![there and which are sailing vessels. Some of these days, perhaps, when the
right kind of moral shall have been drawn from broken propeller shafts and
twisted rudder-heads, the difficulty of distinguishing between the rig of a
sailing ship and the rig of a steamer may prove very much more
considerable than it now is; but, as this matter is at present ordered, the
towering masts, the immensely square yards, should leave even a
ploughman in no doubt as to the character of the vessels to which they
belong.
The number of sailing ships which crowd the docks on either side the
river must prove a real surprise to people who believe that it is all steam
nowadays. Let ancient mariners be consoled by this assurance: there is
plenty of steam indeed, but there is a deal of canvas too, so that all Jack’s
work does not lie in the bunkers yet, and there must still be a large demand
for seamanship of the old sort.
I am not sure that the wonder of the river does not owe quite as much to
the sailing ships as the steamers. The tall spars, the magnificent spread of
yards, the black lines of shrouds, the beautiful tracery of intersecting
running gear, added to the shapely hulls which support these towering
fabrics of hemp and steel and wood, make a most noble and impressive
sight, and give, so to speak, a final touch to the teeming, opulent,
commercial inspirations of the great river. Lower and lower yet down the
grand old stream the spirit of enterprise is settling, and the day is not far
distant when the projected dockyards at Tilbury will veritably transform the
quaint old town of Gravesend into the sea-gate of London. It is almost
startling to contemplate that time. One thinks of Gravesend now as a mere
break in the departure from the Thames. Will the chain of docks end at
Tilbury? At Gravesend, apparently, they are thinking otherwise! and
reckoning—somewhat against their own hopes—that if the Tilbury Docks
people play at leapfrog with the Albert Dock proprietors, the latter company
will repay the compliment and land themselves some distance lower down
yet. The limits of the Port of London, however, will, I believe, be reached
by within a quarter of a mile by the promoters of the Tilbury Dock
undertaking,[D] so that one cannot say in this case that there is room enough
for all. Unquestionably the docks which are nearest the sea will be the
docks best liked; and owners will profit at the expense of tug-masters and
pilots.](https://image.slidesharecdn.com/75196-250507213703-eb7ea1f2/85/Next-Generation-Biomonitoring-Part-1-1st-Edition-Eds-David-Bohan-51-320.jpg)
Next Generation Biomonitoring Part 1 1st Edition (Eds) David Bohan
- 1. Next Generation Biomonitoring Part 1 1st Edition (Eds) David Bohan download https://ebookgate.com/product/next-generation-biomonitoring- part-1-1st-edition-eds-david-bohan/ Get Instant Ebook Downloads – Browse at https://ebookgate.com
- 2. Get Your Digital Files Instantly: PDF, ePub, MOBI and More Quick Digital Downloads: PDF, ePub, MOBI and Other Formats and Forever Next Generation Chaos Bleeds 1 1st Edition Sam Crescent https://ebookgate.com/product/and-forever-next-generation-chaos- bleeds-1-1st-edition-sam-crescent/ Hacking The Next Generation 1st Edition Nitesh Dhanjani https://ebookgate.com/product/hacking-the-next-generation-1st- edition-nitesh-dhanjani/ Algorithms for next generation sequencing 1st Edition Sung https://ebookgate.com/product/algorithms-for-next-generation- sequencing-1st-edition-sung/ Fandom the Next Generation 1st Edition Bridget Kies https://ebookgate.com/product/fandom-the-next-generation-1st- edition-bridget-kies/
- 3. Next generation of data mining 1st Edition Hillol Kargupta https://ebookgate.com/product/next-generation-of-data-mining-1st- edition-hillol-kargupta/ The Next Generation of Corporate Universities Mark Allen https://ebookgate.com/product/the-next-generation-of-corporate- universities-mark-allen/ Next generation network services technologies and strategies 1st Edition Neill Wilkinson https://ebookgate.com/product/next-generation-network-services- technologies-and-strategies-1st-edition-neill-wilkinson/ Next Generation Leadership Insights from Emerging Leaders 1st Edition Sherry Penney https://ebookgate.com/product/next-generation-leadership- insights-from-emerging-leaders-1st-edition-sherry-penney/ Next Generation of International Chemical Additives 1st Edition Thomas F. Derosa https://ebookgate.com/product/next-generation-of-international- chemical-additives-1st-edition-thomas-f-derosa/
- 5. VOLUME FIFTY EIGHT ADVANCES IN ECOLOGICAL RESEARCH Next Generation Biomonitoring: Part 1
- 6. ADVANCES IN ECOLOGICAL RESEARCH Series Editors DAVID A. BOHAN Directeur de Recherche UMR 1347 Agro ecologie AgroSup/UB/INRA P^ ole GESTAD, Dijon, France ALEX J. DUMBRELL School of Biological Sciences University of Essex Wivenhoe Park, Colchester Essex, United Kingdom
- 7. VOLUME FIFTY EIGHT ADVANCES IN ECOLOGICAL RESEARCH Next Generation Biomonitoring: Part 1 Edited by DAVID A. BOHAN Directeur de Recherche UMR 1347 Agro ecologie AgroSup/UB/INRA P^ ole GESTAD, Dijon, France ALEX J. DUMBRELL School of Biological Sciences University of Essex Wivenhoe Park, Colchester, Essex, United Kingdom GUY WOODWARD Imperial College London, Ascot, Berkshire, United Kingdom MICHELLE JACKSON Imperial College London, Ascot, Berkshire, United Kingdom
- 8. Academic Press is an imprint of Elsevier The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 125 London Wall, London EC2Y 5AS, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States 525 B Street, Suite 1800, San Diego, CA 92101-4495, United States First edition 2018 © 2018 Elsevier Ltd. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability 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. ISBN: 978-0-12-813949-3 ISSN: 0065-2504 For information on all Academic Press publications visit our website at https:/ /www.elsevier.com/books-and-journals Publisher: Zoe Kruze Acquisition Editor: Jason Mitchell Editorial Project Manager: Joanna Collett Production Project Manager: Abdulla Sait Cover Designer: Mark Rogers Typeset by SPi Global, India
- 9. CONTENTS Contributors ix Preface xv Acknowledgements xix 1. Biomonitoring for the 21st Century: Integrating Next-Generation Sequencing Into Ecological Network Analysis 1 St ephane A.P. Derocles, David A. Bohan, Alex J. Dumbrell, James J.N. Kitson, François Massol, Charlie Pauvert, Manuel Plantegenest, Corinne Vacher, and Darren M. Evans 1. Introduction 3 2. How Are Ecological Networks Useful for Biomonitoring? 8 3. Ecological Networks Can Be Challenging to Build Using Conventional Approaches 14 4. Combining NGS With ENA: Opportunities and Challenges 18 5. Machine Learning as a Way to Rapidly Build Molecular Ecological Networks in a Rapid and Reliable Way? 28 6. NGS Network Data Sharing 32 7. Conclusion: Towards the Construction of Multilayer Networks in Ecology Using NGS 39 Acknowledgements 42 Glossary 42 References 44 Further Reading 62 2. Why We Need Sustainable Networks Bridging Countries, Disciplines, Cultures and Generations for Aquatic Biomonitoring 2.0: A Perspective Derived From the DNAqua-Net COST Action 63 Florian Leese, Agnès Bouchez, Kessy Abarenkov, Florian Altermatt, Ángel Borja, Kat Bruce, Torbjørn Ekrem, Fedor Ciampor Jr., Zuzana Ciamporová-Zaťovičová, Filipe O. Costa, Sofia Duarte, Vasco Elbrecht, Diego Fontaneto, Alain Franc, Matthias F. Geiger, Daniel Hering, Maria Kahlert, Belma Kalamuji c Stroil, Martyn Kelly, Emre Keskin, Igor Liska, Patricia Mergen, Kristian Meissner, Jan Pawlowski, Lyubomir Penev, Yorick Reyjol, Ana Rotter, Dirk Steinke, Bas van der Wal, Simon Vitecek, Jonas Zimmermann, and Alexander M. Weigand v
- 10. 1. State and Fate of Aquatic Ecosystems 65 2. Advancement of Aquatic Biomonitoring With a Focus on Europe 67 3. A DNA-Based Next Generation of Aquatic Biomonitoring? 72 4. The Grand Challenges for Next-Generation Aquatic Biomonitoring 75 5. The Aim of DNAqua-Net 87 6. Next-Generation Biomonitoring Opens New Doors 90 Acknowledgements 92 References 92 3. Advances in Monitoring and Modelling Climate at Ecologically Relevant Scales 101 Isobel Bramer, Barbara J. Anderson, Jonathan Bennie, Andrew J. Bladon, Pieter De Frenne, Deborah Hemming, Ross A. Hill, Michael R. Kearney, Christian K€ orner, Amanda H. Korstjens, Jonathan Lenoir, Ilya M.D. Maclean, Christopher D. Marsh, Michael D. Morecroft, Ralf Ohlem€ uller, Helen D. Slater, Andrew J. Suggitt, Florian Zellweger, and Phillipa K. Gillingham 1. Introduction 102 2. Factors Leading to Variable Microclimates 108 3. Organisms and Their Environment 117 4. Measuring Microclimates 119 5. Modelling Microclimates 140 6. Looking to the Future of Microclimate Ecology 146 7. Conclusions 148 References 150 Further Reading 161 4. Challenges With Inferring How Land-Use Affects Terrestrial Biodiversity: Study Design, Time, Space and Synthesis 163 Adriana De Palma, Katia Sanchez-Ortiz, Philip A. Martin, Amy Chadwick, Guillermo Gilbert, Amanda E. Bates, Luca B€ orger, Sara Contu, Samantha L.L. Hill, and Andy Purvis 1. Introduction 164 2. Designs of Studies for Assessing Biotic Impacts of Land-Use Change 166 3. Sampling Considerations 174 4. Manipulative vs Correlational Approaches 179 5. Challenges for Syntheses 180 6. Methods for Syntheses 185 7. Research Priorities 188 vi Contents
- 11. 8. Conclusions 189 Acknowledgements 189 Glossary 189 References 190 Further Reading 199 5. Modelling and Projecting the Response of Local Terrestrial Biodiversity Worldwide to Land Use and Related Pressures: The PREDICTS Project 201 Andy Purvis, Tim Newbold, Adriana De Palma, Sara Contu, Samantha L.L. Hill, Katia Sanchez-Ortiz, Helen R.P. Phillips, Lawrence N. Hudson, Igor Lysenko, Luca B€ orger, and J€ orn P.W. Scharlemann 1. Introduction: PREDICTS’ Scientific and Science-Policy Objectives 202 2. Key Design Decisions and Methods 208 3. Modelling Considerations 218 4. Summary of Findings 223 5. Synthesis and Prospects 230 Acknowledgements 233 References 234 6. Mapping Mediterranean Wetlands With Remote Sensing: A Good-Looking Map Is Not Always a Good Map 243 Christian Perennou, Anis Guelmami, Marc Paganini, Petra Philipson, Brigitte Poulin, Adrian Strauch, Christian Tottrup, John Truckenbrodt, and Ilse R. Geijzendorffer 1. Introduction: The Challenges of Monitoring Wetlands Status and Trends With Remote Sensing (RS) Data 244 2. Delineation and Separation of Habitat Types 250 3. Mapping the Water Dynamics of Wetlands 261 4. Detection of Trends Over Time 265 5. Conclusions 270 Acknowledgements 272 References 272 Cumulative List of Titles 279 vii Contents
- 12. This page intentionally left blank
- 13. CONTRIBUTORS Kessy Abarenkov University of Tartu, Tartu, Estonia Florian Altermatt Eawag, D€ ubendorf; University of Zurich, Z€ urich, Switzerland Barbara J. Anderson Manaaki Whenua Landcare Research, Biodiversity and Conservation Team, Dunedin, New Zealand Amanda E. Bates Ocean and Earth Science, National Oceanography Centre Southampton, University of Southampton, Southampton, United Kingdom Jonathan Bennie College of Life and Environmental Sciences, University of Exeter, Penryn, Cornwall, United Kingdom Andrew J. Bladon RSPB Centre for Conservation Science, The Lodge, Sandy, Bedfordshire, United Kingdom David A. Bohan Agro ecologie, AgroSup Dijon, INRA, University of Bourgogne Franche-Comt e, Dijon, France Luca B€ orger College of Science, Swansea University, Swansea, United Kingdom Ángel Borja AZTI, Pasaia, Spain Agnès Bouchez INRA UMR CARRTEL, Thonon-les-bains, France Isobel Bramer Faculty of Science and Technology, Bournemouth University, Poole, Dorset, United Kingdom Kat Bruce NatureMetrics, CABI Site, Surrey, United Kingdom Amy Chadwick University College London, London, United Kingdom Fedor Ciampor Zoology Lab, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Bratislava, Slovakia Zuzana Ciamporová-Zaťovičová Zoology Lab, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Bratislava, Slovakia ix
- 14. Sara Contu Natural History Museum, London, United Kingdom Filipe O. Costa Centre of Molecular and Environmental Biology (CBMA), University of Minho, Braga, Portugal Pieter De Frenne Forest and Nature Lab, Ghent University, Ghent, Belgium Adriana De Palma Natural History Museum, London, United Kingdom St ephane A.P. Derocles Agro ecologie, AgroSup Dijon, INRA, University of Bourgogne Franche-Comt e, Dijon, France Sofia Duarte Centre of Molecular and Environmental Biology (CBMA), University of Minho, Braga, Portugal Alex J. Dumbrell School of Biological Sciences, University of Essex, Colchester, United Kingdom Torbjørn Ekrem Norwegian University of Science and Technology, Trondheim, Norway Vasco Elbrecht Aquatic Ecosystem Research, University of Duisburg-Essen, Essen, Germany; Centre for Biodiversity Genomics, University of Guelph, Guelph, ON, Canada Darren M. Evans School of Natural and Environmental Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom Diego Fontaneto National Research Council of Italy, Institute of Ecosystem Study, Verbania Pallanza, Italy Alain Franc BIOGECO, INRA, Univ. Bordeaux, Cestas, and Pleiade Team, INRIA Sud-Ouest, Talence, France Matthias F. Geiger Zoologisches Forschungsmuseum Alexander Koenig, Leibniz Institute for Animal Biodiversity, Bonn, Germany Ilse R. Geijzendorffer Tour du Valat, Research Institute for the Conservation of Mediterranean Wetlands, Arles, France Guillermo Gilbert Natural History Museum, London, United Kingdom Phillipa K. Gillingham Faculty of Science and Technology, Bournemouth University, Poole, Dorset, United Kingdom x Contributors
- 15. Anis Guelmami Tour du Valat, Research Institute for the Conservation of Mediterranean Wetlands, Arles, France Deborah Hemming Met Office Hadley Centre, Exeter, Devon, United Kingdom; Birmingham Institute of Forest Research, Birmingham University, Birmingham, United Kingdom Daniel Hering Aquatic Ecology; Center of Water and Environmental Research (ZWU), University of Duisburg-Essen, Essen, Germany Ross A. Hill Faculty of Science and Technology, Bournemouth University, Poole, Dorset, United Kingdom Samantha L.L. Hill Natural History Museum, London; UN Environment World Conservation Monitoring Centre, Cambridge, United Kingdom Lawrence N. Hudson Natural History Museum, London, United Kingdom Maria Kahlert Swedish University of Agricultural Sciences, Uppsala, Sweden Belma Kalamuji c Stroil University of Sarajevo—Institute for Genetic Engineering and Biotechnology, Sarajevo, Bosnia and Herzegovina Michael R. Kearney School of BioSciences, The University of Melbourne, Melbourne, Australia Martyn Kelly Bowburn Consultancy, Durham, United Kingdom Emre Keskin Evolutionary Genetics Laboratory (eGL), Ankara University Agricultural Faculty, Ankara, Turkey James J.N. Kitson School of Natural and Environmental Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom Christian K€ orner Institute of Botany, University of Basel, Basel, Switzerland Amanda H. Korstjens Faculty of Science and Technology, Bournemouth University, Poole, Dorset, United Kingdom Florian Leese Aquatic Ecosystem Research; Center of Water and Environmental Research (ZWU), University of Duisburg-Essen, Essen, Germany xi Contributors
- 16. Jonathan Lenoir UR “Ecologie et dynamique des systèmes anthropis es” (EDYSAN, UMR 7058 CNRS- UPJV), Universit e de Picardie Jules Verne, Amiens, France Igor Liska ICPDR Permanent Secretariat, Vienna International Centre, Vienna, Austria Igor Lysenko Grand Challenges in Ecosystems and the Environment, Imperial College London, Ascot, United Kingdom Ilya M.D. Maclean College of Life and Environmental Sciences, University of Exeter, Penryn, Cornwall, United Kingdom Christopher D. Marsh Faculty of Science and Technology, Bournemouth University, Poole, Dorset, United Kingdom Philip A. Martin Conservation Science Group, University of Cambridge, Cambridge, United Kingdom François Massol CNRS, UMR 8198 Evo-Eco-Paleo, Universit e de Lille, SPICI group, Lille, France Kristian Meissner Finnish Environment Institute, General Director’s Office, Jyv€ askyl€ a, Finland Patricia Mergen Botanic Garden Meise, Meise; Royal Museum for Central Africa, Tervuren, Belgium Michael D. Morecroft Natural England c/o Mail Hub, County Hall, Worcester, Worcestershire, United Kingdom Tim Newbold Centre for Biodiversity and Environment Research, University College London, London, United Kingdom Ralf Ohlem€ uller University of Otago, Dunedin, New Zealand Marc Paganini European Space Agency, Frascati, Italy Charlie Pauvert BIOGECO, INRA, Univ. Bordeaux, Pessac, France Jan Pawlowski University of Geneva, Geneva, Switzerland Lyubomir Penev Pensoft Publishers, Sofia, Bulgaria Christian Perennou Tour du Valat, Research Institute for the Conservation of Mediterranean Wetlands, Arles, France xii Contributors
- 17. Petra Philipson Brockmann Geomatics Sweden AB, Stockholm, Sweden Helen R.P. Phillips German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig, Leipzig, Germany Manuel Plantegenest UMR 1349 IGEPP, INRA, Agrocampus-Ouest, Universit e de Rennes 1, Rennes Cedex, France Brigitte Poulin Tour du Valat, Research Institute for the Conservation of Mediterranean Wetlands, Arles, France Andy Purvis Natural History Museum, London; Grand Challenges in Ecosystems and the Environment, Imperial College London, Ascot, United Kingdom Yorick Reyjol AFB, The French Agency for Biodiversity, Direction de la Recherche, Vincennes, France Ana Rotter National Institute of Biology, Ljubljana, Slovenia Katia Sanchez-Ortiz Natural History Museum, London; Grand Challenges in Ecosystems and the Environment, Imperial College London, Ascot, United Kingdom J€ orn P.W. Scharlemann School of Life Sciences, University of Sussex, Brighton; UN Environment World Conservation Monitoring Centre, Cambridge, United Kingdom Helen D. Slater Faculty of Science and Technology, Bournemouth University, Poole, Dorset, United Kingdom Dirk Steinke Centre for Biodiversity Genomics, University of Guelph; University of Guelph, Guelph, ON, Canada Adrian Strauch University of Bonn, Center for Remote Sensing of Land Surfaces (ZFL), Bonn, Germany Andrew J. Suggitt University of York, York, Yorkshire, United Kingdom Christian Tottrup DHI GRAS, Hoersholm, Denmark John Truckenbrodt Friedrich-Schiller-University Jena, Institute of Geography, Jena, Germany Corinne Vacher BIOGECO, INRA, Univ. Bordeaux, Pessac, France xiii Contributors
- 18. Bas van der Wal STOWA, Stichting Toegepast Onderzoek Waterbeheer, Amersfoort, The Netherlands Simon Vitecek University of Vienna, Vienna, Austria; Senckenberg Research Institute and Natural History Museum, Frankfurt am Main, Germany Alexander M. Weigand Aquatic Ecosystem Research; Center of Water and Environmental Research (ZWU), University of Duisburg-Essen, Essen, Germany; Mus ee National d’Histoire Naturelle de Luxembourg, Luxembourg, Luxembourg Florian Zellweger Forest Ecology and Conservation Group, University of Cambridge, Cambridge, Cambridgeshire, United Kingdom; Swiss Federal Research Institute WSL, Birmensdorf, Switzerland Jonas Zimmermann Botanic Garden and Botanical Museum, Freie Universit€ at Berlin, Berlin, Germany xiv Contributors
- 19. PREFACE Biomonitoring the Earth’s ecosystems and their attendant communities, functions and ecoservices underpins decision making in many areas of policy and can have considerable value for the public, particularly in the case of species with high conservation value. In almost all cases, however, current biomonitoring approaches suffer from problems of accuracy, high costs that restrict coverage and limited generality. Biomonitoring schemes are also based upon methods developed in the early or middle part of the last century and have largely ignored subsequent advances in ecological theory and tech- niques, especially those derived from molecular ecology, remote sensing, network science and ecoinformatics. Consequently, the full diversity of functions and species in an ecosystem has rarely been evaluated. This is problematic because it only provides a partial view of the greater whole and cannot account for—or predict—the “ecological surprises” that com- monly arise through indirect food web effects in nature. In this two-volume Thematic Issue of Advances in Ecological Research focusing on Ecological Biomonitoring, we showcase some of the new biomonitoring approaches that have begun to appear in the last 15 years and that have started to tackle these problems directly; to generate the more sophisticated Next- Generation Biomonitoring (NGB) approaches, we will need to cope with our rapidly changing environment. Potentially, NGB could, even within the next decade, revolutionise our understanding of the functioning of Earth’s major ecosystems, allowing us to both measure and predict the effects of a range of abiotic stressors as well as those from the biotic sphere (e.g. species invasion and extinction), which will lead to better-informed and more effective management. Moreover, as they are often rooted in standardised, functional metrics, these approaches could potentially be applied at local to global scales, both accurately and cheaply. The first couple of papers of this two-volume Thematic Issue consider the role that new DNA-based approaches might play in the future of NGB. Derocles et al. (this issue) examine the potential that Next- Generation Sequencing (NGS) of environmental samples of DNA has to provide the means to rapidly build highly resolved species interaction net- works across multiple trophic levels. Their paper details how the analysis of multilayer ecological networks, constructed from NGS data, could be used to characterise the ecological mechanisms that underpin ecosystem xv
- 20. functioning and ecosystem service provision within future NGB frame- works. The authors propose that the future of network ecology and biomonitoring is extremely exciting given that the tools needed to build highly resolved multilayer networks are now finally within reach. In the subsequent paper, Leese et al. (this issue) place the current start of the art in environmental DNA sampling for NGB of freshwaters within the historical context of the limitations and strengths of traditional biomonitoring methods. The authors use a new research consortium, DNA-Aquanet, recently established and supported by the European Union, as the lens through which to view the development of the novel approaches that will augment—and ultimately supersede—current practices. They emphasise the fundamental differences in the traditional and NGB approaches, as well as highlighting some of the key areas of common ground, especially where there is scope for the “handshaking” and cross-calibration that is needed to form the bridge between the old and the new, thus pre- serving the value of the vast store of historical data that have already been amassed. The increase in capacity and a decrease in costs of molecular tools are discussed in relation to the far slower development of traditional methods. Unresolved issues the authors highlight include those that are still holding the field back, such as bioinformatics database errors, amplification bias and problems of estimating relative abundance across taxa from DNA data. These are discussed against the backdrop of end-user community iner- tia due to past investment in older biomonitoring approaches—a classic example of the “sunk cost fallacy”. Leese et al. then focus on the key advances that are now being made in NGB and how the DNA-Aquanet consortium is helping to drive those changes, in both the scientific and non- academic spheres. The paper has a strong applied focus, with a strong link to EU legislative frameworks, but this is complemented by the consideration of the role these new approaches could play in addressing fundamental questions in ecology, reshaping not just our current view of the world but also the questions we will be able to ask in the future. The final four papers of this volume are more explicitly practical in tone, emphasising the application of ecological approaches to biomonitoring and the measurement of ecosystem change. The paper by Bramer et al. (this issue) examines an important, but often overlooked, component of biomonitoring—the local microclimate that supports the focal organisms in the ecosystem of interest. Based on recent discussions from a British Ecological Society Open Workshop (organised by the Climate Change Ecology Special Interest Group), they provide a broad overview of recent xvi Preface
- 21. advances in microclimate monitoring and modelling, highlighting some of the key research challenges and solutions in this field, and scan the horizon for future developments. Ultimately, the spatiotemporal distribution of all organisms is largely controlled by their physiological tolerances to environ- mental conditions. Most research to date examines where and when species exist as a function of broad climatic envelopes operating over many kilometres. However, within these areas, local microclimates can vary immensely, even approaching the physiological limits of life for short periods or in particular patches within an otherwise seemingly benign land- scape. Without understanding microclimatic variability, our understanding of the controls of species distributions is limited, as is our ability to predict how climatic changes may reshape them. Bramer et al. (this issue) tackle these problems directly and provide recommendations for improving NGB and our understanding of the controls on species distributions. In the next paper, De Palma et al. (this issue) explore the strengths and weaknesses of the different study designs that are commonly used in biomonitoring in relation to land-use change, including space-for-time sub- stitution, time series, and before-after-control-impact design. Comparisons of data from different types of studies can be problematic, and different designs may even detect different trends in biodiversity change. The authors discuss how new syntheses can incorporate multiple study types to provide a new and more holistic perspective in NGB. To develop more realistic future projections of biodiversity change, they stress the need for a better under- standing of temporal dynamics. In conclusion, De Palma et al. call for more studies using a before-after-control-impact design, which are relevant for the widest range of questions related to NGB. These studies are still sur- prisingly rare, but disproportionately important because they can be used to validate or correct inferences from simpler designs. The paper by Purvis et al. (this issue) gives a detailed account of the pro- ject “Projecting Responses of Ecological Diversity In Changing Terrestrial Systems (PREDICTS)”. Since 2012, PREDICTS has collated abundance and occurrence data from thousands of sites facing different land-use pres- sures across the globe and now covers over 50,000 species in nearly 100 countries. In their paper, the authors discuss key design decisions for making predictions for biological diversity, including using space-for-time substitu- tion, and detail the modelling approaches they have used. The project focuses on site-level biodiversity data because many ecosystem functions and services depend on the local, rather than global, state of biodiversity. They emphasise how the PREDICTS database can be used to improve xvii Preface
- 22. global biodiversity assessments, which often rely on expert opinion or data from species representing only a small fraction of total global biodiversity (e.g. vertebrates). For instance, PREDICTS has implemented a version of the Biodiversity Intactness Index (BII) that is based on objective primary biodiversity data, rather than subjective expert judgement. This PREDICTS paper gives the most detailed overview of this large project to date and illustrates the value that tools, models, indicators and projections will have for biomonitoring and predicting change in global biodiversity. In the final paper of this issue, Perennou et al. (this issue) describe devel- opments and approaches to improve current space-borne remote sensing of ecosystems, using a case study from wetlands in the Mediterranean biodiver- sity hot spot. Given current challenges that affect wetlands, but which also have corollaries in the remote sensing of all ecosystems, of delineating and separating habitat types, mapping of the internal environmental dynamics and the detection of trends over time that need to be disentangled from natural background variability, Perrenou et al. argue that the solutions to improving current remote sensing approaches will only be achieved by allying the rapidly developing methodologies of remote sensing to ecolog- ical understanding of the ecosystems being monitored. These two volumes present a snapshot of some of the work currently being done in biomonitoring. The combination of papers across them reveals the huge value in using novel NGS, sensing and informatics approaches and better fusions of pure and applied disciplines to monitor and model how natural ecosystems will respond to the accelerating rates and increasing magnitude of environmental change we are already seeing across the globe. There is clearly plenty of exciting and challenging work still to be done, but this Thematic Issue illustrates some of the most impor- tant steps being taken towards developing the NGB approaches we will need to achieve a more sustainable future. ALEX J. DUMBRELL GUY WOODWARD MICHELLE C. JACKSON DAVID A. BOHAN xviii Preface
- 23. ACKNOWLEDGEMENTS David A. Bohan would like to acknowledge the support of the French Agence Nationale de la Recherche project NGB (ANR-17-CE32-0011) and FACCE SURPLUS project PREAR (ANR-15-SUSF-0002-03). xix
- 24. This page intentionally left blank
- 25. CHAPTER ONE Biomonitoring for the 21st Century: Integrating Next-Generation Sequencing Into Ecological Network Analysis St ephane A.P. Derocles*,1 , David A. Bohan*, Alex J. Dumbrell† , James J.N. Kitson‡ , François Massol§ , Charlie Pauvert¶ , Manuel Plantegenestk , Corinne Vacher¶ , Darren M. Evans‡ *Agro ecologie, AgroSup Dijon, INRA, University of Bourgogne Franche-Comt e, Dijon, France † School of Biological Sciences, University of Essex, Colchester, United Kingdom ‡ School of Natural and Environmental Sciences, Newcastle University, Newcastle upon Tyne, United Kingdom § CNRS, UMR 8198 Evo-Eco-Paleo, Universit e de Lille, SPICI group, Lille, France ¶ BIOGECO, INRA, Univ. Bordeaux, Pessac, France k UMR 1349 IGEPP, INRA, Agrocampus-Ouest, Universit e de Rennes 1, Rennes Cedex, France 1 Corresponding author: e-mail address: stephane.derocles@inra.fr Contents 1. Introduction 3 2. How Are Ecological Networks Useful for Biomonitoring? 8 2.1 Traditional Biomonitoring Is Typically Descriptive and Rarely Provides an Understanding of the Underlying Mechanisms Behind Ecosystem Functions 8 2.2 Ecological Networks Provide a Framework to Describe and Monitor Ecological Processes and Ecosystem Functions 9 2.3 Ecological Network Structure Characterizes Ecosystem Properties 10 2.4 Knowledge of Ecological Networks Helps to Assess the Effect(s) of Environmental Changes on Ecosystem Processes and Associated Services 11 2.5 The Robustness of Networks of Ecological Networks: Applications for Understanding Species and Habitat Loss, Restoration and Building Ecosystem Resilience 12 3. Ecological Networks Can Be Challenging to Build Using Conventional Approaches 14 4. Combining NGS With ENA: Opportunities and Challenges 18 4.1 Using NGS to Construct Ecological Networks 18 4.2 PCR Bias and Abundance Estimation in NGS Community Analyses 21 4.3 NGS Without a Prior PCR Step 22 4.4 Detection of Species Interactions Using Molecular Tools 23 4.5 How to Deal With Interactions Not Directly Resolved by NGS: Are Species Association Networks Species Interaction Networks? The Case of Microorganisms 24 Advances in Ecological Research, Volume 58 # 2018 Elsevier Ltd ISSN 0065-2504 All rights reserved. https://doi.org/10.1016/bs.aecr.2017.12.001 1
- 26. 5. Machine Learning as a Way to Rapidly Build Molecular Ecological Networks in a Rapid and Reliable Way? 28 5.1 Learning Ecological Networks From Data 28 5.2 Exploiting eDNA-Derived Information as a Source for Network Data 30 6. NGS Network Data Sharing 32 6.1 The Importance of a Dedicated NGS Network Database: Linking DNA Sequences and Ecological Interactions to Limit Species Identification Errors 33 6.2 Reconstructing Ecological Networks With Different Predicting Methods of Species Interactions 34 6.3 Do Only Sequences and Species Interactions/Cooccurrences Matter in a NGS Network Database? 35 6.4 An Example Output From a NGS Network Database: Phylogenetically Structured Networks 36 6.5 Improving Network Ecology Research With a NGS Network Database 38 7. Conclusion: Towards the Construction of Multilayer Networks in Ecology Using NGS 39 7.1 Towards Larger, Highly Resolved Networks 39 7.2 NGS Networks to Link Above- and Belowground Ecosystems, as Well as Eukaryotes and Prokaryotes 40 7.3 Biomonitoring of Ecosystems With Multilayer Phylogenetically Structured Networks 40 Acknowledgements 42 Glossary 42 References 44 Further Reading 62 Abstract Ecological network analysis (ENA) provides a mechanistic framework for describing complex species interactions, quantifying ecosystem services, and examining the impacts of environmental change on ecosystems. In this chapter, we highlight the importance and potential of ENA in future biomonitoring programs, as current bio- monitoring indicators (e.g. species richness, population abundances of targeted spe- cies) are mostly descriptive and unable to characterize the mechanisms that underpin ecosystem functioning. Measuring the robustness of multilayer networks in the long term is one way of integrating ecological metrics more generally into bio- monitoring schemes to better measure biodiversity and ecosystem functioning. Ecolog- ical networks are nevertheless difficult and labour-intensive to construct using conventional approaches, especially when building multilayer networks in poorly stud- ied ecosystems (i.e. many tropical regions). Next-generation sequencing (NGS) provides unprecedented opportunities to rapidly build highly resolved species interaction net- works across multiple trophic levels, but are yet to be fully exploited. We highlight the impediments to ecologists wishing to build DNA-based ecological networks and discuss some possible solutions. Machine learning and better data sharing between ecologists represent very important areas for advances in NGS-based networks. The future of network ecology is very exciting as all the tools necessary to build highly resolved multilayer networks are now within ecologists reach. 2 St ephane A.P. Derocles et al.
- 27. 1. INTRODUCTION Traditionally, community ecology tends to focus on patterns of spe- cies richness and community composition, while ecosystem ecology focuses on fluxes of energy and materials. Ecological networks (sometimes called food webs for trophic interactions), however, provide a quantitative frame- work to combine these approaches and unify the study of biodiversity and ecosystem function (Thompson et al., 2012). Ecological networks, which describe which species are interacting with which (i.e. qualitative networks) as well as the strength of their interactions (i.e. quantitative networks), are now routinely used to understand ecosystem ‘robustness’ to species extinc- tions (Evans et al., 2013; S€ aterberg et al., 2013), quantify ecosystem services (Derocles et al., 2014a; Macfadyen et al., 2009) or examine the impacts of environmental change (Morris et al., 2015; Thompson and Gonzalez, 2017; Tylianakis et al., 2007). By using a burgeoning range of metrics to describe network structure, complexity and stability (see Arnoldi et al., 2016; Bersier et al., 2002; Donohue et al., 2013; Dunne et al., 2002a,b), ENA is consid- erably improving our understanding of ecology and evolution, with a grow- ing number of applications for biomonitoring (Bohan et al., 2017; Gray et al., 2014). Indeed, ENA is increasingly being used to assess ecosystem response to environmental changes (e.g. climate change, pollution, invasive species; Aizen et al., 2008; Blanchard, 2015; Bohan et al., 2017; Thompson et al., 2016). There is consequently a growing shift in biodiversity monitoring away from conventional species and community-level descriptions towards a more comprehensive and mechanistic approach using species interaction networks (Bohan et al., 2013; Derocles et al., 2014a; Evans et al., 2013; Fontaine et al., 2011; Gray et al., 2014; Ings et al., 2009; K efi et al., 2012; Macfadyen et al., 2009; Pocock et al., 2012; Wirta et al., 2014). Nevertheless, ecological networks can be difficult to construct with conventional approaches and suffer some major pitfalls mainly centred on sampling issues, taxonomic misidentification and/or incorrect species inter- actions (Evans et al., 2016; Gibson et al., 2011). Major errors occurring in either of these steps could ultimately affect network-level structural metrics and thus our understanding of ecosystem functioning (Novak et al., 2011). DNA-based methods (based on combined taxonomic identification and interaction data from DNA sequences) have the potential to overcome many of these issues, providing large, highly resolved, phylogenetically structured networks suitable for rapid and reliable biomonitoring (Bohan et al., 2017; Evans et al., 2016; Vacher et al., 2016; Valentini et al., 2009b). 3 NGS Into Ecological Network Analysis
- 28. Today, next-generation sequencing (NGS) or high-throughput sequencing (see Goodwin et al., 2016 for a review) can rapidly generate mil- lions of DNA sequences. Sequences can describe, very precisely, not only the biodiversity present within an ecosystem, but also species interactions, the data from which can then be used to construct ecological networks (Evans et al., 2016). Recently, ecological network studies have taken advan- tage of NGS to successfully construct networks (e.g. Toju et al., 2014 for a plant–fungus network). Advances in statistical modelling and machine learn- ing approaches bring a new opportunity to predict species interactions and rapidly build multilayer ecological networks from DNA sequences data gen- erated with NGS (Vacher et al., 2016). Despite species identification from DNA sequences commonly being seen as a universal way to identify species (Hebert et al., 2003), the NGS technology to build food webs is not applied uniformly in network ecology. Experimental designs (field sampling and molecular protocols) and the con- struction of ecological networks are heavily dependent on the ecosystem studied, and particularly on the type of interactions (see Box 1). Here, we BOX 1 Species interactions in ecological networks Species interactions are a major component of ecosystem functioning. In ecological communities, a wide range of interactions can be described and visualized as ecological networks. These include direct and indirect interactions. Direct interactions relate to cases where a species directly affects another (i.e. species A impacts species B). Indirect interactions refer to cases where the impact of a species on another is mediated or transmitted by a third species (i.e. a first species A affects a second species B through an intermediary species C). In ecological networks, direct interactions are usually described and collec- tively shape the structure of the networks. However, this does not mean that indi- rect interactions are ignored as ecological networks are also used as a framework to study indirect interactions such as resource competition (Tilman, 1982), appar- ent competition (interactions through shared natural enemies; Derocles et al., 2014a; Holt, 1997; Morris et al., 2004; van Veen et al., 2006) or trophic cascades (Hairston et al., 1960; Oksanen et al., 1981). Indeed, indirect interactions result from the cooccurrence of several direct interactions. Hence, because the purpose of ecological networks is to describe the set of (direct) species interactions in an ecosystem, building networks constitutes a powerful approach to identify poten- tial indirect interactions. Within a network it is, for example, possible to detect shared natural enemies when searching for cases of apparent competition—a particular instance of three-node network motifs (Stouffer et al., 2007). Table 1 summarizes a general classification of direct ecological interactions. Although a wide range of interactions occur in nature, ecological network 4 St ephane A.P. Derocles et al.
- 29. BOX 1 Species interactions in ecological networks—cont’d Continued Table 1 Direct Interactions Between Two Species (According to Lidicker, 1979; see Faust and Raes, 2012) Type of Interaction Effect on Species A Effect on Species B Nature of Interaction Examples in Ecological Networks Mutualism Positive Positive Mutual benefits of the species Plant–pollinator, plant–ant, plant– seed disperser, plant–fungi Interference competition Negative Negative Species have negative effect on each other Trophic/ predation Positive Negative Predator gains at the expense of the prey, which is killed. We include here prey–predator, plant–consumer and host– parasitoid interactions Host–parasitoid, prey–predator, plant–herbivore Parasitism Positive Negative Parasite develops at the expense of the host, which is not killed Host–parasite, host–pathogen Commensalism Positive Null Species A is benefited, species B is not affected Amensalism Null Negative Species A has a negative effect on species B, but species A is not affected Neutralism Null Null Neither species is affected The interaction types studied in depth in network ecology are in bold. 5 NGS Into Ecological Network Analysis
- 30. distinguish two cases in particular. First, NGS can be directly used to build quantitative ecological interactions between organisms by resolving species interactions (e.g. Evans et al., 2016; Kitson et al., 2016; Piñol et al., 2014; Toju et al., 2013, 2014). This use of NGS data is, however, only possible in ecosystems in which relationships between organisms can clearly be established, such as host–parasitoid interactions where the parasitoid can be detected within the host (Derocles et al., 2014a, 2015; Wirta et al., 2014), prey–predator interactions by detecting prey in gut contents (e.g. Piñol et al., 2014; Tiede et al., 2016) or faeces (Clare et al., 2014; Zeale et al., 2011; see Symondson and Harwood, 2014) and plant–pollinator inter- actions by using high-throughput sequencing to identify the pollen carried (Bell et al., 2017; Galimberti et al., 2014; Pornon et al., 2016; Sickel et al., 2015). Second, there are systems in which it is impossible (or logistically very problematic) to detect interactions between organisms and assessing whether these interactions are positive or negative, such as those within microbial (Jakuschkin et al., 2016) or planktonic communities (Lima-Mendez et al., 2015). For these systems, NGS approaches can only identify cooccurring species and their relative abundance. NGS data then need to be combined with theoretical approaches, including statistical modelling (Faust and Raes, 2012) or machine learning (Bohan et al., 2011a), to predict species interac- tions from their abundance patterns and finally to build ecological networks (Bohan et al., 2017; Kamenova et al., 2017; Vacher et al., 2016). These two ways of building ecological networks have their own specificities and chal- lenges to overcome but also share common problems. These problems are BOX 1 Species interactions in ecological networks—cont’d studies to date have tended to focus on three types of interactions: parasitism, mutualism and trophic interactions. Other types of interactions have been studied (see Allesina and Levine, 2011; Coyte et al., 2015; Mougi, 2016 for net- works with competitive interactions), but they are relatively rare in comparison with the large majority of studies dealing with trophic and mutualist networks. A complementary classification was established by Pantel et al. (2017) accounting for the degree of interaction immediacy: whether the interaction takes place over a short or long part of an organism’s life cycle. This distinguishes, for example, parasitism from predation, scramble competition from contest competition and mutualistic symbiosis from external mutualism. However, when discussing species interactions in this chapter, we will be referring to parasitism, mutualism and trophic interactions. 6 St ephane A.P. Derocles et al.
- 31. related to (1) the qualitative and quantitative reliability of NGS data (i.e. polymerase chain reaction (PCR) bias and errors, sequencing bias and esti- mation of species abundances and frequency of interactions with number of NGS reads; Sommeria-Klein et al., 2016); (2) the identification of nodes and interactions in the network (inferring species interactions with statistical models when interactions are not directly resolved by molecular tools); (3) the costs of the sequencing technology and the expertise needed to pro- cess the data (Toju et al., 2013, 2014; Vacher et al., 2016). Here, we bring new insights on how to integrate NGS and ENA into biomonitoring (Fig. 1). We first consider why ecological networks provide a suitable framework for a better understanding of biodiversity and ecosys- tem functioning and how they can be used to complement or supersede con- ventional biomonitoring approaches. Second, we underline the challenges that ecologists face in building ecological networks when DNA-based tools are not available (which represent the vast majority of food web studies in the literature). Third, we demonstrate how molecular methods, NGS in par- ticular, can overcome (at least partially) the numerous constraints inherent in conventional network construction methodologies (e.g. taxonomic identi- fication, insect rearing, fieldwork issues), while considering the challenges of using NGS tools for building networks. Fourth, we give insights on how to overcome NGS data issues and efficiently build networks through machine learning and data sharing. Finally, we discuss new areas of research and development centred on ENA of multilayer networks to ultimately create more resilient ecosystems. Fig. 1 A road map to integrate NGS and ENA into biomonitoring. The successive steps are discussed in this chapter, and the corresponding sections are indicated in blue. Data stored and shared for an efficient biomonitoring are indicated in green. 7 NGS Into Ecological Network Analysis
- 32. 2. HOW ARE ECOLOGICAL NETWORKS USEFUL FOR BIOMONITORING? 2.1 Traditional Biomonitoring Is Typically Descriptive and Rarely Provides an Understanding of the Underlying Mechanisms Behind Ecosystem Functions Biomonitoring of change lies at the core of ecosystem conservation, manage- ment and restoration. As biomonitoring is an obligation today, biomonitoring programs are framed by government organizations (e.g. European Commis- sion, Joint Nature Conservation Committee in the United Kingdom). In its simplest form, biomonitoring consists of recording species diversity and abundances across different locations and times using a range of ecological census techniques and taxonomic identification. Most biomonitoring sam- pling methodologies were developed in the middle of the 20th century (Bohan et al., 2017) and were selected for entirely pragmatic reasons that reflected the current state of knowledge, simplicity and cost. Indicators are sampled to evaluate risks to human health and the environment for com- munication to the public or government policy makers. These include pes- ticide residues, elements and metabolites as pollution indicators, while abundances of target species or community descriptors are used to assess the ecological condition of ecosystems. However, these established meth- odologies are often of low generality. They are also often limited to par- ticular ecosystems of species and communities of study and may not allow comparison between different systems. The evaluation of the myriad of changes in ecosystems that can occur is simply too costly, time-intensive and not necessarily captured by current biomonitoring indicators. Consequently, biomonitoring of the full diversity of species and their interactions within an ecosystem is rarely, if ever, attempted (Bohan et al., 2017). While traditional biomonitoring is useful for simple conser- vation purposes such as identifying biodiversity hot spots or mapping ‘functional gaps’ in ecological communities (Forest et al., 2007; Myers et al., 2000; Raxworthy et al., 2003), such an approach is clearly not suited to the task of predicting the consequences of human actions that specifically target particular species or habitats. This is due to the fact that these human actions can have unintended consequences that spread through the net- work of species interactions at different spatial and temporal scales (Estes et al., 1998; Polis et al., 1997). For instance, traditional biomonitoring schemes have repeatedly failed at predicting the consequences of species 8 St ephane A.P. Derocles et al.
- 33. introductions and have only just begun to look for guidance in interaction network approaches (David et al., 2017; M edoc et al., 2017; Pantel et al., 2017). 2.2 Ecological Networks Provide a Framework to Describe and Monitor Ecological Processes and Ecosystem Functions Networks have become a prominent tool for studying community and eco- system ecology, as they serve as a generic, conceptual framework for under- taking research across a broad range of ecological systems. Ecological networks, famously described by Darwin as the ‘tangled bank’, describe the interactions between species, the underlying structure of communities and the function and stability of ecosystems (Montoya et al., 2006). Histor- ically, the ecologist Charles Elton pioneered the concepts of food chains and food webs, organizing species into functional groups (Elton, 1927; see also Cousins, 1987; Polis, 1991). These concepts formed the basis for ecologist Raymond Lindeman’s classic and landmark paper on trophic dynamics (Lindeman, 1942). The examination of networks has then been spurred by now classic studies such as the keystone predation experiments and theory (Paine, 1966, 1969, 1974), the complexity–stability debate (Gilpin, 1975; MacArthur, 1955; May, 1972, 1973a,b) and the search for invariant patterns linking, for example, species diversity with the number of links in food webs (Briand and Cohen, 1984; Cohen and Briand, 1984; Cohen and Newman, 1985; Cohen et al., 1990; Pimm, 1980; Stenseth, 1985; Williams and Martinez, 2000, 2004). The past decade in particular has seen significant advances in the theoretical understanding, construction, analysis and appli- cation of complex species interactions networks (see Fontaine et al., 2011; K efi et al., 2012 for reviews). This area of ecology has been marked by two trends: (i) the building of more sophisticated models aimed at predicting and/or explaining the structure of ecological networks based on a variety of mechanisms (e.g. Allesina et al., 2008; Bascompte et al., 2003; Canard et al., 2012; Dalla Riva and Stouffer, 2016; Ekl€ of et al., 2013; Jordano, 1987; Jordano et al., 2003; Lewis and Law, 2007; Rohr et al., 2016; Williams and Martinez, 2000; reviewed in Kamenova et al., 2017) and (ii) the search for more precise data (in particular, taxonomic identification), and practical methods to obtain them, that has chiefly been done to coun- teract the tendency to lump together insufficiently described species that reduces the ability to make predictions and identify food web invariants (Novak et al., 2011; Solow and Beet, 1998; Yodzis, 1998). More recently, 9 NGS Into Ecological Network Analysis
- 34. Thompson et al. (2012) proposed using ecological networks as a conceptual framework to reconcile biodiversity and ecosystem function studies. A network approach can be built on current biomonitoring schemes: if interaction data is collected alongside conventional monitoring of biodiversity, then it is possible to start monitoring both biodiversity and ecosystem functioning (see Mulder et al., 2006 for an example for soil microbial communities). For example, plant surveys could be comple- mented with insect flower visitation data to create plant–flower–visitor networks. Conversely, when pollinators are targeted by biomonitoring pro- grams, the pollen carried by the species could be identified and used to create pollen-transport networks. These complementary approaches could be implemented in traditional biomonitoring methodologies and would give a better understanding of ecological processes through the construction of networks. Taking a step further, a combination of NGS and ENA together could provide a radically new approach to understand how environmental change affects ecosystems. 2.3 Ecological Network Structure Characterizes Ecosystem Properties To measure changes in ecosystems, a wide range of metrics have been devel- oped to encapsulate the emergent architecture of the networks (see Bersier et al., 2002). ENA relies on a wide range of network descriptors to assess the effect of environmental changes on ecosystem function. Ma et al. (2017) dis- cuss descriptors of network complexity, such as connectance (a measure of network complexity), modularity (representing compartmentalization within the network) and nestedness (i.e. nodes with few connections linked to a subset of nodes interacting with more connected nodes) and their importance for detecting changes occurring in ecosystems (see Fortuna et al., 2010; Poisot and Gravel, 2014 for a critical view of some network metrics). Metrics of consumer–prey asymmetries are, in addition, very important to consider. The effect of environmental changes may vary across a food web (Thompson et al., 2012), and a change in an ecosystem may go undetected using measures of network complexity but nevertheless can affect consumer–prey asymmetries, with consequences on ecosystem func- tion and services. Such asymmetries can be described as ‘vulnerability’ and ‘generality’ introduced by Schoener (1989) as, respectively, the mean num- ber of consumers per prey and the mean number of prey per consumer within a food web. These consumer–prey asymmetry metrics are particu- larly well suited to the study of host–parasitoid networks (Derocles et al., 10 St ephane A.P. Derocles et al.
- 35. 2014a; Wirta et al., 2014). Other metrics of ecological network structure have also been proposed as determinants of ecosystem properties, such as the existence of fast and slow energy channels (Rooney et al., 2006), neg- ative relations between interaction strength and the length (Neutel et al., 2002) of the trophic loop it is part of, or the frequency of network motifs (Stouffer et al., 2007). Ecological processes such as pollination, pest control and seed dispersal are historically and still currently well studied in network ecology. These processes rely on mutualist and antagonist interactions with structural prop- erties that can be characterized with network descriptors. Mutualist net- works are, for example, often described as nested structures (Bascompte et al., 2003; Th ebault and Fontaine, 2010). Network structure thus consti- tutes an efficient indicator of pollination quality (Kaiser-Bunbury et al., 2017). Similarly, a compartmentalized (or modular) structure often emerges from antagonist networks (Derocles et al., 2014a; Ma et al., 2017). Compart- mentalized networks have important implications for natural pest control as they suggest a high specificity between the pest species and their natural ene- mies. With the current threat to food security (Godfray et al., 2010), ENA could help in our understanding of the underlying mechanisms involved in pest control and provide indicators to help agroecosystem management. Nevertheless, characterizing ecosystem properties through ENA must be done with caution. Most networks metrics are highly dependent on sam- pling completeness (see Bl€ uthgen et al., 2006; Jordano, 2016; Rivera-Hutinel et al., 2012). Consequently, the effort spent to sample and characterize an environment may directly affect the structure highlighted. Since DNA is ubiq- uitous in ecosystems, NGS constitutes a promising way to overcome the sam- pling completeness issues in ENA. 2.4 Knowledge of Ecological Networks Helps to Assess the Effect(s) of Environmental Changes on Ecosystem Processes and Associated Services Ecological networks are increasingly (but not systematically) used to assess the effects of environmental changes on ecosystems as they provide a more complete description of ecological processes than conventional community or species-oriented approaches. For instance, Tylianakis et al. (2007) dem- onstrated that habitat modification altered the structure of networks of cavity-nesting bees, wasps and their parasitoids. The altered network struc- ture had effects on parasitism rate, with consequences on ecosystem services such as pollination and biological control. A striking result from this study 11 NGS Into Ecological Network Analysis
- 36. was that, despite only little observed variation in species richness, marked changes arose in network structure. Evans et al. (2013) demonstrated in an organic farm model system that two particular seminatural habitats (rep- resenting less than 5% of total area of the farm) were disproportionately important to maintain the integrity of the overall network, and thus of the associated ecosystem services (i.e. natural pest control, pollination). More recently, Kaiser-Bunbury et al. (2017) showed that ecosystem resto- ration in mountaintop communities affects the network structure in a pos- itive way with a higher functional redundancy in restored communities. This modification of network architecture had direct and positive effects on the reproductive performance of the most abundant plant species. Thus, the development and application of ENA represent a paradigm shift in the biomonitoring of ecosystems (Kaiser-Bunbury and Bl€ uthgen, 2015). How- ever, empirical studies of this sort are still relatively rare in the literature, mainly because of the underlying network construction process. In partic- ular, theoretical links between network structure and ecological function need to be better established. In this context, ecological network modelling has made some impressive progress in the understanding of ecosystem func- tioning. For example, the allometric food web model designed by Schneider et al. (2016) established the link between the diversity of animal communi- ties and primary productivity. They demonstrated that diverse animal com- munities are more exploitative on plants but do not reduce plant biomass because this communities are composed of energetically more efficient plant and animal species. Network modelling such as the allometric food web model can therefore complement empirical studies. Consequently, more collaborative research between empirical and theoretical network ecologists is urgently needed and could be especially useful in helping to address a number global challenges, such as climate change, biodiversity loss and food security. 2.5 The Robustness of Networks of Ecological Networks: Applications for Understanding Species and Habitat Loss, Restoration and Building Ecosystem Resilience The study of network ‘robustness’ (Dunne et al., 2002a,b; Memmott et al., 2004) has grown rapidly in recent years, partly driven by advances in com- putational modelling (Kaiser-Bunbury et al., 2010; Staniczenko et al., 2010), but mostly by the objective of understanding the threat of biodiversity loss to ecosystem services and functioning (Astegiano et al., 2015; Pocock et al., 2012). Studies have progressed from simple qualitative, bipartite mutualistic 12 St ephane A.P. Derocles et al.
- 37. networks (Memmott et al., 2004), to investigations of patterns across eco- systems (Srinivasan et al., 2007) and to current quantitative approaches that take into account species abundance (Kaiser-Bunbury et al., 2010). Pocock et al. (2012) constructed and analysed a ‘network of ecological networks’ (i.e. 11 groups of animals interacting with shared plants on farm- land), providing new analytical tools for understanding both the conse- quences of species extinctions across multiple animal groups, and the potential for ecological restoration. The study provided a method to calcu- late the relative importance of plants, and thus identified some plants that were disproportionately important in the network of networks (i.e. com- mon agricultural plants such as clover Trifolium and thistle Cirsium spp.). Although yet to be tested empirically, one application of this approach is that important plants could be targets for conservation and restoration that would benefit multiple animal groups. By examining the robustness of the joined networks, the study found that animal groups varied in their robustness to sequences of plant extinction, with the plant–pollinator network exhibiting much lower robustness than the seed-feeding bird network. Therefore, using a network approach, it should be possible to identify more sensitive groups for targeted conservation effort and/or assessment for biomonitoring rather than spending limited funds on charismatic species. Evans et al. (2013) developed this approach further by modelling the cascading effects of habitat loss, driven by plant extinctions, on the robustness of multiple animal groups. Habitat robustness analysis identified two seminatural habitats (i.e. waste ground and hedgerows together comprising 5% of the total area of the farm) as disproportionately important to the integrity of the overall network. This provides another tool for directing the management of multiple-habitat sites and landscape restoration, although it is yet to be tested empirically. Field and landscape-scale manipulations are required to both test and improve robustness models as a way of increasing the resilience of ecosystems. More recently, Pilosof et al. (2017) demonstrated that the multilayer net- work from Pocock et al. (2012) provides much more realistic information on the stability and robustness of ecological communities than the examination of a single disconnected monolayer network (e.g. a bipartite host–parasitoid network). Parasitoid extinctions (representing a major aspect for the natural pest control) differ between scenarios purely based on the plant–parasitoid network and more comprehensive scenarios considering a multilayer net- work of both plant–parasitoid and plant–flower–visitor interactions. As flower visitors are involved in plant pollination, pollinator extinctions lead 13 NGS Into Ecological Network Analysis
- 38. to secondary plant extinctions and tertiary parasitoid loss. This demonstrates that the biomonitoring of ecosystems cannot be realized reliably without considering the myriad of interactions occurring between organisms, as everything is connected in an ecosystem (Evans et al., 2017). The robustness of interactions calculated from multilayer networks rep- resents a powerful indicator of the ecological condition of an ecosystem and should therefore be developed further in the context of biomonitoring pro- grams. Multilayer network approaches allow the long-term monitoring of the fragility of key components of ecological processes and ecosystem ser- vices such as plant–flower visitor networks (i.e. pollination) or insect pest–parasitoid networks (i.e. natural pest control) across spatial scales. With the development of ENA and the availability of NGS, we foresee a comple- mentary use of traditional biomonitoring indicators (i.e. species richness, population surveys) with new indicators based on the architecture of eco- logical networks (in particular, the robustness) which are ultimately much more intimately linked to ecological processes. 3. ECOLOGICAL NETWORKS CAN BE CHALLENGING TO BUILD USING CONVENTIONAL APPROACHES Despite their proven value in ecological research, networks are nev- ertheless limited by the difficulties of building them. These difficulties are centred around three major issues: (i) the sampling effort required to capture a significant range of species interactions; (ii) the reliable identification of specimens; and (iii) the adequate description of interactions between the organisms (see Box 1). First, detecting the majority of species and their interactions within a net- work requires monumental effort. The challenges increase with the species richness in the ecosystem, the spatial scale of the habitat/ecosystem of inter- est and the temporal scale over which interactions are being considered. For example, the biodiversity of tropical ecosystems is much more difficult to assess accurately than its equivalent in arctic environments or temperate agroecosystems (Lewinsohn and Roslin, 2008; Morris et al., 2004), even if the latter is not trivial to study either (Derocles et al., 2014a, 2015; Evans et al., 2013; Macfadyen et al., 2009; Pocock et al., 2012; Wirta et al., 2014). Moreover, quantifying any aspect of species diversity in order to monitor environmental changes in biodiverse regions runs into major issues of scale-dependency (e.g. Dumbrell et al., 2008), raising further logis- tical challenges associated with repeatedly monitoring species diversity, 14 St ephane A.P. Derocles et al.
- 39. while environmental changes modifying the spatial (and most likely tempo- ral) scaling properties of species within these systems. These problems all arise from the sampling effort and associated logistical constraints required to detect a representative and significant proportion of the species living in the ecosystem (Gotelli and Chao, 2013; Gotelli and Colwell, 2011; Jordano, 2016). Furthermore, all species are not sampled equitably (and some of them simply cannot be sampled at all; Valentini et al., 2009b). For example, temporally transient species (e.g. due to migration or phenol- ogy), which when present may have a disproportional influence on network interactions, are almost always ignored. Thus, these issues all lead to a biased view of biodiversity, which favours reporting the presence of species that are the most conspicuous and easiest to sample. As sampling effort and com- pleteness greatly impact the inferred structure of ecological networks, it is now usual to quantify network sampling completeness (see Costa et al., 2016) using estimators such as Chao 2 (Chao, 1984; Colwell and Coddington, 1994). This approach partially alleviates the sampling issues (assuming high sampling completeness is attained), but ecologists still need new tools for a more exhaustive detection of species and interactions. Second, accurate species identification remains a major challenge, with two separate but related issues. The first issue is that accurate and reliable species identification requires specific taxonomic expertise for the studied group (Derocles et al., 2012a; Evans et al., 2016). Consequently, if multiple taxonomic groups are studied, many taxonomists may be required to assess the biodiversity within a network (Valentini et al., 2009a). Hence, reliable morphological identification may not always be possible for all taxa. For example, the existence of cryptic species (i.e. hard-to-identify species using morphological criteria) may lead to an underestimation of the species rich- ness within ecosystems, resulting in biases at the network level (Derocles et al., 2016; Hebert et al., 2004; Kaartinen et al., 2010; Smith et al., 2006, 2007, 2008) and inaccurate model predictions (Novak et al., 2011). The second issue is that numerous taxa cannot be identified in situ (e.g. microbes) and require additional laboratory processing that is often limited due to financial constraints. In the case of microbial species, this is further hindered by the need to culture them in order to provide sufficient numbers for identification. This provides a major identification bias as most microbial taxa are not readily cultivable in the laboratory. In network ecology, mis- identifications can be very problematic as they may bias the structure and dis- tribution of interactions. As species may interact with numerous other organisms at different network levels, each identification error is compounded 15 NGS Into Ecological Network Analysis
- 40. with each interaction across the network. Consequently, sufficient sampling and accurate identification are crucial steps in the construction of highly resolved ecological networks. Third, exhaustively describing the possible range of species interactions that structure ecosystems is an onerous task (Bohan et al., 2013) and sampling all interactions of even a single type is conditioned by the number of obser- vations (Bl€ uthgen et al., 2008). While most species interactions are hard to identify in the field, some of them simply cannot be detected or observed with traditional sampling methods (see Jordano, 2016). As discussed by Gotelli and Colwell (2011), sampling biodiversity is very labour-intensive and often fails to detect most of the species in an ecosystem. For example, the construction of mutualist networks (e.g. plant–pollinator and plant– flower visitor interactions in Pocock et al., 2012) requires laborious and time-consuming field observations: it is therefore very hard to exhaustively capture mutualist interactions. Similarly in food webs built from host– parasitoid interactions (Derocles et al., 2012b; Gariepy et al., 2008), speci- men sampling and rearing in the laboratory are imperfect for most taxo- nomic groups (Derocles et al., 2012b, 2015; Evans et al., 2016). For instance, rearing hosts sampled in the field until the emergence of adult par- asitoids is a very challenging task. Indeed, both hosts and parasitoids have a high risk of dying during the rearing process, hence compromising the iden- tification of host–parasitoid interactions. In webs based on prey–predator interactions, a ‘Russian doll’ effect may lead to the detection of false inter- actions from morphological (or molecular) identification of gut contents (Woodward et al., 2012). In this latter case, the prey is not directly consumed by the predator from which the gut content was analysed, but in the gut con- tent of an intermediate consumer present in the focal predators gut. Conse- quently, overlooking these cases of secondary predation may lead to unrepresentative ecological networks. Finally, some interactions cannot realistically be observed in the field despite providing valuable information on ecosystem services, such as seed–ground beetle interactions associated with weed regulation by Cara- bids (Bohan et al., 2011b), or belowground plant–microbe interactions, such as arbuscular mycorrhizae that influence terrestrial ecosystem productivity (Fitter et al., 2005). When relying purely on classic approaches (i.e. field observations, specimen rearing, morphological identification) to build eco- logical networks, the construction of networks becomes risky. In order to limit the complexities and costs of describing complete eco- logical networks for a given ecosystem, most ecological network studies to 16 St ephane A.P. Derocles et al.
- 41. date have assessed ecosystem function and services by studying a subsample of the network and focusing on particular types of interactions (e.g. mutu- alist or trophic interactions). The choice of subsampling (according to the question addressed) makes sampling, field observation, specimen rearing and taxonomic identification logistically possible. Focussing on a subset of interactions also illustrates the a priori expectations ecologists have on the underlying role of some taxonomic groups on ecosystem functions and ser- vices. Many ecological networks studies to date may better reflect the inter- actions which are easy to study or that ecologists think more important, rather than an actual representation of ecosystem functioning. This can lead to key aspects of networks being overlooked: there are still vast numbers of as yet ‘unknown’ interactions that need to be described and their role in eco- system function evaluated. For example, in an agricultural network, a machine learning approach discovered an unexpected role for predatory spi- ders as prey (Bohan et al., 2011a; Tamaddoni-Nezhad et al., 2013), a finding confirmed by subsequent gut content analyses (Davey et al., 2013), giving a new mechanistic insight into the role of spiders in agroecosystems. The application of combined approaches, such as machine learning and NGS, that are less limited by the a priori expectations and assumptions of ecologists could greatly expand and speed up the discovery of links to build a more holistic and exhaustive view of ecological networks (Bohan et al., 2017). Pocock et al. (2012) were among the first to assess multiple types of inter- actions that were pooled in a ‘network of ecological networks’ in the context of farmland ecosystem services and functioning, providing new insights into the robustness of these interconnected networks (Evans et al., 2013). These networks of networks were built using conventional methodologies that rely on field observations or rearing specimens followed by morphological iden- tification by taxonomists. Although species interactions were highly resolved and well quantified for many of the subnetworks (e.g. plant–insect pollinators), others were potentially subject to bias (e.g. plant–leafminer– parasitoids) because of the limitations of taxonomically selective rearing suc- cess and the reliance on accurate morphological identification. Given that the construction of such networks is labour-intensive, building larger, highly resolved ecological networks in a wide range of ecosystems is likely to be hindered until more cost-effective methodologies can be developed. The application of NGS technology is one such method that is likely to revolu- tionize network ecology. Advances in DNA-sequencing technologies are answering previously intractable questions in functional and taxonomic biodiversity and provide 17 NGS Into Ecological Network Analysis
- 42. enormous potential to determine hitherto difficult to observe species inter- actions. Thus, DNA-based approaches, NGS in particular, hold the poten- tial to provide many of the solutions to the problems described earlier (Bohan et al., 2017; Evans et al., 2016; Vacher et al., 2016). Combining DNA-barcoding technologies with ENA offers important new opportuni- ties for understanding large-scale ecological and evolutionary processes (such as invasive species, see Kamenova et al., 2017), as well as providing powerful tools for building ecosystems that are resilient to environmental change (see Evans et al., 2016 for a conceptual framework). Until recently, ecological networks represented therefore a powerful but challenging approach to establish and were consequently difficult to integrate in biomonitoring. As discussed in the next section, NGS technologies together with the pre- diction of interactions with statistical modelling and machine learning rep- resent now an exciting opportunity to include ENA more systematically into biomonitoring. 4. COMBINING NGS WITH ENA: OPPORTUNITIES AND CHALLENGES 4.1 Using NGS to Construct Ecological Networks Currently, ecological networks constructed using DNA-based approaches are not used to regularly monitor ecosystems. This may be partially due to the historical reliance on classic field survey methods in network ecology, which rely on observation, specimen sampling, laboratory rearing and mor- phological identification to construct bipartite networks. Recent work has demonstrated that NGS can be rapid, universal and relatively cheap, in com- parison to conventional (i.e. ‘traditional’ taxonomy based) approaches to assess biodiversity (Beng et al., 2016; Ji et al., 2013; Liu et al., 2013). Beyond the characterization of biodiversity, NGS can also be used to efficiently build ecological networks (see Evans et al., 2016; Toju et al., 2014; Vacher et al., 2016). First, NGS has the potential to directly establish species interactions (Evans et al., 2016; Kitson et al., 2016; Toju et al., 2014). Second, with metabarcoding and eDNA approaches, NGS can also generate millions of DNA sequences which then can be processed and used in statistical models to construct ecological networks (Vacher et al., 2016). However, molecular approaches and NGS in particular are yet to be widely used to build ecological networks. Non-NGS molecular approaches such as diagnostic PCRs (using taxon-specific primers to detect targeted species in samples) and Sanger sequencing approaches (see Table 2) 18 St ephane A.P. Derocles et al.
- 43. Table 2 Comparison of the Main Sequencing Technologies Sequencing Platform Sequencing Generation Amplification Method Sequencing Method Read Length (bp) Error Rate (%) Error Type Number of Reads Per Run Time Per Run (Hours) Cost Per Million Bases (USD) Sanger ABI 3730xl 1 PCR Dideoxy chain termination 600–1000 0.001 Indel– Substitution 96 0.5–3 500 Ion Torrent 2 PCR Polymerase synthesis 200 1 Indel 8.2107 2–4 0.10 454 Roche GS FLX+ 2 PCR Pyrosequencing 700 1 Indel 1106 23 8.57 Illumina HiSeq 2500; high output 2 PCR Synthesis 2125 0.1 Substitution 8109 (paired) 7–60 0.03 Illumina HiSeq 2500; rapid run 2 PCR Synthesis 2250 0.1 Substitution 1.2109 (paired) 24–144 0.04 Illumina MiSeq v3 2 PCR Synthesis 2300 0.1 Substitution 3108 27 0.15 SOLiD 5500xl 2 PCR Ligation 260 5 Substitution 8108 144 0.11 PacBio RS II: P6-C4 3 Real-time single-molecule template Synthesis 10,000–15,000 13 Indel 3.5–7.5104 0.5–4 0.40–0.80 Oxford Nanopore MinION 3 None Nanopore 2000–5000 38 Indel– Substitution 1.1–4.7104 50 6.44–17.90 Based on Schendure, J., Ji, H., 2008. Next-generation DNA sequencing. Nat. Biotechnol. 26, 1135–1445; Glenn, T.C., 2011. Field guide to next-generation DNA sequencers. Mol. Ecol. Resour. 11, 759–769; Niedringhaus, T.P., Milanova, D., Kerby, M.B., Snyder, M.P., Barron, A.E., 2011. Landscape of next-generation sequencing technologies. Anal Chem. 83, 4327–4341; Liu, L., Li, Y., Li, S., Hu, N., He, Y., Pong, R., Lin, D., Lu, L., Law, M., 2012. Comparison of next-generation sequencing systems. J. Biomed. Biotechnol. 2012, 251364; Escobar-Zepeda, A., Vera-Ponce de León, A., Sanchez-Flores, A., 2015. The road to metagenomics: from microbiology to DNA sequencing technologies and bioinformatics. Front. Genet. 6, 348; Rhoads, A., Au, K.F., 2015. PacBio sequencing and its applications. Genomics Proteomics Bioinformatics 13, 278–289; Weirather, J.L., de Cesare, M., Wang, Y., Piazza, P., Sebastiano, V., Wang, X.-J., Buck, D., Au, K.F., 2017. Comprehensive comparison of Pacific Biosciences and Oxford Nanopore Technologies and their applications to transcriptome analysis. F1000Res. 6, 100.
- 44. remain more intuitive and easier for network ecologists to understand than NGS (Derocles et al., 2014a, 2015; Traugott et al., 2008; Wirta et al., 2014). These two explanations focus on why network ecologists have yet to fully embrace NGS approaches over more traditional methods. The flip side to this argument is that molecular ecologists using NGS for metabarcoding studies have yet to fully realize the potential of the data they generate. The vast majority of NGS studies quantifying the diversity of ecological communities have heavily relied on descriptive statistics based on classical measures of community diversity, and/or changes in species composition between samples. However, data are often collected in such a way that (ecological) networks could be con- structed, but are not, and the vast potential of the NGS data thus remains unrealized. Conventional approaches to ecological network construction have some major drawbacks that could make them inefficient in the biomonitoring of ecosystems. Visual species identification (with or without microscopy) can sometimes be slow and labour-intensive at best, or unreliable at worst. Diag- nostic PCRs need very good prior knowledge of the species composition of the ecosystem monitored, because they require the design of multiple PCR primers to detect the full range of species. For Sanger sequencing, costs increase linearly with experiment size and quickly become too expensive for large-scale biomonitoring. In contrast, NGS metabarcoding may scale more efficiently to large sam- ples compared with microscopy, diagnostic PCRs and Sanger sequencing, providing opportunities for much more intensive sampling of species inter- action networks than has previously been possible. The investment in time and materials goes ‘per plate’ (96, 384 or 1536 samples) rather than ‘per sam- ple’, although for large numbers of samples, additional sequencing runs may be required, which increases the cost. However, the number of samples that can be processed with a single sequencing run varies widely depending on a range of factors, including sequencing technology, chemistry techniques, and the quality of DNA extraction and amplification procedures (see Table 2). While in principle highly useful, it is these technical, and some of the theoretical issues linked to the use of NGS to quantify interactions that may have limited their adoption by researchers. We discuss ways of overcoming these limitations below and foresee that the construction of ecological networks using NGS will soon become commonplace and be integrated into biomonitoring schemes. 20 St ephane A.P. Derocles et al.
- 45. 4.2 PCR Bias and Abundance Estimation in NGS Community Analyses Understanding the limitations of molecular approaches in detecting species interactions is fundamentally important when correctly designing a sequenc- ing approach and interpreting the produced network. Primer and amplifi- cation biases are well-known phenomena in the PCR. Mismatches between primer and binding site sequences or structural and compositional variation in the DNA strand can lead to variation in PCR efficiency (Polz and Cavanaugh, 1998), causing two distinct issues, respectively: (1) the inability to detect species present within the sample as the primer mis- matches exclude their detection, and (2) the preferential amplification of DNA from some species at the expense of amplifying DNA from others (a cloning step is added to separate mixed DNA sample; e.g. Dunshea, 2009). However, for metabarcoding where the aim is to parallel sequence an entire community (or to identify two parties in an ecological interaction), this becomes more critical as differing PCR efficiencies among species can result in a final PCR product composition that is not representative of the input DNA composition (although some authors have reported broad cor- relations e.g. Elbrecht and Leese, 2015; Leray and Knowlton, 2017; Razgour et al., 2011). In practical terms, biases can lead to false negatives and read depths that are of no use for determining quantitative or even rel- ative community composition (Elbrecht and Leese, 2015; Leray and Knowlton, 2017; Piñol et al., 2014). The most common approach to dealing with this is to develop PCR primers that are as general and unbiased as pos- sible (Elbrecht and Leese, 2015; Leray et al., 2013), but even these are prone to the above-mentioned biases and a certain degree of PCR-induced bias is now commonly acknowledged in PCR-based metabarcoding studies (Leray and Knowlton, 2017). That said, in some instances by using carefully designed primers and targeting genes which vary in base-pair composition but not structural properties among species, elimination of PCR amplification biases is entirely possible (Cotton et al., 2014), and researchers should continue to pursue development and validation of these unbiased approaches. Some authors have attempted to circumvent this by using a meta- genomic approach (e.g. Tang et al., 2015) where they sequence all DNA present in their extraction and then filter the resulting sequences to only retain the data of use for identifying species. In theory, with no PCR step, there is no amplification bias, so read depths are more representative of the input DNA composition. In practice, the relationship between community 21 NGS Into Ecological Network Analysis
- 46. composition and metagenomic read depth is not so simple. The availability of mitochondrial DNA (mtDNA) for extraction varies significantly with tis- sue mass and metabolic activity (e.g. there is a significant nonlinear increase in mitochondrial count in developing oocytes; Cotterill et al., 2013), and this relationship can be further modified by tissue type and age (Barazzoni et al., 2000). This bias concerns mtDNA (mainly used to identify animals), but similar issues surround plastid DNA (used for plant identification), and the overall metagenome is often swamped with 16S ribosomal DNA gene reads (used for microbe identification), which may mask the presence of rarer higher taxa. Even if it were possible to know how read counts vary with tissue mass/type/age for all the organisms in our community (e.g. the con- tribution of multicellular organisms to eDNA has been modelled by this chapter; Sommeria-Klein et al., 2016), relative read counts can be further skewed by differences in ease of DNA extraction across taxa (Schiebelhut et al., 2017) and the extraction method used to obtain the DNA (Deiner et al., 2015; Vesty et al., 2017). Taken together we are forced to conclude that, as currently performed, metabarcoding is not generally suitable for esti- mating tissue biomass from sequence data (Clare, 2014) and thus any such metabarcoding-based estimations would have to be idiosyncratically cali- brated using conventional abundance surveys (as in Tang et al., 2015). Esti- mation of abundances is also problematic except for single-celled organisms for which they can be assessed accurately by targeting genes with minimum amplification biases (see Fischer et al., 2017). For higher taxa, however, only relative abundances can be recovered from NGS data. Relative abundances may nevertheless have a limited use in a conservative biology perspective and thus in biomonitoring as well (Clare, 2014). 4.3 NGS Without a Prior PCR Step A significant drawback of the metagenomic approach is cost (see Table 2). By sequencing all DNA in an extraction, researchers greatly limit the sample size per sequencing run and thereby either increase the sequencing costs for the study or, in fixed cost studies, reduce the statistical power of the study dramatically. They also discard much of the available read depth when they filter reads to only those of direct interest. One general approach to solve this is to enrich the DNA to be sequenced for a specific genomic region without PCR, avoiding PCR bias. This can be achieved in several ways, but the most common is to use an hybridization approach that employs sets of degenerate probes that can bind to target DNA and then themselves be bound to 22 St ephane A.P. Derocles et al.
- 47. magnetic beads (Gnirke et al., 2009) or a solid substrate (Albert et al., 2007) with nontarget DNA simply being washed away. The enriched DNA is biased towards useful genomic regions so a smaller proportion of the sequenc- ing reads are discarded and more samples can be included in a sequencing run (see ‘sequencing coverage’ in the glossary). Variations of this approach exist, which range from very simple centrifugation-based approaches (Macher et al., 2017) to much more complex methodologies using isothermal DNA replication (Dapprich et al., 2016) that can allow researchers to enrich for extremely long genomic regions suitable for the latest sequencing technologies such as Pacific Bioscience (PacBio) Single Molecule, Real- Time (SMRT) sequencing and Oxford Nanopore MinION (see Table 2). Depending on the laboratory, these approaches can be highly scalable, but their utility for community assessment is yet to be proven. 4.4 Detection of Species Interactions Using Molecular Tools As an alternative to attempting to infer relative abundances via read depths of bulk extracted communities, it is possible to simply analyse individual organ- isms separately and link the metadata for each sample (i.e. individual). In this situation, the number of samples for each species is a proxy measure of rel- ative abundance. If the sample contains multiple DNA templates arising from a species interaction (e.g. predator gut contents, host/parasite systems or plant–pollinator systems), then it is possible to use molecular tools to detect these species interactions in a quantitative or semiquantitative man- ner. Prior to the advent of parallel sequencing technologies, this would have been achieved by one of the following two broad approaches. First, PCR diagnostic approaches use sets of primer pairs that each produce species- specific bands of different lengths (or with different attached fluorophores as in microsatellite genotyping) that can then be separated by gel or capillary electrophoresis (e.g. aphid/parasitoid interactions Traugott et al., 2008; predatory beetle gut contents King et al., 2011). Second, PCR amplification of all DNA in a sample using general primers, separating PCR products via cloning (e.g. Dunshea, 2009) or gel electrophoresis, followed by Sanger sequencing (e.g. Kitson et al., 2013). Third, the design of primers specific to a taxonomic group (e.g. a parasitoid family or a genus of prey) amplifying a short but variable region (such as a fragment of COI) is another approach to resolve species interactions (Derocles et al., 2012b; Fayle et al., 2015). This method allows identification of an interaction that is occurring by relying first on a PCR diagnostic (e.g. a parasitoid within a host, a prey with a 23 NGS Into Ecological Network Analysis
- 48. gut content) and then to identify the nature of the interaction by sequencing the organisms detected (Derocles et al., 2012b; Rougerie et al., 2011). Thus, this approach was successfully applied to build ecological networks in a farm- land system (Derocles et al., 2014a) and an arctic system (Wirta et al., 2014). However, because of the linear cost of the Sanger sequencing and/or the time to process samples with these molecular tools, these approaches allow to examine a relative limited number of organisms (e.g. a low number of hosts). The advent of NGS technologies allows researchers to parallelize this process and work more effectively on a larger scale. The use of PCR primers tagged with known sequences to track samples is well established in NGS (Binladen et al., 2007), and this has been shown to be effective not only for community metabarcoding (Yu et al., 2012) but can also be used to build webs. Toju et al. (2013, 2014) were one of the first to use NGS (454 pyrosequencing) to resolve species interactions between trees and arbuscular mycorrhizal fungi and then build ecological networks from that data. One step further, Shokralla et al. (2015) and Cruaud et al. (2017) showed that a ‘nested tagging’ approach for amplicons involving two rounds of PCR per- mits (see nested PCR in the glossary) extensive multiplexing to increase throughput of barcoding programs, and Evans et al. (2016) have proposed this as an approach to building larger, replicated networks in ecological stud- ies. In the future, it is likely that these sorts of nested tagging approaches will be combined with PCR-free approaches to sequencing to allow quantified networks to be produced while reducing concerns over PCR bias and miss- ing interactions caused by false negatives. All the molecular approaches described earlier represent tools able to rap- idly characterize the biodiversity of ecosystems or describe species interac- tions. There is no doubt that this area of research is expending very rapidly as that new advances must be expected, pushing the limits of the description of biodiversity and the understanding of ecosystems. In the future, we believe that NGS will be fully integrated by ecologists to build networks and will be a usual approach of biomonitoring programs. 4.5 How to Deal With Interactions Not Directly Resolved by NGS: Are Species Association Networks Species Interaction Networks? The Case of Microorganisms For several decades, ecological networks have been constructed from the observations of both the species and their interactions (Ings et al., 2009; Poisot et al., 2016b). Databases of observation-based ecological networks, 24 St ephane A.P. Derocles et al.
- 49. Discovering Diverse Content Through Random Scribd Documents
- 50. the active little figures aloft, the bustle and business in her, cannot impair the pregnant suggestiveness of her leave-taking. You think of the people aboard who have said “Good-bye” to their friends, perhaps for ever. Poor Jack, sitting astride on the fore-topgallant yardarm, catches hold of the lift, whilst he turns his head in the direction of where he reckons Stepney or Poplar lies, and, as he thinks of his wife or sweetheart and the perplexities of the new allotment notes, he discharges a stream of tobacco juice into the air, and, with a melancholy countenance, wipes his mouth with the back of his hand and goes on with his job. There may be plenty of bustle and loud calls, but there is bound to be a share of sorrow too. It is not long since the skipper took his wife to his heart, and his head is full of her and the youngsters as he paces the quarter-deck, sometimes pausing to peep over the side at the cluster of boats round the gangway ladder, and sometimes singing out to the mate, who has his hands full forward. Indeed, it is impossible to look at an outward-bound ship without sympathy and a kind of respect that comes near to being reverence in some minds. What will be her fortune? you think. She holds herself bravely on the bosom of the calm river; the current wrinkles itself sharply against her solid bows, and breaks away along her side in a cadence like the tinkling of bells. Who can doubt that tears are being shed in her darksome interior? It is hard to leave the old home. The glimpse of the church spire through the open scuttle brings up memories which tighten the throat. When shall the next meeting be? and when time brings it about, will not absent faces and a change in the spirit of old associations make it sadder than this going is? Pray God that no harm befall the stout ship! As you sweep past her your hearty hope is that prosperous winds may attend her, and that in the new country fortune and happiness await those whose sad eyes dwell fixedly on the land that will be far astern of them before the sun has thrice sunk beyond the deep. It may be that thoughts of this kind are suggested more by sailing than by steam ships, because the existence of the propeller does to a large extent mitigate the bitterness of the contemplation of distance. But let no in-shore dweller flatter himself that the sailing vessel is very nearly extinct. She may have one leg in the grave, but the other seems to me still to possess an astonishing amount of animation. The hulls of the vessels in the docks on the Blackwall side of the river are not, for the most past, visible from the water; but, unhappily for steamers, there is not the least difficulty in telling, by the look of spars bristling out of a hidden dock, which are steamships
- 51. there and which are sailing vessels. Some of these days, perhaps, when the right kind of moral shall have been drawn from broken propeller shafts and twisted rudder-heads, the difficulty of distinguishing between the rig of a sailing ship and the rig of a steamer may prove very much more considerable than it now is; but, as this matter is at present ordered, the towering masts, the immensely square yards, should leave even a ploughman in no doubt as to the character of the vessels to which they belong. The number of sailing ships which crowd the docks on either side the river must prove a real surprise to people who believe that it is all steam nowadays. Let ancient mariners be consoled by this assurance: there is plenty of steam indeed, but there is a deal of canvas too, so that all Jack’s work does not lie in the bunkers yet, and there must still be a large demand for seamanship of the old sort. I am not sure that the wonder of the river does not owe quite as much to the sailing ships as the steamers. The tall spars, the magnificent spread of yards, the black lines of shrouds, the beautiful tracery of intersecting running gear, added to the shapely hulls which support these towering fabrics of hemp and steel and wood, make a most noble and impressive sight, and give, so to speak, a final touch to the teeming, opulent, commercial inspirations of the great river. Lower and lower yet down the grand old stream the spirit of enterprise is settling, and the day is not far distant when the projected dockyards at Tilbury will veritably transform the quaint old town of Gravesend into the sea-gate of London. It is almost startling to contemplate that time. One thinks of Gravesend now as a mere break in the departure from the Thames. Will the chain of docks end at Tilbury? At Gravesend, apparently, they are thinking otherwise! and reckoning—somewhat against their own hopes—that if the Tilbury Docks people play at leapfrog with the Albert Dock proprietors, the latter company will repay the compliment and land themselves some distance lower down yet. The limits of the Port of London, however, will, I believe, be reached by within a quarter of a mile by the promoters of the Tilbury Dock undertaking,[D] so that one cannot say in this case that there is room enough for all. Unquestionably the docks which are nearest the sea will be the docks best liked; and owners will profit at the expense of tug-masters and pilots.
- 52. Meanwhile Gravesend may be complimented on its prospects. But what do the watermen think? They are loud just now in their complaints of the steam ferries. They say that they are not allowed to board the ocean steamers, even to put Gravesend passengers ashore. Everybody must go to Tilbury first. How much of their vocation will be left when the new docks are opened? But assuredly if some old interests vanish, many new interests will start into life under the magic wand of the harlequin Progress. One may look for a complete transformation of the low, flat, treeless shore of Tilbury Ness and an ever-increasing clustering of industries along the banks of those reaches whose skirts now mainly consist of mud. Our fourpenny voyage will have to be extended if we are to compass all the wonders of our river below bridges. The New Zealander who is to muse over the ruins of St. Paul’s may come as soon as he likes, only it is quite certain that his meditations will not be excited by any spectacle of decay. Life and industry were never more active on the Thames than now—enterprise never more bold, speculation never more prophetic. The time is not remote when Gravesend, which I may say for centuries has been thought of as a port of call, will be connected with London by lines of edifices and piers and wharfs, as Blackwall is connected, and future passengers by the little Thames steamboats—which, it is to be earnestly hoped, in the good time coming will be considerably more river-worthy than they now appear to be —will be conveyed past a continuous panorama of commercial life and marine interests to limits which will make Gravesend and the opposite shore the actual sea-gate of the Port of London; in other words, the entrance to a scene of civilization comparable to nothing that we can imagine even by the building up of fancy from the wondrous facts at present submitted to any man bold enough to adventure upon a fourpenny voyage down the Thames.
- 53. POOR JACK. I climbed the steep hill that runs from the Belvedere railway-station, pausing now and again for breath and to glance at the summer beauty of the distant green land through which the river toiled, like a stream of quicksilver sluggishly rolling, and presently, passing through a gateway, found myself in a fine park-like stretch of grounds, shaded by a multitude of tall far-branching trees, in the midst of which, and upon the highest point of the billowy soil, stood a spacious and exceedingly handsome mansion. There were circular seats affixed to many of the trees, and upon them I noticed several bent and aged figures leaning their breasts upon stout walking-sticks, and holding themselves in very quiet postures. Here and there, walking to and fro near the house or upon the grass under the trees, were similar figures, all of them bowed by old age, though some of them paced the turf with a certain nimbleness of tread. They were dressed in pilot-cloth trousers and sleeved waistcoats, with brass buttons, and ancient as these men were, yet it was wonderful to observe, even where decrepitude was at its height, how the old sea-swing and lurching gait of the sailor lived in their hobbling and determined their calling, as though the word “seaman” had been branded upon every man’s forehead. I stood looking at them, and at the house and at the great trees, beyond which the distant prospect was shining under the high sun, for many minutes before advancing. The sense of repose conveyed to me by the shadows of the trees, the restful shapes of cattle upon the slopes beyond the mansion, the motionless postures of the old men seated, and the movements of the few figures who were walking, cannot be expressed in words. I listened. There was no note of human life in the air; no sound broke the fragrant summer stillness but the piping of birds in the trees, the humming of bees and flies, the silken rustling of leaves. The landscape was like a painted picture, save where here and there, upon the far-off shining silver of the river, a vessel slowly gliding broke the still scene with a fugitive interest. I walked to the house and entered the spacious hall, and as I did so, a single stroke on a bell to denote that it was half an hour after noon resounded through the building. A number of ancient men hung about this entrance, and I examined them curiously, for of all the transformations which old age works in the human countenance I
- 54. never beheld stranger examples than were submitted by many of these venerable seamen. Let me own to a feeling of positive awe in my inspection, for there was no face but that time had invested it with a kind of sanctity. “How old are you, my man?” I said to one of them. He turned his lustreless eyes upon me and bent his ear to my mouth. I repeated the question, and he answered that he was ninety-three. Years had so honeycombed his face that such likeness of humanity as there was in it appealed to the eye rather as a fantasy than as a real thing. A sailor is usually an old man at fifty, thanks to exposure, to hardship, and to the food he has to live on. Many of these men had used the sea for above half a century; some of them were drawing near to a hundred years of age; little wonder, therefore, that they should be mere dim and feeble vestiges of creation, and that vitality in conformations so decayed should excite the awe and reverence of those who explore the vague and crumbling features, and behold the immortal spirit struggling amid lineaments which have the formlessness of the face of a statue dug from the sand which entombs an ancient city. I turned my eyes from these old men to the hall in which I stood. Pretty columns of malachite supported the roof; woodwork and ceiling were lavishly decorated; marine hints helpful to the prejudices of the decayed mariners were not wanting in the shape of models of full-rigged ships—men-of-war and East Indiamen of the olden time; through the door I could see the green grass sloping away into a spacious lawn; and the warm air, full of sunshine, gushed in sweet with the smell of clover and wild flowers. In a few minutes I was joined by the house-governor, himself a skipper, and fresh from the command of a sailing-ship-a genial, hearty gentleman, and the fittest person in the world for the command of such a quarter-deck as this. “The old men will be going to dinner at one o’clock,” he said; would I like to see them at their meal? I answered “Yes;” so we stood in the door of a long, handsome room, fitted with tables and benches, and watched the aged seamen come in one by one, hobbling on their sticks, many of them talking to themselves. “Have you any shipmasters among these men?” I inquired. “Several,” answered the house-governor; and he instantly called out a name. An old man approached us slowly; he was bald, with a very finely-shaped head and
- 55. a long grey beard, and stood deferentially before us, his hands clasped, waiting to be addressed. “This man had command of vessels for many years,” said the house- governor. I looked at the poor old creature, and received one of the gentlest, saddest smiles I ever saw on a man’s face. I asked him how it was that he came to need the charity of this institution in his old age. “I was in the General Steam Navigation Company’s service, sir, for many years, and had charge of vessels running to Boulogne. But my memory began to fail me; I was attacked with dizziness, and had to give up. I had saved some money, and took a little hotel at Boulogne, on the Quay. I could not make it answer, and, being ruined and an old man, sir, I had to come here.” He broke down at this, his eyes filled with tears, and he turned his back upon me. I waited a little, and then, taking his arm, I asked him if he was happy in this house. Yes, he said, he was quite happy. “You may talk to me without fear,” I continued; “I am here to learn the truth and to speak it. Do they feed you well?” “Very well, sir.” “Have you no complaints to make?” “None, sir.” “You think this institution a good and honest charity?” “God knows what we should do without it,” he exclaimed, looking round at the old men who were taking their seats at the dinner-tables. Here the house-governor brought up some other aged men, whom he introduced as shipmasters. One of them was a North Shields captain, eighty years of age; he supported himself on two sticks, was a little, white-faced, ancient creature, with strange silver hair, and he spoke with a wistful expression of countenance. He had been seized with paralysis by “farling doon” the main hatch of his vessel. He told me in his rich, plaintive, North-country brogue, how the doctor had measured his leg and thigh with a tape—for some purpose I could not clearly understand—and how the accident had flung him upon the world, a beggar, and forced him to take a refuge in this institution. Was he happy? Ay, it was a man’s own fault if he wasn’t happy here. He was grateful to God for the care taken of him. At eighty a man was
- 56. “na’ langer a laddie,” and with a bright old laugh he hobbled hungrily towards one of the dinner-tables. In a few moments two bells were struck, signifying one o’clock, and all hands being seated, I followed the house-governor to the bottom of the room to have a look at the tables before the old men fell-to. The dinner consisted of salt fish, butter, potatoes, and plain suet pudding. “This is Tuesday’s fare,” said the house-governor. “On Sundays they get boiled beef, potatoes, and plum pudding; on Mondays, vegetable soup, boiled mutton, and vegetables at discretion; on Tuesdays, what you see; on Wednesdays, soup, boiled beef, and potatoes; on Thursdays, roast mutton, vegetables, and bread and cheese; on Fridays, salt pork, pea soup, and calavances; and on Saturdays, soup and boulli—not soap and bullion, as Jack says, one onion to a gallon of water—but a very good preserved soup, with potatoes or rice and bread-and-cheese. Taste this fish.” I did so, and found it excellent; so, likewise, was the suet pudding. The potatoes were new. The beer was the only doubtful feature of the repast; it was thin, insipid, and flat. I made haste to taste and approve, for I could see that the old fellows were very hungry. The governor left me, and went to the top of the room, where, in a loud and impressive voice, he said grace, bidding the ancient mariners be thankful for what they were about to receive; they all half rose, and in one feeble, rustling old pipe, sung out “Amen,” and then, like schoolboys, made snatches at the dishes, and in a minute were eating with avidity. It warmed my heart to see them. It made me feel that there must yet be plenty of goodness left in this world, when— through the benevolence of strangers and their large-hearted concern for poor Jack—ninety-three old, very old seamen, tottering on the verge of the grave, so poor and so destitute, so feeble and so friendless that but for the benevolence of those whom Providence had brought to their succour, they must have miserably starved and died, were clothed, and fed, and sheltered, and tenderly watched over. I know not that I have ever been so moved as I was in my passage through that dining-room. It was not only the pathos that lies in the helplessness of old age; I could not but think of the great compass of time these men’s experiences embraced, of the changes they had witnessed, of the sorrows and struggles which had made up the sum of their long lives, and how eighty and ninety years of privation, endurance, and such pleasures as sailors take, and such ambitions as sailors have, had ended in these bowed and toothless shapes, clutching at their plain repast
- 57. with child-like selfishness, indifferent as death itself to the great machine of life that was whirring with its thousand interests outside the silent sphere of their present existence, and dependent for the bread their trembling hands raised to their poor old mouths upon the bounty of those who love the noble profession of the sea, and who will not let the old and bruised and worn-out seaman want for such help as they can send him. Here and there were men too infirm to feed themselves; and I took notice how thoughtfully their aged messmates prepared their meal for them. Some of those thus occupied were more aged than the men they assisted. “Bless your honour, he’s but a child to me,” said one of them, in answer to my questions; “he’s but three and seventy, and I shall be eighty-nine come next September.” One pitiful sight deeply affected me. It was an old man stone deaf and stone blind. How is the helplessness in his face to be conveyed? “He’s losing his appetite fast,” said a seaman of about eighty who sat near him. “His senses is all locked up. Ye never hear him speak.” There were sadder sights even than this; but I dare not trust myself to write of them. I followed the house-governor out of the dining-rooms into a large apartment, well stored with books, magazines, etc., the gifts of friends of the charity. This I was told was the reading-room. It looked on to the green grounds, and was a most cheerful and delightful chamber. Further on was another room furnished with bagatelle boards and side tables for cribbage, etc. There was a particular cleanness and neatness everywhere visible, and I asked who did the work of the house. The house-governor answered, “The inmates. The more active among them are put to washing down and dusting at ten o’clock, and they finish at twelve. This is all the work required of them. Throughout the rest of the day they have nothing to do but to lounge about the grounds and amuse themselves as they please in the bagatelle or reading rooms, or in the smoking-room, which is a large apartment in the basement.” Mounting the wide stone staircase, and admiring as I went the singularly handsome and lavishly-embellished interior of the very fine building, I found myself on a floor devoted to the sleeping-rooms. These consist of rows of bulkheads partitioning off little cabins, each with a door and a number, and furnished with a comfortable bed, and some of them were movingly decorated by photographs of a mother, a sister, a child, with
- 58. humble memorials saved from the wreck of the past; such relics of the old home as a few china chimneypiece ornaments, a coloured picture, and the like, with here and there a sea-chest, though, as a rule, these little cabins, as they are called, were conspicuously empty of all suggestions of marine life. Now and again the opening of a door would disclose an old man seated on his bed, darning a sock or mending a shirt. It might have been that they were used to the visits of strangers; but I could not help observing in all these old seamen an utter indifference to our presence and inspection, a look of deep abstraction, as if their minds were leagues astern of them or far ahead, and existence were an obligation with which they had no sympathy, and of which they never took notice unless their attention was compelled to it. “Here,” said the governor, taking me into a room in which three or four old men were assembled—for dinner had been finished some time, and the seamen had quitted the tables—“is a veteran who has taught himself how to write. Show us your copy-book, my man,” said he, giving him his name. The old fellow produced his book with a great air of pride, and I was struck by the excellence of the writing. “Is this all your own doing?” I asked. “Ay, sir, every stroke. It’s been a bit of a job; for, you see, when a man’s nearing eighty ye can’t say that his brain’s like a young ’un’s.” “This would shame many a youngster, nevertheless,” said I. “I’d be prouder if I could read it, though,” he exclaimed, with the anxious and yet gentle expression that seemed a characteristic of the faces in this institution. “Ah, I see,” said I. “You can copy, but cannot read what you copy. Never mind! that will come too, presently.” “I’m afeard not,” said he, shaking his head. “Writin’s one thing, readin’s another. I have learned to write, but dunno as ever I shall be able to read it.” The governor, with an encouraging smile, told him to persevere, and then led the way to one of the sick wards, where I found a very aged man in bed, and two others seated at a table. “That poor old fellow,” said he, pointing to the bed, “begged to be allowed to attend the funeral of a man who died in the institution a short time since; he was so much affected that he was struck with paralysis, and
- 59. had to be carried back here. He was for years a shipmaster, had command of several fine ships, and is a man of excellent education. He has been in this institution some years.” And then, addressing him, “Well, and how do you feel yourself now?” “Mending, sir, mending,” answered the old man. “It’s death to me to be lying here. Why, for seventy-nine years I never had a day’s illness, never took a ha’porth of physic.” “You must have patience,” said the governor; “you’ll be up and doing presently.” “Ay, the power of forereaching is not taken out of me yet,” he answered, breaking into a laugh, the heartiness of which somehow pained me more to hear than had he burst into sobs. There were more “cabins” upstairs, and in one of them we found an old Irishman standing, lost in thought, looking out of the window. I addressed him, and he answered me in a rich brogue. I never remember meeting a more winning old face, nor being won by a voice more cordial and pleasant to hear. He told me he had been in the Kent, East Indiaman, when she was burnt. This was so long ago as 1825, and he was then a hearty, able-bodied man. It was like turning back the pages of the history of England to hear him talk of that famous and dreadful disaster. “There’s another man in the institution who was along with me in the Kent,” said he. I thought of the description given of the Kent by the master of the Caroline as I looked at this ancient man. “Her appearance was that of an immense cauldron or cage of buoyant basketwork, formed of the charred and blackened ribs, naked, and stripped of every plank, encircling an uninterrupted mass of flame.” Again and again had I read the story of that terrible fire at sea, thinking of it always as something deep-buried in history, and infinitely remote; and now here was a man who had been an actor in it, talking of it as if it had been but of yesterday, quavering out his “says I’s” and “says he’s,” and eager to let me know that if he liked he could tell me something about the behaviour of certain responsible persons on board that would not redound to their credit. It was pantaloon with harlequin’s wand in his hand; the faded old picture was touched, and became a live thing, the seas rolling, the ship burning, the terror and anguish of nearly sixty years since growing quick again under the magic of
- 60. this ancient man’s memory, and in the presence of a living witness of that long-decayed night of horror. Of such a charity as this of the Royal Alfred Aged Merchant Seamen’s Institution how can any man who honours the English sailor and values his calling hope to speak in such terms of praise as shall not seem hyperbolical? Not for one instant will I say that as a charity it is superior to others which deal with the sick, with the destitute, with the infirm, with little children. “There is misery enough in every corner of the world as well as within our convent,” Sterne’s monk is made to imply by his cordial wave of the hand. But I do claim for this institution the possession of a peculiar element of pathos such as no man who has not beheld the aged, the stricken, the helpless, the broken-down men congregated within its walls can form any idea of. As you survey them their past arises; you think of the black and stormy night, the frost and snow, the famine and the shipwreck—all the perils which sailors encounter in their quest or carriage of that which makes us great and prosperous as a nation; and then reflections on the dire ending which must have befallen these tempest-beaten, time-laden men but for the charity that provides them with a refuge break in upon you, and you feel that no words of praise can be too high for such an institution, and that no money dedicated by generous hearts to the alleviation of human suffering can be better directed than to the exchequer of this aged seamen’s home. Ninety-three old sailors are at present lodged in the institution. The house is big enough to accommodate two hundred, but the funds of the charity are already stretched to their last limits, and many an old and broken-down seaman whom this home would otherwise receive, and whose closing days would be rendered happy by all that tender ministration, by all that pious kindness can effect, must die in the cold and cheerless silence of the Union unless the charity that is prayerfully entreated for him is given.
- 61. ON THE GOODWINS. On a fine, calm day from the height of the cliffs betwixt Ramsgate and Broadstairs you may spy at low-water time a yellow vein, like a thin winding of pale gold, a hand’s breath this side of the horizon—the famous and fatal Goodwin Sands. I suppose there is no shoal in the whole world that a man whose sympathies are with sailors can view with more interest. Starting from the North Sand Head, which is almost abreast of Ramsgate, and looking east, the eye follows the south-westerly sweep of the Goodwins until the Downs are embraced with all their dim tracery of spars and rigging and faint sinuous lines of steamers’ smoke beyond, whilst the giant South Foreland acclivity stares down upon the lightship abreast of St. Margaret’s Bay, marking the extreme limits in the south and west of the deadliest stretch of sands upon the face of the globe. “Who can view the Goodwins without thinking of the treasures which lie buried in their heart, of the hundreds of ships which have gone to pieces upon them, of the thousands of human corpses which have floated out of their flashing surf to be stranded upon some distant beach, or to drift, maybe for days, upon the bosom of the tides, looking up with blind faces to heaven through the green transparent lid of their sea coffin? There is no spot that has ever been the theatre of wilder human suffering. Again and again as you sail past you see forking up out of them some black gibbet-like relic of a wreck a week, a fortnight, a month old. Something of the kind is always visible, as though even on the tenderest of summer days, when the blue water sleeps around, and the heavens are a violet hollow, with a rayless sun making gold of the sea in the west, the deadly suggestiveness of that long sweep of yellow sand should be as plain as when its presence is denoted amid the black tempestuous night by the ghastly gleam of boiling white waters. I remember once passing these Goodwins and seeing a number of little black figures running about them. A pleasure vessel from one of the adjacent ports was lying at anchor a short distance off, and her boat was against the slope of the shoal. It was a very calm day indeed, the sea just blurred here and there with small draughts of air that gave the water in
- 62. those places a look of ice, with a pallid streak of the French coast beyond the white mainsail of the pleasure-cutter, hove up by the refraction of the light above the sea-line. I brought a small pocket telescope to bear, and observed that those little black figures running about like the savages Robinson Crusoe saw were Cockney excursionists, engaged in playing cricket. They played as if they wanted to be able to talk of having played rather than as if they enjoyed the game. Talk of contrasts! A man may be rendered pensive by watching children sporting in a graveyard, by mingling in a festivity held upon a space of ground where once a famous battle was fought, and where the feet of the merrymakers are separated from the bones and skulls of warriors by a couple of spades’ length of earth. But to see those little black-coated creatures running about after a ball on top of such an ocean burial-place that the like of it for the horror of its annals and for the number of those it has sepulchred is not to be found in this habitable world, might well have made the gayest heart sad and thoughtful for a spell. As I leaned over the rail, looking at those happy pigmies—those lords of creation who, viewed half a mile further away, might have passed for a handful of black crabs crawling about—the scene in imagination changed, the darkness came rushing out of the east with a moan of approaching storm, the three lanterns winked like stars beyond the North Sand Head, and there was a sound of weltering waters and the seething and hissing of surf rising up through the gloom out from the whole length of the shoals. The wind rose fresh and eagerly, with a raw edge in it; the ebony of the swelling water was broken by the glimmer of the froth of breaking seas. I could hear the muffled thunder of the confused play to windward of the surf, with the shrieking of the blast overhead, whilst a deeper shadow yet gathered in the air. Then, with a blinking of my eyes, back would come the facts of the thing again, and yonder were the little figures merrily chasing the ball, the sea spreading like a sheet of silk to the yellow rim of the hard sand, and the blue sky bright overhead. Yet another touch of the magician Fancy’s wand, and it was all howling storm and flying blackness and the steam of hurling spume again, with a sudden glare of lightning between, flinging out the shapes of the piles of whirling clouds like monstrous brandished wings going to pieces in the hurricane, and throwing up the black fabric of a big ship on her beam ends, her masts gone, and a fury of white water veiling her.
- 63. There are lifeboat coxswains who need but close their eyes to see fearfuller things. Just where those little creatures are brandishing their tiny bats and flourishing their shrimp-like legs, the great ship struck, and four hundred men and women shrieked out to God for mercy in one breath. A man’s fancy must be feeble even on the softest of summer days not to hear the crash of her timbers, the thunder-shocks of the smiting seas, the rending noises of hemp and wire and spar torn by the tempest from their strong fastenings; not to see the ghastly picture she makes in the wild gleam of the signal flare whose tongues of fire are blown horizontal, like streaming flags, by the furious breath of the storm, illuminating with a dull horrible crimson light the throngs of human beings who cry and struggle upon her decks, or hang, like streaming suits of clothes, in what remains of her rigging. Is this an exaggerated picture? Alas! the pen never yet was wielded that could pourtray, in the barest form, any one of the countless horrible scenes which have taken place on that stretch of sands where one summer day I watched, leaning over the rail of a vessel, a number of light-hearted excursionists playing cricket. Among the things which never can be known may be placed the thoughts which possess a man in the moment of shipwreck. Of the hundreds of published narratives none satisfies the reader; and of those who relate their experiences, how infinitely remote from the truth do their statements strike them as being when they put what they have written side by side with what they remember having felt! The reason is, I take it, because in no other situation is death more awful than upon the sea. It is commonly slow—at least, it gives time for anguish to become full-blown—and the hope of rescue must be very strong indeed, and well founded, to qualify that agony of expectation, sinking into paralyzing despair, which confounds and in a manner stuns a person stranded far out upon the water in a black night, seeing nothing but the glare of lightning or the spectral flashing of froth flying past, hearing nothing but the grinding and trembling and dislocating noises of the hull upon the ground. It is supposed because sailors cannot or do not describe the horrors they pass through that they lack the capacity of expression. But you may put the most eloquent writer now living, call him by what name you please, on board a ship foundering amid a tempest or going to pieces in a storm on such a shoal as the Goodwins or the Sunk Sand, and when he has been long enough rescued and ashore to recover the use of his brains, you may defy
- 64. him to write such a narrative of the disaster as will come, to his own conscience and memory, one jot nearer to the truth than the newspaper paragraph of five lines in which the wreck was chronicled. A man can describe what he has suffered in a railway collision, in a house on fire, down in a mine where there has been an explosion, in a theatre where there has been a panic; but put him aboard a ship and let him clearly understand that he is going to be drowned, and when succoured he can tell you little more than that the waves ran mountains high, that some people were brave, and that some people shrieked, and that what he best remembers is catching hold of something, and hearing the water in his ears, and being dragged into a boat. Very true is the old saying, “If you want to learn how to pray, you must go to sea.” So distracting, so paralyzing, so utterly despairful are all the conditions of shipwreck in its worst forms, that I cannot but think, when a man is known to act bravely and coolly in that situation, unmindful of himself, thinking of others, encouraging and heartening them, the heroism he exhibits is of a kind not to be matched by any kind of courage a man may show in a position that lacks the overwhelming features which distinguish the foundering or the stranding of a ship. Some days ago I met a seaman who had made one of the crew of a brig that a few months since was stranded on the Goodwin Sands, and went to pieces there. The circumstances of the wreck were so recent that I was sure it could not but be a very sharp, clear memory in this sailor; and, wanting to hear what sort of thoughts come into a man’s head at such a time, and how he will act, what kind of impulses govern him, and the like, I carried this mariner to where a seat and a glass of beer were to be had, and conversed with him. “She was a wessel,” said he, “of 220 ton, and we was in ballast, bound from Can (Caen) to Seaham. All went well, nothen particular happening, I mean, till we comes abreast o’ the South Foreland. It might then be twelve o’clock in the middle o’ the night. The weather was as thick as mud, plenty of rain driving along, and the wind west, blowin’ a fresh breeze. We was under upper and lower main-tops’l, lower fore-tops’l, and foresail.” Here he took a drink. “And the weather as thick as mud, you say?”
- 65. “Ay, thick as mud in a wine-glass. The Sou’ San’head light was on our starboard beam, and ye may guess how clear it was when I tell you that that light took a deal of peering at to make out. As to the East Good’in, why, all that way was black as my boot: not the merest glimmer to betoken a lightwessel there. I was at the side, heavin’ the lead, getting nine fathom, and then seven, and then eight, and then seven again. Eight fair betwixt the Callipers and the Deal coast I’ll allow ye’ll get eleven and twelve fathom good till you come on to past the Downs—headin’ up, I mean—and then it shoals down to height and seven and five and a ’arf. So in a night as black as a dead wall, when there’s no moon, who’s to know, when the last light seen has drawed out of view, and there’s ne’er another to be sighted, where you are in that water? We was going along tidy fast, when a squall of rain drives right up over our starn in a wild smother, and I had just made seven fathom by the lead when the wessel took the ground, chucking me off the rail on to the deck. The skipper begins to bawl out like mad, ‘Let go the main-torps’l halliards! Haul up the foresail! Let go the ——’ Wash at that moment comes a lump of sea right over the port quarter, cantin’ our starn to the south’ard and smotherin’ the decks. You didn’t want to see—you could feel that the brig was hard and fast, though as the sea thumped her she’d kinder sway on her keel.” Here he took another drink. “Well?” said I. “Well,” he continued, “what was to do now, master? Everything being let go aloft, the canvas was slatting like thunder up there, and though I’m not goin’ to tell you it was blowing a gale of wind, yet it seemed to come twice as hard the moment we took the ground, and the seas to rise as if our falling helpless on a sudden had swelled ’em up with joy. We lay with our head about nor’-nor’-east, and over the starboard bow you could see the white water jumping. But that was all that was visible. The wind seemed to blow up the thickness all round us, there was not a light to be seen, and looking around anywhere away from the white water was like putting your head in a pitch-kettle. Cold! master, that was the worst part of it. I’ll allow that in all sitivations of this kind the cold’s the part that’s hardest to bear. Somehow clanger ain’t so frightful when it’s warm. Can’t explain it, I’m sure; matter o’ constitootion, perhaps: but I doubt if ye’d find much bravery among the Hesquimos and the Roosians up near the pole, and the likes o’ them. Can’t see how it’s possible; but it’s only my ’pinion.”
- 66. Another drink. “Well,” he continued, holding up the fresh glass of ale I had ordered for him to the light, with a look of pensiveness in the one bloodshot eye he kept open, “we tarns to and makes a flare—a sort o’ bonfire. But if we couldn’t see anything, who was to see us? However, we kept all on burning flares, whilst first the fore-top-gall’nmast came down with a run, causing us all to jump aft out of the road, and then the main-topmast carries away at the cap and falls with a roar over the side, and set us all running forrard. I for one made up my mind we was all to be drownded. I couldn’t see no help for it. The noise of them spars cracking and tumbling away in the blackness overhead, and the shindy set up by the slatting canvas, along with the creaking of the hull and the washing of the water that came as white as milk over the starboard rail, was enough, I reckon, to make any man suppose his time had come, and that his ghost was to be turned out of him. However, we took heart after a spell, by noticing that the seas burst with less weight as the tide left us, though every butt in her must have yawed open after she had been grinding awhile, for she was full of water and a few hours more of such dusting was bound to have made staves of her. Well, at about half-past four o’clock in the morning, we being by that time pretty near froze to death, the weather thinned down, and we caught sight of the Gull Light shining—about three mile off, I dare say. What was to be seen of our wessel was just a fearful muddle; masts overboard washing alongside, the lower masts working in her like loose teeth with every heave, decks full of raffle, and the water every now and again flying over us as though detarmined if it couldn’t wash us overboard it would keep us streamin’ wet. When we spied the Gull Light we turned to and made another flare, and presently they sent up a rocket, and to cut this yarn short,” continued he, having by this time emptied his second tumbler, and finding me slow in offering him a third, “just as the light was abreakin’ in the east one of us sings out that there was a steamer headin’ for us, and when the mornin’ grew stronger we spied a tug makin’ for us with a lifeboat in tow. Well, by this time there was little enough sea, and the lifeboat, letting go off the tug, came alongside, but two of our men was so badly froze up that they had to be lifted into her, and such had been our sufferings, though I’m not going to say they equalled what others have gone through on those cussed sands, that we couldn’t have looked worse, with salt in our eyes and our faces washed into the appearance of tallow, had we been spendin’ forty-eight hours on that shoal.
- 67. We lost all our clothes, every bloomin’ thing we had with us; and that same forenoon, just afore twelve o’clock, half a gale of wind sprung up, and by two o’clock there was nothing to be seen of the brig.” “And that’s the story,” said I. “That’s it,” he answered; “every word gospel true.” “How did the others behave,” said I, “in this awful situation? Pretty well?” “It was too dark to see,” he answered. “Did you encourage one another?” “Well,” he replied, “the cook at first kept on singin’ out, ‘We’re all drownded men! Lord have mercy upon me!’ and the like of that, until the cold took away his voice. I don’t know that there was any other sort o’ encouragement.” “And what were your feelings,” said I, “when the brig took the ground and the water washed over her?” “My feelings?” he replied. “Why, that we was in a bloomin’ mess. That was my feelings.” “How did the prospect of death affect you—I mean the idea of being swept into the black water and strangling there?” “Are you chaffin’ me, sir?” he asked. “Certainly not,” said I. “Well,” he said, “I’m blessed if I was asked such a question as that afore,” grinning. “It’s like a meetin’-house question.” “Didn’t you think at all?” said I. “Yes,” he answered; “I thought what a jolly fool I was to be ashore on the Good’ens on a winter’s night, gradually dyin’ of frost, instead of bein’ in a warm bed ashore, with a parlour to take breakfast in when I woke up. That’s about it, sir.”
- 68. THE STRANGERS’HOME. A plain red-brick building stands in the West India Dock Road, with the following lengthy name or description written along the front of it:—“The Strangers’ Home for Asiatics, Africans, and South Sea Islanders.” On the day I visited this house there were three or four people standing on the doorsteps, with faces which did more in an instant to express the character of the place than could have been effected by reams of reports of annual meetings and descriptive pamphlets. They were, it is needless to say, persons of colour, and of very decided colour too: one as black as a hat, another of a muddy yellow, a third a gloomy brown. They were dressed in European clothes: they might have belonged to nations which were in a high state of civilization when the Thames was clean water, and rolled its silver stream through a land whose scanty population hung loose and unclothed among the trees; but for all that, they had the look of wild men in breeches, and the very black person needed little more than a boomerang or a bow and arrows to give him the aspect at least of an unsafe object. I had, however, but little time to inspect these men, for a commotion in the hall of the building, coupled with an assemblage of some dozen or twenty people on the street pavement, called my attention to a spectacle of real interest. This consisted of the starting of a troupe of Javanese musicians for the place of entertainment where they were then performing. There were a number of men and four women—at least, I think there were four women; yet it is possible that I may have mistaken a man for one of the other sex, for some of the men and women were very much alike, especially the men. They streamed out in a great hurry, their bright black eyes sparkling in their brown faces, the men smoking short pipes of a decidedly West India Dock Road pattern, and the women bundling along in such queer raiment that it would be as hopeless to attempt to describe its colours and cut as to catalogue the stock of a rag-and-bottle merchant. A kind of large private omnibus stood at the door, into which these strange people got, some of them climbing upon the roof; and striking indeed was the appearance of the windows of the vehicle, framing, as they did, every one of them, a dark, contented face, whilst the roof of the omnibus was crowded with blacks and whites, like the keys of a pianoforte.
- 69. “Who are those people?” said I to a Chinaman, as the omnibus rolled away. “Hey?” answered John. “Those people,” I said, pointing towards the retreating vehicle, “they are not sailors, are they? There are women among them.” “No, no, not sailor, no, no,” cried the Chinaman with great earnestness, and wagging his head so violently that he nearly shook his hat off. “Music- man, not sailor; play tic-a-tic, tic-a-tic;” and here he screwed an imaginary fiddle into his throat and fell to sawing the air with his elbow. At this moment I was joined by the secretary—a gentleman, let me say at once, who, after spending many years of his life in India, is now gratuitously devoting his services to the poor Asiatic who finds himself homeless in this great wilderness of London, often penniless, and speaking a tongue with which he may journey from Mile End Gate to Hammersmith without finding an ear capable of comprehending a word he says. This gentleman told me who those queer-looking people were, how they were in charge of a Dutch entrepreneur, and how they were “putting-up” at the Strangers’ Home because there, and at no other place in London, they were likely to meet people who, even if they did not speak their language, would impart a sense of home. We now proceeded to inspect the building. As at the Well Street Sailors’ Home, so here, the common room, if I may so term it, is the central hall, a large place furnished with seats and tables and heated by an immense stove. Here of an evening, when it is cold or damp out of doors, the inmates of the home assemble, and the bright lamps shed their light upon as many diverse countenances and costumes as there are nationalities to the eastward of Russia and in the great oceans which wash the Capes of Africa and South America. Strange, indeed, is the admixture to a European eye: the Hindoo sitting cross-legged on a bench listening, with dusky eyes rolling in his black attenuated features, to the pigeon-English of a round-faced Chinaman; a Malay endeavouring by gestures to make himself understood by a Kanaka; a native of Ceylon smiling over the porcine gutturals of a couple of Zulus; with here an Arab reis pacing the floor in lonely dignity, or a red man of a paternity indistinguishable in his features, which seem compounded of the Nubian, the last of the Mohicans, a dash of Polynesia,
- 70. Welcome to Our Bookstore - The Ultimate Destination for Book Lovers Are you passionate about books and eager to explore new worlds of knowledge? At our website, we offer a vast collection of books that cater to every interest and age group. From classic literature to specialized publications, self-help books, and children’s stories, we have it all! Each book is a gateway to new adventures, helping you expand your knowledge and nourish your soul Experience Convenient and Enjoyable Book Shopping Our website is more than just an online bookstore—it’s a bridge connecting readers to the timeless values of culture and wisdom. With a sleek and user-friendly interface and a smart search system, you can find your favorite books quickly and easily. Enjoy special promotions, fast home delivery, and a seamless shopping experience that saves you time and enhances your love for reading. Let us accompany you on the journey of exploring knowledge and personal growth! ebookgate.com