sediments, the Arctic Ocean, and the Mediterranean Sea, in chapters by P. Raymond and R. Spencer, D. Burdige and T. Komada, L. Anderson and R. Amon, and C. Santinelli, respectively). Topics new to this edition include marine microgels, introduced by M. Orellana and C. Leck, and the long-term stability of marine DOM by T. Dittmar. The book closes with the advances of DOM in ecosystem and global circulation models by T. Anderson, J. Christian and K. Flynn.

Many in the ocean science community have developed a strong biogeochemical view of the ocean. This book serves as a tool to provide foundation for their forays into the biogeochemistry of marine organic matter. The book maintains a particular focus on DOM in elemental cycling, and therefore does not revisit the many, well-documented advances made in organic geochemistry during the previous decades. Attention is largely to the marine environment, with little coverage of the fresh water systems other than the important rivers and dynamics adding DOM to the coastal seas. The book is directed at professional ocean scientists and advanced students of biological and chemical oceanography.

Many individuals and organizations must be thanked for support of the science that provided content for this book, as well as to development of the book itself. The U.S. federal agencies supporting much of what has been reported here, including individual research by the chapter authors, are the National Science Foundation (NSF), the National Oceanographic and Atmospheric Administration (NOAA), and the National Aeronautics and Space Administration (NASA). The agency program managers most important in supporting research conducted by the editors are Don Rice, David Garrison, and Eric Itsweire at NSF, David Legler and Joel Levy at NOAA, and Diane Wickland and Paula Bontempi at NASA. We greatly appreciate their leadership and support, and hope they are proud of these accomplishments.

Chapter 1

Why Dissolved Organics Matter

DOC in Ancient Oceans and Past Climate Change

Andy Ridgwell; Sandra Arndt    School of Geographical Sciences, University of Bristol, Bristol, UK

Abstract

Earth history is punctuated by a huge variety of transitions and perturbations in climate and global biogeochemical cycles. Some of these events exhibit evidence for greenhouse warming and CO2 release and hence potentially provide clues regarding future changes. Negative spikes and transients in the isotopic composition of carbon in sedimentary rocks bear the testimony of these carbon cycle perturbations and are generally taken to indicate the introduction of carbon with a relatively depleted isotopic signature into the ocean and atmosphere. One reservoir of isotopically depleted carbon is the dissolved organic carbon (DOC) present in the ocean. This has led to the idea that both the cycling of DOC and its reservoir size could have been fundamentally different in the past and that change in the oceanic DOC cycle may be mechanistically linked to major events in the geological record. This chapter provides an overview of how the proposed link between DOC and major global carbon cycle perturbation events in the geological record arises. We start by presenting a summary of the traditional view of how the marine carbon cycle operates. We then introduce how the geological record is interpreted, focusing on the ways in which the carbon isotopic signature of sedimentary rocks reflects past changes in global carbon cycling. We go on to critically assess recent thinking on the potential role of DOC as a driver for extreme climate events before touching on the future implications of a DOC cycle active on geological timescales.

Keywords

Geological record

Marine carbon cycling

Palaeoclimate

Carbon isotopes

Dissolved organic carbon (DOC)

Contents

I Overview   1

II Marine Carbon Cycling   2

A A Tale of Three Ocean Carbon Pumps   3

B A Fourth Appears—The Microbial Carbon Pump   6

III Interpreting the Geological Past   8

A Carbon Isotopes as Proxies for Past Global Carbon Cycle Changes   8

B Reconstructing Past Steady-State Modes of Global Carbon Cycling   10

C Interpreting Transient Carbon Cycle Perturbations   11

D Ocean DOC and Ancient Carbon Cycling: An Example from the Paleocene and Eocene   13

E Ocean DOC and Ancient Carbon Cycling: An Example from the Precambrian   14

IV Implications for Future Global Change?   16

Acknowledgements   17

References   18

Acknowledgements

AR and SA acknowledge support from The Royal Society in the form of University Research Fellowship, and the Natural Environmental Research Council in the form of a Postdoctoral Fellowship.

I Overview

The ocean and its underlying sediments are the largest sinks of CO2 within the Earth system that are able to respond to changes in atmospheric CO2 on both human-induced (anthropogenic) and geologically relevant time scales. We need a complete understanding of their dynamics and strength of feedbacks with climate and other drivers of global change if we are to make confident projections regarding the full consequences of continued fossil fuel CO2 emissions. To this end, in the past couple of decades, we have made rapid progress in elucidating the roles that basic physical (e.g. ocean circulation) and inorganic geochemical processes (e.g. gas exchange) play in regulating the uptake of CO2 from the atmosphere. In contrast, the role and response of the biological carbon pump—the interplay of biological, geochemical, and physical processes that transfer carbon from the surface ocean where it is fixed by primary producers to the depths where it is either consumed or buried in the underlying sediments—is much less well understood. Even the sign of some of the main feedbacks involved and whether the response of the biological carbon pump will act to amplify or reduce future climate warming and ocean deoxygenation is somewhat uncertain.

Two research directions are providing new insights into the marine carbon cycle and how it responds to perturbation. The first revolves around ongoing efforts to understand the mechanistic operation of the biological carbon pump and, in particular, the cycling of carbon through dissolved organic carbon (DOC) in the ocean, as described in this book. The second is an increased appreciation of what might be learned from the geological record. Earth history is punctuated by a huge variety of transitions and perturbations in climate and global biogeochemical cycles, with some events exhibiting evidence for greenhouse warming and CO2 release and hence potentially providing clues regarding future changes. The conjunction of these two developments has led to the idea that both the cycling of carbon through DOC and its reservoir size could have been fundamentally different in the past and that changes in that cycling may be mechanistically linked to major events in the geological record. The breath of speculation about how the marine carbon cycle may have operated during the Precambrian (prior to 541 Ma) and under very different conditions of oxygenation and ecosystem function from today also highlights the importance of first being able to ground our geologic interpretation in a full mechanistic understanding of the sources and sinks of DOC in the modern ocean.

This chapter provides an overview of how the proposed link between DOC and major global carbon cycle perturbations in the geological record arises. We start by presenting a brief summary of how the marine carbon cycle operates. We then introduce how the geological record can be interpreted, focusing on the ways in which the carbon isotopic signature of sedimentary rocks reflects past changes in global carbon cycling. We finish by critically assessing recent thinking regarding the potential role of DOC dynamics as a driver or amplifier of extreme climate events in the past as well as the potential future implications.

II Marine Carbon Cycling

Geological rock reservoirs dominate the global inventory of carbon on Earth (Figure 1.1). However, the response time for the formation or any substantive depletion of these reservoirs is counted in 10s, if not 100s, of millions of years, being largely governed by plate tectonics and major biological evolutionary innovations such as the advent and proliferation of calcifying plankton (Ridgwell, 2005). At ~ 38,000 PgC (1 PgC = 10¹⁵ gC), the present-day ocean dissolved inorganic carbon (DIC) reservoir is the next largest carbon store on Earth and is an order of magnitude larger than the likely extractable resources of fossil fuel carbon or the terrestrial biosphere (and soils). One way to influence atmospheric pCO2 and climate via greenhouse warming is to create an imbalance in the inputs versus sinks of carbon to the ocean. However, the response time of the ocean plus atmosphere as a whole—calculated as its carbon inventory (Figure 1.1) divided by the rate of carbon throughput from weathering and mantle CO2 outgassing (Figure 1.2b and c)—comes out to be of the order of 100,000 years (100 ky). Perturbations of the global carbon cycle that change the DIC inventory of the ocean as a whole in this way are hence only arguably relevant on geological time scales. Ocean alkalinity (and pH) influence the speciation of DIC (between CO2(aq), HCO3−, and CO3² −), but similar time scales apply if one wishes to change the alkalinity (ALK) inventory of the ocean as a whole.

Figure 1.1 Illustration of the primary reservoirs and processes constituting the (natural) global carbon cycle. Approximate carbon inventories (brackets) and representative carbon isotopic compositions (bold and in parentheses) are shown for the main reservoirs and fluxes. Adapted from Hönisch et al. (2012) with representative isotopic compositions from Maslin and Thomas (2003) and inventories from IPCC (2007).

Figure 1.2 Schematic workings of the four components of the ocean’s carbon pump. (a) The solubility. (b) The particulate organic carbon pump in the ocean. (c) The carbonate pump in the ocean. (d) The cycle of dissolved organic carbon in the ocean and the microbial carbon pump. Fluxes of DOC are broken down into semi-labile (light gray) and semi-refractory (dark gray), neither of which are transported far into the ocean interior. Fluxes of refractory DOC (RDOC) are indicated with black arrows. Reservoir inventories, in brackets, are from Hansell (2013) . In all figures, carbon fluxes into the ocean are shown in normal font and sinks in italics. (a–c) These are quantified following Hönisch et al. (2012) in units of PgC year − 1 .

In addition to changes in the total reservoir size, DIC (and alkalinity) can also be partitioned spatially within the ocean. This is important as the atmosphere only sees the surface ocean. The rapid exchange of CO2 between ocean surface and atmosphere (Figure 1.2a) means that atmospheric pCO2 and hence climate are highly responsive to changes in surface DIC which gives the ocean a unique importance to the global carbon cycle and climate system for reasons beyond its relative total carbon mass.

A A Tale of Three Ocean Carbon Pumps

The processes that control the partitioning of DIC between the surface ocean and ocean interior as well as its speciation at the surface are traditionally divided into three conceptual pumps in the ocean (e.g., Sarmiento and Gruber, 2006): (1) the solubility pump, (2) the organic matter (organic carbon or soft tissue) pump, and (3) the carbonate (or counter) pump, illustrated in Figure 1.2. In this conceptual framework, all three oceanic carbon pumps act primarily by vertically relocating carbon away from the surface, hence the concept of the pumping of carbon from the surface to depth. At the same time, the latter two also partition alkalinity (see Stumm and Morgan, 1981 or Zeebe and Wolf-Gladrow, 2001 for details) with depth, thus dictating the speciation of DIC (the balance between CO2(aq), HCO3−, CO3² −) to maintain overall charge neutrality. Modulating all three and hence influencing atmospheric pCO2 are the large-scale patterns and rates of ocean circulation that control the return of DIC, ALK, and nutrients back to the surface.

The operation of the solubility pump in partitioning DIC between surface and the deep ocean is the simplest to conceptualize (Figure 1.2a). CO2 is more soluble in colder compared to warmer seawater. As a consequence, carbon dissolution is enhanced in cold, high-latitude surface waters, where deep waters form. The rapid sinking of these dense, DIC-rich surface waters and their redistribution in the deep ocean results in efficient pumping of carbon from the atmosphere to the deep ocean. Apart from temperature, the solubility of CO2 is also affected by salinity. Thus, differences in salinity between major water masses and between surface and deep ocean will also make a contribution. Nevertheless, the contribution of salinity is generally minor compared to the effect of temperature.

The organic carbon (organic matter or soft tissue) pump also acts to induce a vertical DIC gradient (Figure 1.2b). In this case, DIC is removed from solution as a result of photosynthesis and is fixed in the form of both particulate organic carbon (POC) and DOC. A proportion of POC escapes the intense recycling in the surface ocean, sinking vertically through the ocean interior where heterotrophic degradation results in a release of DIC. In general, the heterotrophic degradation of POC during sinking is highly efficient, with < 1-6% of the POC export production reaching the seafloor, where of this flux, just 0.3% escape benthic degradation to be buried (Dunne et al., 2007). The further the POC sinks before the associated carbon is released back to the DIC pool, the more effective the partitioning between surface and deep ocean. Any reduction (or increase) in the efficiency of this degradation during sinking, perhaps as a result of changes in sinking speed or heterotrophic degradation rate, thus has the potential to affect pCO2. The influence of the organic carbon pump is partially offset by ALK changes associated with nitrogen transformations, particularly from NO3− to organic N during photosynthesis and from organic N to NH4+ during heterotrophic degradation (Zeebe and Wolf-Gladrow, 2001), which at the surface has the effect of partitioning more DIC in the form of CO2(aq). In addition, the organic carbon pump controls nutrient cycling in the ocean. It creates nutrient-rich deep waters that are redistributed by global circulation and upwelling and thereby influences the spatial distribution of primary production in the ocean.

The third of the traditional carbon pumps, the carbonate pump, involves the biological precipitation of calcium (and to a lesser extent, magnesium) carbonate (CaCO3) at the ocean surface. Perhaps as much as 50% of carbonate dissolves as it sinks (Feely et al., 2004). The remainder reaches the sediment surface where ~ 20% of the deposited carbonate escapes further dissolution to be buried (Feely et al., 2004), ultimately forming a new geological carbon reservoir (Figure 1.1). Although the carbonate pump also vertically redistributes carbon, the transfer of ALK dominates in terms of its influence on the speciation of dissolved carbon species and overall acts to increase, not decrease, surface ocean pCO2.

By virtue of their influence on surface ocean pCO2, all three of the pumps play an important role in the short-term carbon cycle. We define short-term here as the approximate mixing time scale of the ocean (nominally ~ 500-1000 years) and before the geochemical redistribution initially arising from any transient change in any of the C pumps will have been effectively homogenized by ocean mixing. On short time scales, the organic carbon pump dominates over the carbonate pump in terms of net influence on atmospheric pCO2, simply as a result of its order-of-magnitude larger export flux. The organic matter pump also dominates over the solubility pump (Cameron et al., 2005).

All three pumps are also important components of the long-term carbon cycle, driving the accumulation, biogeochemical transformation, and burial of carbon in marine sediments, directly in the example of the two solid pumps and indirectly in controlling the saturation state (all three pumps) and oxygenation (organic carbon and solubility pumps) of bottom water. For instance, under conditions conducive to the preservation of biogenic calcium carbonate (CaCO3)—primarily the shells of haptophyte algae (coccolithophores) and zooplankton (foraminifera)—sediments can become rich in CaCO3, with large amounts of carbon locked away for millions of years in the form of chalk deposits. As mentioned earlier, the sediment preservation of organic matter is generally less efficient, with only a tiny fraction of the organic matter that is exported from the surface ocean ultimately buried. Nevertheless, organic carbon burial rates can vary significantly in space and time. Particular intervals of Earth history, for instance the Cretaceous (145.5-65.5 My ago (Ma)) and Jurassic (201.3-145.5 Ma), saw a widespread deposition of organic carbon rich (> 1 wt%) strata. The factors that may explain this spatial and temporal variability in organic carbon deposition and burial rates are a matter of debate. They include, but are not limited to, organic matter composition, terminal electron acceptor, and in particular oxygen availability, benthic community composition, microbial inhibition by specific metabolites, priming, physical protection, deposition rate, and macrobenthic activity (see review by Arndt et al., 2013). In addition, an enhanced deposition and burial of organic matter may promote the production of methane (CH4) in the anoxic sub-seafloor. In conjunction with cold temperatures and/or high pressure, hydrates can form molecules of CH4 trapped in a water-ice cage structure (Maslin et al., 2010). Methane hydrates are special among the geological reservoirs in that they also influence the global carbon cycle and climate on short time scales if the environmental conditions change sufficiently rapidly and substantially to destabilize the hydrate structure and release the CH4 gas. An increase in ocean temperatures and depressurization by slumping and loss of the overlying sediment burden are possible drivers of significant and rapid CH4 release (e.g., see Maslin et al., 2010).

The last glacial maximum (LGM; about 21,000 years ago) is a good example of the role that the ocean carbon pumps play in regulating atmospheric pCO2.. At the time of the LGM, the concentration of CO2 in the atmosphere was about 190 ppm compared to ~ 260 ppm at the start of the Holocene (11,700 years ago) and ~ 280 ppm just prior to the industrial revolution (Monnin et al., 2001). An enduring mystery is why?, or in other words, in what way did the global carbon cycle function differently, resulting in lower atmospheric CO2 during glacial periods (Kohfeld and Ridgwell, 2009). Aside from ideas revolving around lower glacial ocean temperatures and hence an increase in the strength of the solubility pump, a greater supply of eolian iron to surface ocean ecosystems and hence partial relief of iron limitation in regions such as the Southern Ocean may have played a role (Watson et al., 2000). Iron deposition works its magic on atmospheric CO2 via an increase in plankton productivity and POC export and hence a stronger organic matter pump in the ocean—increasing the partitioning of DIC away from the surface and to depth. The biological pump could have been more efficient as well as stronger. This is an important distinction to make. In terms of lowering atmospheric pCO2, it is not sufficient to simply export POC away from the surface more rapidly through increased biological export; the CO2 that is eventually released through heterotrophic degradation needs to be kept away from the ocean surface and hence the atmosphere for as long as possible. For instance, carbon released from the degradation of POC in a more stratified glacial deep ocean (Adkins et al., 2002) would be less effectively mixed back towards the surface, while colder ocean temperatures could potentially have reduced the degradation rate of settling POC (e.g., Crill and Martens, 1987; Jørgensen and Sørensen, 1985; Matsumoto et al., 2007), pushing the average depth at which DIC is released to greater depths. Both processes would act to increase the efficiency of the biological carbon pump, even if a side effect was decreased export from the surface due to reduced nutrient return. That the glacial CO2 question has yet to be settled, despite the existence of a wide variety of hypotheses (Kohfeld and Ridgwell, 2009), points to a still incomplete understanding of the marine carbon cycle and leaves the potential for new mechanisms and processes to be discovered.

B A Fourth Appears—The Microbial Carbon Pump

Appreciation of the microbial carbon pump (Figure 1.2d), a part of the biological pump involving the net production of DOC by microbial processing of organic matter, has increased rapidly over the past decade (Jiao et al., 2011). It represents a further mechanism for repartitioning carbon among the non-geological (surficial) reservoirs. Alongside the production of POC and particulate inorganic carbon (PIC) by marine phytoplankton at the surface ocean, a variable but substantial fraction of primary production (30-50%) is released as DOC (e.g., Biddanda and Benner, 1997; Ducklow et al., 1995). DOC is also created by a variety of other activities in the surface ocean, including viral lysis, sloppy feeding by metazoan grazers, the egesta of protists and metazoa (e.g., Jiao et al., 2010), and the initial extracellular hydrolysis step involved in heterotrophic POC degradation. The most labile fractions of DOC are consumed rapidly and account for a mere 1% of the total ocean DOC inventory (Hansell, 2013). Only the more biologically recalcitrant DOC fractions last long enough to be transported into the deep ocean by ocean circulation. Even so, few of these newly produced, recalcitrant DOC fractions survive to reach depths of 1000 m (Hansell, 2013). The degradation product (DIC) will still continue to be advected into the ocean interior, however, helping partition DIC into the ocean interior.

At the highly refractory end of the reactivity spectrum, refractory DOC (RDOC) is removed only on multi-millennial time scales and hence on average survives multiple cycles of ocean turnover. Presumably, direct exudation during bacterial production, viral lysis of microorganisms or microbial necromass further contribute to the refractory pool. Microbial activity thus transfers, by the successive processing of DOC, organic carbon from low concentrations of reactive carbon to progressively higher concentrations of refractory carbon (hence the pump designation; Jiao et al., 2010). Despite its relatively low rate of production—perhaps only 0.043 PgC year− 1 (Hansell, 2013) as compared to 1-10 PgC year− 1 for PIC and POC export (Figure 1.1)—the most refractory DOC component accumulates in the ocean with estimated inventories of 630 ± 12 PgC for RDOC and > 12 PgC for an ultra-refractory (e.g., black carbon) DOC (Hansell, 2013). As such, RDOC constitutes a reservoir of similar size to the atmospheric burden of carbon present as CO2 (Figure 1.1). Thus, although the existence of this refractory carbon pool has been known for decades (e.g., Hedges, 2002), its potential size and thus its potential significance for carbon sequestration and past and future climate change has only been recognized recently (e.g., Jiao et al., 2010).

In theory, any substantial change in the consumption (or degradation) rate of this DOC pool could exert a large effect on atmospheric pCO2 and climate. A variety of environmental factors such as oxygen availability and ecosystem or temperature changes could potentially trigger substantial changes (see several other chapters in this book for insights on the processes that control the composition, reactivity, production, and removal of DOC in the ocean). In particular, the availability of oxygen often plays a central role in the discussion of substantial and large-scale changes in organic carbon consumption rates on climate-relevant time scales. Although the jury is still out on the direct effect of oxygen availability on heterotrophic organic carbon degradation, preservation of organic carbon tends to increase under anoxic conditions (e.g., Canfield et al., 1993). This increase could be the result of a thermodynamically limited degradation of refractory organic compounds in the absence of the powerful electron acceptor oxygen, the inability of anaerobic organisms to directly and completely degrade organic matter to CO2, a decreased enzymatic activity, and/or a decreased availability of DOC adsorbed to mineral surfaces under anoxic conditions (e.g., see review by Arndt et al., 2013). Therefore, large-scale ocean redox changes could, in theory, result in a larger global ocean DOC inventory during periods of widespread ocean anoxia. This pool of DOC would then be rapidly oxidized to DIC once oxic conditions are re-established, releasing CO2 to the atmosphere. This hence provides a new mechanism to alter the DIC inventory of the whole ocean and one that can operate with a response time much faster than the ~ 100 ky time scale driven by imbalances between weathering on land and sediment burial at the ocean floor. The possibility that such DOC reservoir variations occurred in the geological past is the subject of the remainder of this chapter.

III Interpreting the Geological Past

The strength and efficiency of the biological pump is controlled by a plethora of biological, geochemical, and physical factors, such as ecosystem composition, the reactivity and fate of organic matter, oxygen levels of the ocean’s interior, and the cycling of nutrients (Hain et al., 2014). These factors interact with and respond to changes in climate. The resulting complexity makes it extremely challenging to make projections about the consequences of continuing global warming on the marine carbon cycle. The geological record helps here as Earth’s history is punctuated by a huge variety of transitions and perturbations in global biogeochemical cycles and climate. These perturbations may exhibit evidence for greenhouse warming and CO2 release (Hönisch et al., 2012), with some associated with major extinctions or biotic disruption. Such events potentially provide calibrations regarding, for instance, the response of marine ecosystems to ongoing global change (e.g., Gibbs et al., 2006, 2012), for which model-based projections are difficult.

A Carbon Isotopes as Proxies for Past Global Carbon Cycle Changes

To understand past carbon cycle feedbacks and biotic sensitivities, we need to know how key environmental conditions have changed, such as atmospheric CO2 concentrations, surface temperatures (Dunkley Jones et al., 2013), ocean pH and carbonate saturation (Hönisch et al., 2012), or depletion of the ocean’s oxygen inventory (Keeling et al., 2010). However, reconstructing environmental variables is an imperfect science that becomes increasingly difficult as one goes further back in time. To about 800 ky ago (800 ka), ice cores drilled from the Antarctic ice cap reveal the composition of the ancient atmosphere (including, critically, pCO2) encoded in bubbles (Etheridge et al., 1996; Lorius et al., 1993). Additional environmental variables, such as the fluxes of dust and aerosols to the ice sheet surface, are recorded as impurities in the ice itself (Lambert et al., 2012). Beyond this, we have no direct record of the concentration of CO2 in the atmosphere and therefore have to rely on geological proxies.

Proxies (here, entities or measures representing the value of something else) are rooted in measurements made of some physical property in a geological sample such as, for instance, the mass fraction of a particular solid component (e.g., CaCO3) or a geochemical property, such as the ratio between different trace elements or isotopes of the same element (Wefer et al., 1999). The intent with proxies is to empirically (or in rare cases, theoretically) calibrate a physical or geochemical property against an environmental variable of interest. Such calibrations are based on a combination of laboratory manipulation experiments and correlations inherent in modern observations. For instance, the density of stomata—the physiologically controlled pores in the underside of leaves that regulate both CO2 ingress and water vapor loss—tends to decrease with increasing ambient pCO2 (Royer, 2001). The reason is that while plants require CO2, they equally need to conserve water. Higher atmospheric pCO2 simply means that fewer stomata are required by the plant in order to obtain the same diffusive flux of CO2. Fewer stomata provide the benefit of reduced water loss. Hence, the density of stomata, which can be counted in fossil leaves, should inverse correlate with concentration of CO2 in the atmosphere if the plant requires that water loss be minimized. However, for many other environmental variables no direct proxy may exist. Therefore, reconstruction of potential past environmental conditions and the testing of hypotheses often rely on more indirect proxies. The ratio of the stable isotopes of carbon recordedsedimentary rocks has been particularly importantthis respect and has been central to effortsreconstructing past global carbon cycling.

Carbon has two stable isotopes with masses 12 and 13 and a mean global abundance ratio of ~ 99:1. By convention, the 12:13 ratio in a sample is written in the delta-notation:

(1.1)

where Rsample is the ¹³C/¹²C ratio measured in the sample and Rstandard is the ¹³C/¹²C ratio in a substance of known composition (a standard). The sample isotopic composition (δ¹³Csample) is scaled by a factor of 1000 and given in units of per mil (‰).

Isotope fractionation—a change in the ratio of 12:13 during conversion of one carbon-containing substance to a second—can occur under a number of circumstances (see Chapter 6). First, at the same temperature and hence same kinetic energy, the lighter isotope (¹²C) has a higher velocity. During diffusion, the destination will become isotopically depleted (in ¹³C) while the source of carbon will become enriched. Second, covalent bonds involving ¹³C are stronger than those involving ¹²C. Hence, the photosynthetic splitting of CO2 and the breaking of bonds in organic molecules tends to occur to a greater extent for ¹²C, again leaving the residual enriched in ¹³C. The result of kinetic and particularly bond-breaking fractionation, as carbon moves between different carbon reservoirs and is transformed, is that the carbon reservoirs on Earth display a wide range of ¹²C:¹³C ratios. Typical δ¹³C values for the major carbon reservoirs are illustrated in Figure 1.1. The importance of bond breaking in creating a distinctly light (¹³C-depleted) composition of organic matter is obvious, and the additional fermentation step during methanogenesis gives rise to ever more depleted methane carbon—that is, there is a fractionation during photosynthesis and carbon assimilation as inorganic carbon is turned into organic molecules, and then a second stronger biological fractionation as organic matter is broken down and partly converted into CH4.

Because the various reservoirs are characterized by different values of δ¹³C, moving carbon from one to another will change not only the amount of carbon in the reservoirs but also potentially their δ¹³C. Repeated sampling of any one reservoir over time therefore enables input from (or loss to) a second reservoir to be inferred, even if changes in the size of either reservoir cannot be directly reconstructed. For instance, the isotopic composition of CO2 in the atmosphere has declined from a pre-Industrial (year 1765) value of about − 6‰ to a present-day value of about − 8.3‰, reflecting the release of isotopically depleted fossil fuel carbon (ca. − 27‰) to the atmosphere. Using this information together with a model representation of the exchange of carbon between atmosphere and ocean, it is possible to reconstruct the expected history of atmospheric pCO2 increases since the industrial revolution consistent with the observed δ¹³C changes. In this particular example, measurements of historical atmospheric pCO2 change together with some information on the rate at which fossil fuel carbon has been burned and terrestrial biomass degraded are directly available and changes in pCO2 do not have to be determined indirectly. However, in the absence of more direct measurements, δ¹³C changes recorded in dated plant remains (the proxy), for example combined with models, would have allowed us to reconstruct the history of atmospheric pCO2 and the associated time-varying emissions of fossil fuel CO2 to the atmosphere.

B Reconstructing Past Steady-State Modes of Global Carbon Cycling

The geological record of carbon isotopic variability, illustrated in Figure 1.3, provides important insights into past carbon cycling. For the Phanerozoic (542 Ma to present; Figure 1.3a), the records are primarily derived from measurements made on the calcareous shells and skeletons of shallow dwelling and reef-forming organisms, supplemented by analyses on much smaller carbonate shells of planktonic and benthic foraminifera for the interval for which ancient oceanic crust and its overlying sediment burden still exists (the last ~ 180 My). Organic matter may also help reconstruct δ¹³C trends, although marine carbonates (CaCO3) have the advantage of being much more abundant in sedimentary rocks than organic matter.

Figure 1.3 Carbon isotopic variability through Earth history. Shown are compilations of δ ¹³ C measured in marine carbonates for the Phanerozoic (a) ( Veizer et al., 1999 ) and for the Precambrian (b) ( Shields and Veizer, 2002 ). (c) A more detailed benthic foraminiferal calcite δ ¹³ C record spanning the Paleocene-Eocene boundary ( Zachos et al., 2010 ). The timing of hyperthermal events are indicated by red triangles. (d) A more detailed record spanning the mid-through-late Neoproterozoic ( Halverson et al., 2005 ; Macdondald et al., 2012 ). The approximate age of occurrence of glaciations are marked with inverted blue triangles. In the geological time scale at the top, the abbreviated Periods, from left to right are: Neogene (Ng), Paleogene (Pg), Cretaceous (K), Jurassic (J), Triassic (T), Permian (P), Carboniferous (C), Devonian (D), Silurian (S), Ordovician (O), and Cambrian (Cm).

Overall, changes in the global carbon cycle through the Phanerozoic (Figure 1.3a) are reflected in relatively slow, multimillion year transitions between higher and lower δ¹³C. These intervals of elevated (or depleted) δ¹³C can last 10s to 100s of million years and are generally ascribed to changes in the balance of organic versus inorganic (carbonate) carbon burial. This is because the surficial carbon reservoirs have insufficient capacity (Figure 1.1) to accumulate (or release) carbon at rates anywhere near comparable to weathering or sediment burial (Figure 1.2) for such durations. Assuming steady state, that is, time scales much longer than the ca. 100 ky residence time of δ) burial fluxes:

   (1.2)

where δ¹³Cinput is the isotopic signature of the average carbon input into the ocean (is the isotopic difference between the δ¹³C of buried organic matter versus that of CaCO3 (see, e.g., Kump and Arthur, 1999 for details). Because the system is expressed as a function of the observed δ¹³C of the carbonate sediments (δ¹³Cobs), for approximately invariant rates of carbon inputs and burial, Eq. (1.2) reconstructs Earth’s organic carbon burial history from the observed δ¹³C record (Figure 1.3). (One also needs to assume appropriate values for the characteristic isotopic composition of carbon inputs and outputs (δ) as summarized in Figure 1.1.) In other words, δ¹³Cobs can be used as a proxy for the global burial rate of organic matter, an important global carbon cycle flux that would otherwise be impossible to reconstruct.

As an example, the increasing δ¹³Cobs trend associated with the Carboniferous (359-299 Ma) in Figure 1.3a is commonly interpreted as reflecting increasing organic matter burial in response to the evolution of lignin-forming plants. Lignin is a relatively recalcitrant terrestrial compound that could have caused a step increase in the efficiency of organic carbon preservation (Berner, 2004), for which the extensive coal measures of the Carboniferous (a series of coal strata characteristic of the period) are considered as supporting evidence. Proxies such as leaf stomata density suggest that atmospheric CO2 declined to relatively low concentrations around this time (Royer, 2006) and a long interval of glaciation occurred—both consistent with an increase in organic carbon burial that sequestered CO2 from the ocean and atmosphere. The important lesson here is that large observed changes in δ¹³C may indicate fundamental reorganizations of the global carbon cycle.

C Interpreting Transient Carbon Cycle Perturbations

Superimposed on the general underlying trends of gradually increasing and decreasing δ¹³C are a wide variety of negative spikes and transients, illustrated in Figure 1.3c for the interval surrounding the Paleocene-Eocene boundary (ca. 56 Ma). Assuming that δ¹³C transient changes reflect changes in the δ¹³C of the seawater from which the carbonates are precipitated, rather than postdepositional diagenetic alteration, these negative spikes indicate that the ocean DIC and atmospheric CO2 reservoirs were being augmented by carbon with a distinct isotopic signature. In a simple mass balance approach, which deliberately ignores the fluxes into and out of the system unlike the approach above (Eq. 1.2), we can write:

(1.3a)

where the subscripts 0 and final represent the initial and final values of mass (M) and isotopic composition (δ¹³C) of the surficial carbon reservoir and ΔM and δ¹³CΔM are the magnitude of the added carbon and its isotopic composition, respectively. Equation (1.3a) simply states that the final mean isotopic composition of a carbon reservoir (here assumed to be surficial system of ocean + atmosphere + terrestrial biosphere ) can be taken to be equal to the mean isotopic composition of the initial reservoir plus that of the added (or removed) carbon, all weighted by their respective masses. For a given isotopic signature of the new carbon, Eq. (1.3a) can be rearranged, substituting Mfinal = M0 + ΔM (final isotopic composition equals the initial isotopic composition plus the recorded isotopic anomaly), to estimate the amount of carbon needed to explain a given magnitude of isotopic excursion (Δδ¹³C):

(1.3b)

This relationship is illustrated in . An important caveat to this analysis is that the carbon addition (or removal) is assumed to be instantaneous. That is, that the excursion is not significantly modified as a result of dilution caused by the continual throughput of carbon from terrestrial weathering to marine sedimentation (Figure 1.2) while the new carbon is still being added (or removed).

Figure 1.4 The relationship between carbon release, isotopic signature, and resulting excursion magnitude. Contours delineating the relationship between source magnitude and isotopic composition for a − 4‰ carbon isotopic ( δ ¹³ C) composition as given by Eq. (1.4). The assumed δ ¹³ C of the carbon source is on the x -axis, with the resulting inferred mass of carbon (in PgC) on the y -axis. Contours are plotted for initial mean surficial carbon reservoir δ ¹³ C values spanning the observed range of long-term Phanerozoic variability: − 2‰ (pink), 0‰ (orange), 2‰ (green), 4‰ (light blue), and 6‰ (dark blue).

The transient event that occurs at the boundary between the Paleocene and Eocene Epochs (Figure 1.3c) is a well-studied example of such a negative excursion event. The Paleocene (66-56 Ma) and Eocene (56-33.9 Ma) epochs are relatively warm intervals associated with elevated atmospheric pCO2 compared to modern (Hönisch et al., 2012) and a peak in warmth towards the early-middle Eocene followed by a long-term cooling trend cumulating in the emergence of substantial ice on Antarctica in the Oligocene (33.9-23.0 Ma). Tectonically, these intervals saw the movement of India toward and then collision with the Asian plate and the disappearance of the remnant ancient Tethys Sea, together with the gradual opening in the late Eocene and Oligocene of seaways separating Antarctica from South America and Australia. The Paleocene-Eocene boundary itself is marked by a prominent extinction among deep-sea foraminifera (Thomas, 2007), associated with a pronounced surface warming characterized by a global mean temperature increase of about 5 °C (Dunkley Jones et