LUCCI Annual Report 2014/2015

LUCCI Annual Report 2014/2015 Lund Centre for Studies of Carbon Cycle and Climate Interactions

Editors: Susanna Olsson and Anna Ekberg ISBN: 978-91-85793-52-5 (print) ISBN: 978-91-85793-53-2 (e-publ) Cover photo: Dicroidium by Vivi Vajda

Printed in 2015 by Tryckeriet I E-huset, Lund University

2

LUCCI Annual Report – Table of contents

Contents Contents .................................................................................................................................. 3 Preface ..................................................................................................................................... 5 WORK PACKAGE 0 ................................................................................................................. 7 WP 0 – Administration and coordination ...................................................................................... 9 LUCCI-supported infrastructure and technical assistance ............................................................ 10 Communication .......................................................................................................................... 12 LUCCI R3i: Research Integration, Innovation and Inspiration ....................................................... 13

WORK PACKAGE 1............................................................................................................... 15 WP 1: Progress report ................................................................................................................ 17 MEMEG, Department of Biology ............................................................................................................. 17 Division of Nuclear Physics, Physics Department .................................................................................... 19 Biogeosciences and Remote Sensing groups, Department of Physical Geography and Ecosystem Science ..................................................................................................................................................... 22 The response of fungal communities to nitrogen and phosphorus fertilization in a spruce forest . 24 Responses of soil microbial community to warming across a natural geothermal gradient in grasslands of Iceland .................................................................................................................. 27 δ 13 C – A useful tool of source characterization of the carbonaceous aerosol at rural measurement sites? ................................................................................................................... 30 BVOC emission variation caused by intra-genotypic variation, canopy height and change in time ........................................................................................................................................... 33 Salt sensitivity of soil microbial processes .................................................................................. 36 Using pollution-induced community tolerance (PICT) to identify direct salt effects on the soil microbial community ................................................................................................................. 38 Soil microbial decomposer responses to salt: A review ............................................................... 41 Aging soil organic matter: concentration dependence and microbial control of the priming effect ......................................................................................................................................... 43 Does the fungal-to-bacterial dominance characterize the turnover of soil organic matter and nutrients? .................................................................................................................................. 46 Primary producer-release of labile carbon enhances litter degradation in aquatic experimental systems ...................................................................................................................................... 49 A comparable approach to assess variation in surface energy fluxes in northern high-latitude ecosystems ................................................................................................................................ 51 Succession of ectomycorrhizal fungi in mesh bags over a three year period in a fertilized and unfertilized Norway spruce forest .............................................................................................. 54 Seasonal variation of BVOC emissions from Norway spruce ......................................................... 56 Hygroscopicity of photochemically processed particles from three different sources .................. 59

3

LUCCI Annual Report – Table of contents

WORK PACKAGE 2............................................................................................................... 63 WP 2: Progress Report ............................................................................................................... 65 A Modern Automatic Chamber Technique As A Powerful Tool For CH 4 And CO 2 Flux Monitoring. . 67 High Arctic tundra greenhouse gas emissions: new insights from Adventdalen, Svalbard ..................................................................................................................................... 68 Controls of spatial and temporal variability in CH 4 flux in a high arctic fen over three years ........ 69

WORK PACKAGE 3............................................................................................................... 71 WP 3: Progress Report ............................................................................................................... 73 Persistent solar influence on Greenland climate during the last Glacial Maximum ....................... 80 A comparison of size fractions in faunal assemblages of deep-water benthic foraminifera – a case study from the Benguela Upwelling System ......................................................................... 83 Reconstructing the Holocene landscape development of SE Blekinge using pollen and the REVEALS model .......................................................................................................................... 86 Atmospheric circulation patterns in the Southern Ocean during the Holocene - a terrestrial view from Iles Kerguelen (49°S, Indian Ocean) on Southern Hemisphere Westerly belt variability .................................................................................................................................. 90

WORK PACKAGE 4............................................................................................................... 93 WP 4: Progress Report ............................................................................................................... 95 First evidence of the Cretaceous decapod crustacean Protocallianassa from Sweden and its implications for paleoenvironment and climate ........................................................................ 101 The Claret Conglomerate: Evidence of an abrupt change in the hydrological cycle at the Paleocene- Eocene Boundary.................................................................................................... 104 A multitype asteroid shower in the Late Eocene: Does this indicate an astronomical trigger of the late Cenozoic ice house world?........................................................................................... 106 The carbon cycle and its relation to astronomical signals during the Triassic-Jurassic mass extinction event (201 Mya) – studies from the Junggar Basin, China ......................................... 108

WORK PACKAGE 5............................................................................................................. 111 WP 5: Progress Report ............................................................................................................. 113 Main dynamics and drivers of boreal forests fire regimes during the Holocene ......................... 118 Planned LUCCI contributions to CMIP6 with EC-Earth ................................................................ 122

4

LUCCI Annual Report – Preface

Preface

Never before has global climate change been more in focus than today, and work on reaching global agreements and mitigating the impact of climate change has intensified. The requirements put pressure on all countries to reassess developments with the impact of climate change in mind. The implications of climate change are often highlighted by both scientists and the broader community, and include the lack of freshwater in some regions as a consequence of aridification. Another issue is the extinction of species that occurs as a combined result of deforestation and climate change. During the last six years within the Lund University Carbon Cycle Centre, LUCCI, we have collaborated in integrating a broad range of research fields with the aims to highlight knowledge about the carbon cycle, understand its feedback mechanisms, and clarify its future implications. We have employed a holistic approach – integrating information from modern systems and ‘deep-time’ data from the geological record. A strong research platform has been achieved by applying not only cutting edge research within individual disciplines, but also integrating strong research fields into new and creative constellations and involving scientists at different career levels. Through this strategy, our centre has received some excellent evaluations from the Swedish Research Council, something we should all be proud of. The flagships of the LUCCI environment are our strong research groups that are able to bring in substantial external funding and our R3i group of enthusiastic young scientists who are ready to carry the baton of research on climate change to new heights. During the coming years we will focus even more on the collaboration between the different work packages integrating our themes, such as atmosphere, biosphere, oceans—cutting across geographical areas and bridging time spans from ecosystems hundreds of millions of years in the past to seasons within modern environments.

Vivi Vajda LUCCI coordinator

5

6

WORK PACKAGE 0

Administration and coordination

8

LUCCI Annual Report – administration and coordination

WP 0 – Administration and coordination Vivi Vajda, Anna Ekberg & Susanna Olsson

LUCCI administration, coordination and communication are hosted under the umbrella of WP 0 and it is a pleasant task to be involved in the management of top research about such urgent and pressing issues as climate and climate change. Another pleasant occupation is to read the section about LUCCI in the “Midterm evaluation report of the 2008 Linnaeus Centres” (Vetenskapsrådet 2014, ISBN 978-91-7307-243-4). Our research environment came out strong in the evaluation of the first six years and now receives continued funding for the remaining 4 years until 201806-30. You will find the evaluation of LUCCI below. If you are interested in reading the full report, it can be accessed from https://publikationer.vr.se/. After the positive midterm evaluation and the following decision of continued funding, the steering committee decided to keep the earlier between-WP allocation of resources. This ensures that LUCCI can continue to provide long-term research support, building on already existing activities and initiatives in various forms, such as positions, technical support, infrastructure and project seed money. However, funding to Sweden’s Linnaeus Centres cannot be anticipated to continue after 2018, at least not in its current form. The steering committee has therefore initiated a discussion of alternative strategies towards continued, post-2018 strong research of carbon cycle climate interactions and feedbacks at Lund University. For WP 0, it means that the work now will focus both on relevant research facilitation (for example internal communication, outreach, annual meetings, seminars), and the task to make sure that the environment delivers as promised in the original application. More information about communication and outreach can be found in one of the following chapters. The R3i-group (Research Integration, Innovation and Inspiration), consisting of PhD students and early career researchers, is technically a part of WP 0, but they decide on their own agenda and you can read more about their activities and future plans in their own chapter of this annual report. The group leadership changed in early 2015 and we are happy that the new leaders, Ylva Persson and Niklas Olén, continue the same tradition of dedication and passion for the task that the former leader, Claire McKay, left behind. LUCCI research and links to projects associated with Future Earth (a global platform for scientific collaboration and research on global environmental change, www.futureearth.org) was presented at a seminar aiming to inform the Swedish Secretariat for Earth System Sciences (SSEESS, www.sseess.org) about relevant research at Lund University. The panel was impressed by the diversity of topics within LUCCI, representing areas of relevance for at least seven of the research projects currently associated with Future Earth. The event was jointly organized by the strategic research area BECC (www.becc.lu.se) and the Lund University Sustainability Forum (www.sustainability.lu.se). Finally, a few words about the annual meeting that will be held in early September 2015. The two-day meeting will have a paleoclimatic focus on the first day with invited keynote speakers Patricia Vickers-Rich (Monash University, Australia) and Alain Franc (University Bordeaux, France). We will also listen to a presentation by Per Persson (CEC, Lund University) about synchrotron light research opportunities at the MAX IV lab. Per will also share some of his insights about the future funding strategies of the Swedish Research Council (VR). Per Bodin (INES, Lund University) will then round off the day by presenting possible opportunities for young scientists at the International Institute for Applied Systems Analysis (IIASA). The second day of the meeting will be a tribute to 2015 being the UN international year of soils. Two members from LUCCI’s scientific advisory board will present their research, followed by presentations from the individual work packages.

9

LUCCI Annual Report – administration and coordination: crosscutting projects

LUCCI-supported infrastructure and technical assistance Anna Ekberg (a) (a) Centre for Environmental and Climate research, Lund University

Introduction Research on climate and carbon cycling within the LUCCI interest sphere is often strongly dependent on laborintensive field and lab work based on well-functioning infrastructure, and also on professional operation and maintenance of high-tech equipment. The benefits of such investments in infrastructure and technical support have been embraced within LUCCI, resulting in high added value to many projects both within the research environment and in cooperation with other partners. Several basic and doctoral education programs are also reaping the benefits of the lab and field facilities since students and teachers are given access to relevant expertise and analytical equipment positioned at the research front. The following is a brief summary of activities and infrastructure supported by LUCCI.

Chemical analysis, incubation chambers and radioactive labelling The chemical analysis laboratory is run by Marcin Jackowicz-Korczynski. The main chromatography instrumentation consists of HPLC-MS, GC-MS and GC-FID for analysis of organic chemical composition of liquid samples and trace gases in air. The instrument park also includes a Gasmet portable FTIR analyzer for trace gases in ambient air samples. The laboratory further facilitates temperature-, humidity- and light controlled growth chambers. The chambers are mainly used for plant incubation experiments, but can also be used for instrument calibration. One of the two available chambers is situated in a separate lab used for incubation, sampling and analysis of 14C- and 3H labelled material. Collected samples are analyzed by sample oxidation followed by scintillation counting. The number of researchers using the labs vary, e.g. depending on activities carried out by international visiting scientists, but normally five to ten researchers are actively operating the facilities, including several PhD students. Approximately three to four master student projects per year are also making use of the analytical instrumentation and expertise provided by the lab.

Sediment lab and research environment The Astrogeobiology (AGB) Laboratory situated at Medicon Village is a cross-disciplinary lab and research environment, founded with the aim to study and understand the interrelationship between astronomy, geology and biology. The laboratory has a high capacity for acid leaching of sedimentary rock for various applications such as stable isotopic analyses of organic fractions or searches for extraterrestrial mineral grains. Approximately ten researchers are actively involved in the AGB laboratory, and a cross-disciplinary PhD course, gathering students from geology, astronomy and physics, will be linked to the lab activities in autumn 2015.

Sample preparation and general lab support Reliable technical support and sample preparation is a crucial platform for any type of successful lab work. Sieving, microscopy and polishing of sediment and mineral samples is performed by Faisal Iqbal. Mattias Olsson works with carbon extraction from both aerosol- and sediment samples later to be analyzed for, or dated by, 14C. Git Ahlmark provides technical support in the labs.

Field operations Field measurements are fundamental to much of the research carried out within LUCCI. Several of the major sites are located in remote areas, which places high demands on well-functioning logistics and reliable in-place infrastructure. LUCCI supports major field operations in Greenland (Nuuk and Zackenberg), Svalbard (Adventdalen), and Sweden (Fäjemyr, Nimtek and Rumperöd).

10

LUCCI Annual Report – administration and coordination: crosscutting projects Zackenberg, Greenland The field measurements in Zackenberg, NE Greenland, are linked to the Zackenberg GeoBasis monitoring programme which has a focus on selected abiotic characteristics in order to describe the state of Greenlandic terrestrial environments and their potential feedback effects in a changing climate. LUCCI supports the GeoBasis measurements of CH4- and CO2 fluxes, energy exchange and meteorological measurements (autochamber system and flux tower). Mikhail Mastepanov is responsible for maintenance and development of the equipment and analytical instrumentation. Data and results from the site is used in basic and doctoral education, as well as for master thesis projects.

Nuuk, Greenland The field site situated outside Nuuk in the southwest of Greenland is hosting an autochamber system for measurements of CO2- and CH4 fluxes. There are also two energy exchange stations linked to the International Network for Terrestrial Research and Monitoring in the Arctic (INTERACT). Magnus Lund and Mikhail Mastepanov are responsible for development and maintenance of measurements and equipment. Data and results from the site is used in basic and doctoral education, as well as for master thesis projects.

Adventdalen, Svalbard This high-arctic field site is hosting an autochamber system for measurements of CO2- and CH4 fluxes, and one energy exchange station. Frans-Jan Parmentier is responsible researcher for the site and Norbert Pirk is maintaining the instrumentation as part of his PhD project.

Fäjemyr, Sweden The station has been in place since 2005 and LUCCI is supporting its operation, maintenance and consumables. Measurements include eddy covariance fluxes, meteorological measurements, gradient measurements of CO2, CH4 and H2O fluxes, and an autochamber system for CH4 and CO2 exchange. In 2013, an infrastructure grant from the Faculty of Science was used for sensor upgrade. Magnus Lund is the main responsible researcher for the site. On top of the many researchers that utilize the station, Fäjemyr has become an important facility also for educational purposes, much because of its location in such close vicinity to Lund. At least one basic education course and one PhD course per year visit the mire and several master thesis projects have been based on work carried out at the site. It has also come to function as a valuable training station both for students and for field assistants who are later to work at the more remote field sites in e.g. Greenland.

Nimtek and Romperöd, Sweden LUCCI is supporting personnel resources as well as instrument maintenance and calibration of long-term CO2 net exchange (NEE) measurements in two boreal forest ecosystems: a natural unmanaged forest in the north of Sweden, Nimtek and a continuous cover forest in southern Sweden, Rumperöd. The rationale for this is that longer measurements period are required in order to better understand the interannual variability in NEE. The Nimtek site has been running for about 3 full years and the Rumperöd site for a little more than one year. Both of these sites are quite unique for the Fennoscandian boreal forests; the Nimtek site because it is unmanaged, to be compared to managed forests, and the Rumperöd site because it represents a form of non clear-cut management that is much under discussion and debate.

11

LUCCI Annual Report –administration and coordination: communication

Communication Susanna Olsson

The communication office works with internal and external communication, with one employee (Susanna Olsson) at 40 % of full time. The work includes administration of events, internal information and meetings, support to R3i, outreach material and press releases, updating and handling of the website, newsletters, editing the annual report and supporting the Coordinator and Scientific secretary.

Outreach LUCCI hosted an event in November 2014 at the Lund library, which included a photo exhibition and as the main event an evening with presentations from researchers with a following public debate. It was appreciated by researchers, audience and library alike. The photos were up for one week before the talks and debate evening, with the idea that the exhibition would be a teaser and introduction to the science presentations. We are aiming at doing an event like this again, possibly at more libraries in the region close to Lund if there is interest from libraries and researchers. The planning work with this has started. During the spring we started up a series of LUCCI lunch seminars, where each work package hosted one seminar each. The aim was to encourage cross-disciplinary cooperation or at least exchange of ideas mainly from the discussion part of the seminars. The seminars were visited by about 20 people every time. This year the talks were held by the WP leaders, and the planning for a continuation next year is aiming at bringing something new to the lunch seminar concept and encourage phD students to be involved in presentations. The newsletters have been sent out about 4 times a year with an update of what is going on involving LUCCI participants. The newsletters are also available at the website. The LUCCI website is being continuously updated, most frequently the news, and the participant lists with new young researcher joining LUCCI. We are currently starting the work to migrate the content of the website to the LU Drupal system. This will open for easier management of the website and the possibilities for those who are interested to manage their own WP pages. Further, as we hope that LUCCI will continue after 2018, at least as a research area within LU, it is good to already now establish a good website to build on for the future, and to make it easy to handle for whoever will work with it.

The way forward The communication work in the years to come will be more aimed at outreach, both written products and events, and most importantly the plans are to arrange an international conference as a final for LUCCI in its present form. The aim is to show what LUCCI contributes with in terms of knowledge that can be of use for stakeholders, schools and the general public, as well as to fullfill the goals that was set up in the start of LUCCI.

12

LUCCI Annual Report –administration and coordination: R3i

LUCCI R3i: Research Integration, Innovation and Inspiration Ylva Persson (a), Niklas Olén (a) (a) Dept of Physical Geography and Ecosystem Sciences, Lund University

The LUCCI R3i group was initiated during the autumn of 2010 with the intention of being an interdisciplinary forum where all LUCCI PhD students and postdocs can discuss their research and interact with one another. In the beginning of 2015 there were 66 members of the LUCCI R3i group. In January 2015, Claire McKay stepped down as a LUCCI R3i leader and was replaced with Niklas Olén and Ylva Persson on a shared leadership role. Both Niklas and Ylva are from the Physical Geography and Ecosystem Science department, where Ylva is embarking on her third year as a PhD student and Niklas on his fourth. As the new R3i leaders, we are both determined to listen and develop the ideas and needs presented in the R3i group and we take on this responsibility with much dedication and anticipation. R3i meetings and after-work socials have continually been arranged as the importance to meet peers over the department levels to discuss ideas, future collaborations, workshops and excursions are vital aspects of what the R3i group represents. But further work is needed in order to increase communication and interaction, not only within the group, but over the group boarders and scientific fields as well. Our aim is to increase communication by welcoming the interaction from other work packages in some of our R3i activities. We also want to bring up more opportunities where group members can share their work and work approach with their peers in order to create more networking opportunities. During the year the R3i group has arranged two main events. In autumn 2014, 20 PhD students and postdocs went to House of Challenges in Malmö for a fun day of team building organized by LUCCI R3i. The participants were divided into mixed groups with people from different fields. Each group went around inside the house and worked together to solve different tasks and by that improved their collaboration skills as well as getting to know each other better. All in all it was a great day with many laughs and it improved the communication between the group members that were present. In the end of January 2015 the second event took place in form of a workshop called “Scientific writing and publishing with impact”. The workshop was led by Dr Dan Csontos, a publishing consultant at Elevate Scientific, with the aim to give the participants a better understanding and the tools to plan, write and publish scientific papers more successfully. The workshop was of high relevance for LUCCI as the demand of learning how to publish scientific material successfully always is high and of importance not only for each individual researcher but for the continuing outreach of LUCCI as well. Some of the points that were addressed were the editorial and publication processes, what editors look for when deciding on publishing, the importance of a strong main message and various techniques for clear, concise and coherent writing. The workshop was highly appreciated by the 19 participants, which were a mixture of both PhDs and postdocs. As leaders of the R3i group we are always looking for new inspiration and/or suggestions of activities that can be arranged within the group. If you have any suggestions please do not hesitate to contact any of us and we will see what can be done.

13

14

WORK PACKAGE 1

Understanding today´s carbon cycle and its interaction with the climate system and ecosystem

15

LUCCI Annual Report – work package 1

16

LUCCI Annual Report – work package 1

WP 1: Progress report Anders Lindroth (a), Kristina Eriksson Stenström (b), Erik Swietlicki (b), Håkan Wallander (c) (a) Dept of Physical Geography and Ecosystem Sciences, Lund University (b) Div of Nuclear Physics, Lund University (c) Dept of Biology, Lund University

MEMEG, Department of Biology Scientists involved: Prof Erland Bååth, Prof Håkan Wallander, Dr Johannes Rousk, Dr Adam Bahr, Dr Stephanie Reischke, Dr Zhanfeng Liu and PhD students Kristin Rath, Margarida Soares and Juan Pablo Almeida.

This year a new study was established on the volcanic Island Surtsey, south of Iceland, to study the importance of mycorrhizal fungi for carbon sequestration. Soil under plants associated with arbuscular (AM), ericoid (ErM) or ectomycorrhizal fungi (EM), as well as non-mycorrhizal (NM) plants, was sampled, and carbon and nitrogen contents, as well as the composition of microbial communities has been analysed. Preliminary results suggest that ErM and EM plants sequester more carbon than NM or AM plants. Continuing studies on another experimental site in Iceland, Forhot, has been going on over the year. In these sites soil temperatures in a grass land and a Sitka spruce plantation rose between 0 and 50°C above ambient after an earthquake in 2008. EM growth is followed in the forest, and AM growth in the grassland. Analysis of neutral and phospholipid fatty acids have revealed that microbial communities change over these gradients but the biomass of AM fungi in the grassland remained constant up to a level of around 10°C above ambient. At higher temperatures AM biomass dropped, probably due to poor plant growth. Growth and EM induced carbon sequestration in the forest site did not change along the gradient up to a similar level of 10°C above ambient. But the trees started to die at higher temperatures, and concurrently the growth of EM fungi dropped. The importance of EM fungi for nitrogen leaching is followed in an experimental site at Tönnersjöheden Research Park in Halland where N, P and N+P have been added in replicated plots. In these studies we found that N additions, and especially N+P additions, reduced the growth of EM fungi. Furthermore the capacity to retain N dropped significantly in N treated compared to control plots (Bahr et al. 2015). We also found that addition of P containing apatite stimulated EM growth in plots that had limiting P levels (Control and N treated plots). Furthermore, the composition of the EM community changed in response to apatite amendment, especially in plots that were limited by P availability. One EM species, Boletus badius, increased considerably in abundance after apatite amendment in plots that were P limited, and the importance of this species for apatite weathering with be followed-up in more controlled laboratory experiments. In another study the effect of disturbances on the composition of ectomycorrhizal communities was studied in mesh bags buried in the soil. It was found that inter-annual variation was the most important factor to shape these communities, but trenching and intensive fertilization with a complete fertilizer added every second year also had significant effects. The activities of decomposers in the soil, such as fungi and bacteria, have been studied in a number of different soil types and under different environmental conditions. For instance, food-web model based work has established a widely held belief that the bacterial decomposer pathway in soil supports high turnover rates of easily available substrates, while the slower fungal-dominated decomposition pathway supports the decomposition of more complex organic material. In a field experiment, the Detritus Input and Removal Treatments – DIRT – experiment at Harvard Forest Long-term Ecological Research site in the USA was used to test this hypothesis. In contrast to the general belief, no support was found for a positive relationship between the relative dominance of bacteria and high soil organic carbon (SOC)-quality. Furthermore, decomposition of SOC in more fungal dominated soils was not different from that

17

LUCCI Annual Report – work package 1 degraded in relatively more bacterial dominated systems. These results have important implications for improving end developing soil carbon models (Rousk and Frey 2015). The amount of carbon stored in soil is a regulator for soil fertility and the global climate, and is the balance between formation and decomposition of soil organic matter. Addition of high quality carbon is known to stimulate the decomposition of soil organic matter in a process called “priming”, but it is not well known to what extent this effect is due mainly to bacteria or fungi. Rousk et al. (2015) found that the decomposition of soil carbon increased by up to 350% after adding sugar, but no connection was found between growth responses of bacteria or fungi and the priming of soil carbon decomposition. Priming was also studied in pond water in laboratory experiments where the high quality carbon was delivered in a more natural by primary producers such as algae rather than adding sugar to the soil. It was concluded that enhanced primary production induced by more light stimulated fungal growth which had capacity to ‘prime’ the decomposition of recalcitrant organic matter. One of the most pressing environmental challenges to resolve for the next century is the salinization of soils. Rath et al. (2015) conducted a synoptic review of the available research on how salt affect decomposer microbial communities and carbon (C) cycling in soil. In addition to summarizing known physiological responses of microorganisms to salinity, they also provided an overview of a selection of widely applied methods to assess microorganisms in soil. The overall conclusion of the review was that salt exposure has a powerful influence on soil microorganisms and that this influence also could be used as an experimental probe to better understand how microorganisms control the biogeochemistry in soils. It is difficult to distinguish direct salt effects from indirect effects due to other confounding factors, such as pH or organic matter content when studying microbial communities in soil. In order to solve this problem the ecotoxicological concept of pollution-induced community tolerance (PICT) was employed to study salt effects in gradients of naturally saline soils. The basic idea behind the PICT method is that if a community is exposed to a toxicant at an ecologically relevant concentration, the average tolerance of the community to the toxicant should increase. Along these lines, enhanced tolerance was found for bacterial communities collected from saline soils, but fungal communities did not show a clear trend. The work of the group has during the year 2014 and 2015 been presented and discussed at EGU in Vienna in April 2014 and 2015, at the Global Soil Biodiversity Conference in Dijon in December 2014, at the Cost action ClimMani in Copenhagen in March 2105, and at the Cost action FP1305 in Krakow in March 2015.

References Bahr A, Ellström M, Bergh J, Wallander H. (2015) Nitrogen leaching and ectomycorrhizal nitrogen retention capacity in a Norway spruce forest fertilized with nitrogen and phosphorus. Plant Soil (in press). Birgander J, Rousk J, Olsson PA, 2014. Seasonal variation in the soil microbial community along a temperate grassland fertility gradient. Soil Biol Biochem 76, 80-89. Maienza A, Bååth E (2014) Temperature effects on recovery time of bacterial growth after rewetting the soil. Microb Ecol 68: 818-821. Rath KM, Rousk J, 2015. Salt effects on the soil microbial decomposer community and their role in C cycling – a review. Soil Biol Biochem 81, 108-123. Reischke S, Rousk J, Bååth E, 2014. The effects of glucose loading rates on bacterial and fungal growth in soil. Soil Biol Biochem 70, 88-95. Reischke S, Kumar MGK, Baath E (2015) Threshold concentration of glucose for bacterial growth in soil. Soil Biol Biochem 80: 218-223. Rousk J, Dempster DN, Jones DL, 2013. Transient biochar effects on decomposer microbial growth rates: evidence from two agricultural case-studies. Europ J Soil Sci 64, 770-776. Rousk J, Bengtson P, 2014. Microbial regulation of global biogeochemical cycles. Front Microbiol 5, article 103. Rousk J, Hill PW, Jones DL, 2014. Using the concentration-dependence of respiration arising from glucose addition to estimate in situ concentrations of labile carbon in grassland soils. Soil Biol Biochem 77, 81-88.

18

LUCCI Annual Report – work package 1 Rousk J, Hill PW, Jones DL, 2015. Priming of the decomposition of ageing soil organic matter: concentrations dependence and microbial control. Functional Ecology 29: 285-296. Rousk J, Frey SD, 2015. Revisiting the hypothesis that fungal-to-bacterial dominance characterizes turnover of soil organic matter and nutrients. Ecology (in press) doi: 10.1890/14-1796. Smits MM, Johansson L, Wallander H. (2014) Soil fungi appear to have a retarding rather than a stimulating role on soil apatite weathering. Plant Soil (DOI 10.1007/s11104-014-2222-6)

Division of Nuclear Physics, Physics Department Scientists involved: Prof Kristina Eriksson Stenström, Prof Erik Swietlicki, Prof Bengt Martinsson, Dr Birgitta Svenningsson, Dr Adam Kristensson, Dr Göran Frank, MSc Moa Sporre, Dr Pontus Roldin, MSc Axel Eriksson, MSc Cerina Wittbom, MSc Erik Ahlberg, MSc Maria Berghof, MSc Emilie Hermansson, MSc Johan Friberg, MSc Sandra Andersson, MSc Johan Martinsson. Aerosols, of natural as well as anthropogenic origin, affect the climate by scattering and absorbing sunlight and by promoting cloud formation. According to the recent IPCC AR5 (WG1), these properties result in a net cooling effect on the climate. However, the magnitudes of the different effects are shrouded by very large uncertainties. Thus a better understanding regarding the aerosols and their impact on climate is required. A substantial fraction of the aerosol mass is carbon-containing substances, including non-fossil carbon, and many intricate processes and feedback mechanisms are yet to be properly elucidated and their significance quantified. Some important recent results and activities in the aerosol field resulting from the LUCCI environment, including development of new methods, are: • Brown Carbon (BrC) is a fraction of organic carbon that is highly light absorbing in UV. One of the main sources of BrC are residential wood combustion. Martinsson et al. (2015a) studied the emissions of light absorbing carbon during a combustion cycle in a conventional wood stove. It was found that BrC dominated the light absorbing emissions during the fuel addition phase, i.e. the phase were logs were added to the stove. The other burn phases were dominated by black carbon emissions which absorbs light efficiently through all wavelengths in the UV-visible spectrum. • Martinsson et al (2015b) have performed an investigation of the applicability of stable isotope analysis for aerosol source characterization at rural measurement sites. • An Asian Tropopause Aerosol Layer (ATAL) layer has been observed from satellites (CALIPSO and SAGE II) in connection with the Indian monsoon season. The ATAL is found in the upper troposphere/lower stratosphere at altitudes 13 - 18 km, composed mainly of carbonaceous and sulphurous aerosol. The layer was not detectable in the late 1990th. The last 15 years the ATAL has grown in intensity, presently exerting a regional radiative forcing of -0.1 W/m2 (Vernier et al, 2015). The salient features of the ATAL is reproduced by a sectional aerosol model coupled with the Community Earth System Model (Yu et al, 2015). • A negative feedback mechanism involving terrestrial vegetation, GPP, aerosols and clouds was quantified for a boreal forest environment on the basis of long-term observational data (Kulmala et al, 2014). This negative feedback mechanism is however not strong enough to balance the warming induced by anthropogenic greenhouse emissions. • We have been involved in several European projects dealing with organic aerosols (Beddows et al, 2014, Mann et al, 2014, Fountoukis et al 2014, Paglione et al, 2014). • We have developed satellite retrieval algorithms to deduce cloud microphysical properties in the Nordic and sub-Arctic regions. We show that higher aerosol number concentrations result in smaller cloud droplet effective radius, a sign of an indirect effect of aerosols on clouds. An increase in aerosol loadings results in a

19

LUCCI Annual Report – work package 1 suppression of precipitation rates (Sporre et al, 2014a). A follow-up study on stratocumulus clouds revealed an aerosol effect on droplet effective radius and precipitation rates also for this cloud type, but surprisingly no effect on cloud optical thickness was found (Sporre et al, 2014b). This adds uncertainty to the magnitude of the cooling indirect effect of aerosols on continental clouds. • Our detailed process model for aerosol dynamics, atmospheric chemistry and radiative transfer (ADCHEM) has been modified extensively to better account for secondary organic aerosol (SOA) formation from biogenic precursor gases (Hermansson et al, 2014). The model was applied to a test case in northern Scandinavia. • Our ADCHEM process model has also been modified to better examine chamber data of SOA formation from biogenic precursor gases (Roldin et al, 2014). Our studies show that the gas-particle partitioning of carbonaceous material cannot be described as an equilibrium process, as previously postulated. Several reactions involving organic compounds in a highly viscous particle phase need to be considered to account for the observations. • The Lund University Aerosol Mass Spectrometer (AMS) was also used to study highly time-resolved particulate emissions of organics and PAH from residential biomass combustion (Eriksson et al, 2014). •

A new methodology (NanoMap) was developed to study the geographical extent of atmospheric new particle formation events through analysis of particle number size distribution data (Kristensson et al, 2014).

• Soot ageing and transformation was investigated, both in a smog chamber at Lund University and in an urban near-traffic environment in downtown Copenhagen (Rissler et al, 2014, Wittbom et al, 2014. New smog-chamber experiments on soot ageing took place at Gothenburg University during December 2013January 2014 (Pei et al. 2014, Poulsen, 2015) • The PAM (Potential Aerosol Mass) chamber can be used to simulate up to two weeks of atmospheric ageing within minutes (see abstracts by Ahlberg et al. 2014). It is an important tool to that enables us to study the atmospheric fate of VOC and the resulting SOA, both in the field and in laboratory chamber and flow-tube studies. Professor W. Brune, who introduced the PAM concept, was a guest professor in our group in September 2014. • The Lund University Aerosol Mass Spectrometer (AMS) and Cloud Condensation Nucleus Counter (CCNC) was also used to study particulate emissions of organics, inorganics and soot from ships in the Gothenburg harbor area.Particle concentrations, content of organic and inorganic material and cloud drop formation ability was studied for a set of samples covering one year of the aerosol at the Zeppelin station, Svalbard (Silvergren 2014).

References Ahlberg, E., J. Falk, A. Eriksson, P. Roldin, E. Swietlicki, W.H. Brune, A. Kristensson, B. Svenningsson. 2014. Investigations into enhanced biogenic secondary organic aerosol from ageing experiments of VOC-mixtures. Gordon conference on Biogenic Hydrocarbons & the Atmosphere Interactions in a Changing World, June 29 - July 4, 2014 Spain. Beddows, D.C.S., M. Dall'Osto, Roy M. Harrison, M. Kulmala, A. Asmi, A. Wiedensohler, P. Laj, A. M. Fjaeraa, K. Sellegri, W. Birmili, N. Bukowiecki, E. Weingartner, U. Baltensperger, V. Zdimal, N. Zikova, J.-P. Putaud, A. Marinoni, P. Tunved, H.-C. Hansson, M. Fiebig, N. Kivekäs, E. Swietlicki, H. Lihavainen, E. Asmi, V. Ulevicius, P. P. Aalto, N. Mihalopoulos, N. Kalivitis, I. Kalapov, G. Kiss, G. de Leeuw, B. Henzing, C. O'Dowd, S. G. Jennings, H. Flentje, F. Meinhardt, L. Ries, H. A. C. Denier van der Gon, A. J. H. Visschedijk. Variations in tropospheric submicron particle size distributions across the European continent. 2008–2009. Atmos. Chem. Phys., 14(2014)4327-4348, 2014. Eriksson, A.C., E. Z. Nordin, R. Nyström, E. Pettersson, E. Swietlicki, C. Bergvall, R. Westerholm, C. Boman, J. H. Pagels. Particulate PAH Emissions from Residential Biomass Combustion: Time-resolved Analysis with Aerosol Mass Spectrometry. Env. Sci. Technol., 48(2014)7143-7150. Fountoukis, C., A. G. Megaritis, K. Skyllakou, P. E. Charalampidis, C. Pilinis, H. A. C. Denier van der Gon, M. Crippa, F. Canonaco, C. Mohr, A. S. H. Prévôt, J. D. Allan, L. Poulain, T. Petäjä, P. Tiitta, S. Carbone, A. Kiendler-Scharr, E. Nemitz, C.

20

LUCCI Annual Report – work package 1 O'Dowd, E. Swietlicki, S. N. Pandis. Organic aerosol concentration and composition over Europe: insights from comparison of regional model predictions with aerosol mass spectrometer factor analysis. Atmos. Chem. Phys., 14(2014)9061-9076. Hermansson, E., P. Roldin, A. Rusanen, D. Mogensen, N. Kivekäs, R. Väänänen, M. Boy, E. Swietlicki. Biogenic SOA formation through gas-phase oxidation and gas-to-particle partitioning – a comparison between process models of varying complexity. Atmos. Chem. Phys., 14(2014)11853-11869. Kristensson, A., M. Johansson, E. Swietlicki, N. Kivekäs, T. Hussein, T. Nieminen, M. Kulmala, M Dal Maso. NanoMap: Geographical mapping of atmospheric new particle formation through analysis of particle number size distribution data. Boreal Environment Research, 19(2014)329-342. Kulmala, M., T. Nieminen, A. Nikandrova, K. Lehtipalo, H. Manninen, T. Petäjä, H.-C. Hansson, E. Swietlicki, A. Lindroth, T. R. Christensen, A. Arneth, P. Hari, J. Bäck, T. Vesala, V.-M. Kerminen. CO2 induced terrestrial climate feedback mechanism: From carbon sink to aerosol source and back. Boreal Environment Research, 19(2014)122-131. Mann, G.W., K. S. Carslaw, C. L. Reddington, K. J. Pringle, M. Schulz, A. Asmi, D. V. Spracklen, D. A. Ridley, M. T. Woodhouse, L. A. Lee, K. Zhang, S. J. Ghan, R. C. Easter, X. Liu, P. Stier, Y. H. Lee, P. J. Adams, H. Tost, J. Lelieveld, S. E. Bauer, K. Tsigaridis, T. P. C. van Noije, A. Strunk, E. Vignati, N. Bellouin, M. Dalvi, C. E. Johnson, T. Bergman, H. Kokkola, K. von Salzen, F. Yu, G. Luo, A. Petzold, J. Heintzenberg, A. Clarke, J. A. Ogren, J. Gras, U. Baltensperger, U. Kaminski, S. G. Jennings, C. D. O'Dowd, R. M. Harrison, D. C. S. Beddows, M. Kulmala, Y. Viisanen, V. Ulevicius, N. Mihalopoulos, V. Zdimal, M. Fiebig, H.-C. Hansson, E. Swietlicki, J. S. Henzig. Intercomparison and evaluation of aerosol microphysical properties among AeroCom global models of a range of complexity. Atmos. Chem. Phys. Discuss., 14(2014)4679-4713. Martinsson, J., A. C. Eriksson, I. E. Nielsen, V. Berg Malmborg, E. Z. Nordin, R. Lindgren, R. Nyström, E. Swietlicki, C. Boman, J.H. Pagels. 2015a. Relationship between light absorbing carbonaceous particulate emissions and biomass combustion conditions. Abstract for the NOSA-FAAR conference in Kuopio 11-12 March 2015. Martinsson, J., K. Eriksson Stenström, E. Swietlicki, P.A. E., Olsson P. δ13C – A useful tool of source characterization of the carbonaceous aerosol at rural measurement sites? Abstract in the current annual report. Paglione, M., S. Saarikoski, S. Carbone, R. Hillamo, M. C. Facchini, E. Finessi, L. Giulianelli, C. Carbone, S. Fuzzi, F. Moretti, E. Tagliavini, E. Swietlicki, K. Eriksson Stenström, A. S. H. Prévôt, P. Massoli, M. Canaragatna, D. Worsnop, S. Decesari. Primary and secondary biomass burning aerosols determined by proton nuclear magnetic resonance (1H-NMR) spectroscopy during the 2008 EUCAARI campaign in the Po Valley (Italy). Atmos. Chem. Phys., 14(2014)5089-5110. Poulsen, M. B., 2015, Black Carbon, Effects on the Climate System and Human Health. Master Thesis at Copnehagen Center for Atmospheric Research (CCAR), Copenhagen University. Pei, X, R.K. Pathak, A. Eriksson, E.Z. Nordin, M.B. Poulsen, B. Svenningsson, J. Pagels, E. Swietlicki, M. Hallquist, 2014, Enhancement of light absorption of soot particles coated with sulfuric acid and limonene SOA, International Aerosol Conference IAC 2014, Korea. Rissler, J., E. Nordin, A. Eriksson, P. T. Nilsson, M. Frosch, M. Sporre, A. Wierzbicka, B. Svenningsson, J. Löndahl, M. Messing, J. G. Hemmingsen, S. Loft, S. Sjögren, J. H. Pagels, E. Swietlicki. Aerosol particle effective density and mixing state in a neartraffic urban environment. Env. Sci. Technol., 48(2014) 6300-6308. Roldin, P., A. C. Eriksson, E. Z. Nordin, E. Hermansson, D. Mogensen, A. Rusanen, M. Boy, E. Swietlicki, B. Svenningsson, A. Zelenyuk, J. Pagels. Modelling non-equilibrium secondary organic aerosol formation and evaporation with the aerosol dynamics, gas- and particle-phase chemistry kinetic multi-layer model ADCHAM, Atmos. Chem. Phys., 14(2014)7953-7993. Sporre, M.K., E. Swietlicki, P. Glantz, M. Kulmala. A long-term satellite study of aerosol effects on convective clouds in Nordic background air. Atmos. Chem. Phys., 14(2014a) 2203-2217. Sporre, M., E. Swietlicki, P. Glantz, M. Kulmala. Aerosol indirect effects on low-level clouds over Sweden and Finland. Atmos. Chem. Phys., 14(2014b)12167-12179). Silvergren, S., U. Wideqvist, J. Strom, S. Sjogren, B. Svenningsson. Hygroscopic growth and cloud forming potential of Arctic aerosol based on observed chemical and physical characteristics (a 1 year study 2007-2008), J. of Geophysical ResearchAtmospheres, 119(2014)14080-14097. Vernier J.-P., T.D. Farlie., M. Natarajan, F.G. Weinhold, J. Bian, B.G. Martinsson, S. Crumeyrolle, L.W. Thomason, K. Bedka K. Increase in upper tropospheric and lower stratospheric aerosol levels and its potential connection with Asian Pollution. J. Geophys. Res. Atmos., 120(2015), doi:10.1002/2014JD022372. Yu P., O.B. Toon, R.R. Neely, B.G. Martinsson, C.A.M. Brenninkmeijer. Composition and physical properties of the aerosols in the Asian tropopause aerosol layer and in the North American tropopause aerosol layer. Geophys. Res. Lett., 42(2015) Doi: 10.1002/2015GL063181. Wittbom C., J. Pagels, J. Rissler, A.C. Eriksson, J.E. Carlsson, P. Roldi, E.Z. Nordin, P.T. Nilsson, E. Swietlicki, B. Svenningsson. Cloud droplet activity changes of soot aerosol upon smog chamber ageing. Atmos. Chem. Phys. 14(2014)9831-9854.

21

LUCCI Annual Report – work package 1

Biogeosciences and Remote Sensing groups, Department of Physical Geography and Ecosystem Science Scientists involved: Ass. Prof. Jonas Ardö, Prof. Lars Eklundh, Dr. Jutta Holst, Dr. Sadegh Jamali, PhD student Hongxiao Jin, Prof. Anders Lindroth, Dr. Maj-Lena Linderson, Ass. Prof. Meelis Mölder, PhD student Christian Stiegler, Dr. Elin Sundqvist, PhD student Patrik Vestin. The Biogeosciences and Remote sensing teams are continuing the research on processes related to exchanges of carbon, water and energy in the soil-plant-atmosphere continuum in different type of ecosystems, from drylands in Africa to high Arctic. Some of the highlights are listed below: Methane exchange in forest ecosystems is affected by management and particularly management actions that alter the ground water level and soil moisture. Soil efflux measurements with chambers in undisturbed, thinned and clearcut forests show that these management options turn a small sink (undisturbed forest) into a substantial source (clearcut) during the first years after clear-cutting (Sundqvist et al. 2014). Using a simple empirical model for methane soil efflux in combination with high resolution lidar data and soil moisture modeling made it possible to scale up soil effluxes to larger area (Sundqvist et al. 2015a) and compare with whole ecosystem exchange using tower gradient methods (Sundqvist et al. 2015b). The whole ecosystem exchange showed a diurnal pattern which indicates climate dependent processes beyond temperature and soil moisture. There were also indications of effects from sources at long distance from the tower but consistent with footprint analyses (Sundqvist et al. 2015b) Using eddy covariance method (EC) for methane flux measurements above different type of ecosystems is gaining increasing interest. A comparison between chamber and EC measurements over a lake in central Sweden revealed large differences between the methods (Podrajsek et al 2014). Peltola et al. (2014) compared different types of gas analysers for EC flux measurements and concluded that in spite of differences in performance, fluxes were in general in good agreement. Hydrology is a crucial factor controlling energy exchange in different ecosystems. Peich et al. (2014) concluded that pre-growing season temperature and water table depth were critical factors in control of CO2 exchange from a boreal fen. Öqvist et al. (2014) found the full carbon balance of boreal forests to be highly sensitive to precipitation. The full carbon balance include also the lateral fluxes. Kasurinen et al. (2014) analysed latent heat exchange in different types of high latitude ecosystems including Arctic ones, and they found that the surface resistance showed distinctly different characteristics depending on type of ecosystem. The surface resistance which is mainly controlled by radiation and air humidity is the main controlling parameter for latent heat exchange. Lund et al. (2014) analysed energy exchange in high Arctic tundra ecosystems and found that it was mainly controlled by vapour pressure deficit. The effect of micrometeorological variations on gross photosynthesis was analysed for a boreal site in Sweden showing that it was the variation in light intensity that explained most of the variations within the canopy. Air humidity and CO2 concentration played minor roles. Bosiö et al. (2014) demonstrated the importance of snow cover and connected active layer depth on photosynthesis in a subarctic tundra. The shorter vegetation period in plots with thicker snow cover and deeper active layer was compensated for by more vigorous vegetation growth and thus higher total photosynthesis over the season. In a global study Yuan et al. (2014) demonstrated the importance of the moss layer in boreal forests for assessment of absorbed radiation by remote sensing and, consequently, for the gross photosynthesis. Lin et al. (2015) analysed a global dataset on stomatal conductance and concluded that the optimal stomatal model was a good explanatory factor for the between biome differences in photosynthesis and transpiration. Detection of changes in vegetation cover or properties is important for assessment of vegetation interaction with the climate system and assessments of e.g. net primary production. Jamail et al. (2014 & 2015) analysed different methods to estimate such changes in the Sahel region and proposed modifications to the methods to improve the remotely sensed data. Jin & Eklundh (2014) developed new remotely sensed index for estimation of plant phenology

22

LUCCI Annual Report – work package 1 and Jin & Eklundh (2015) described methods for in situ calibration of light sensors for long term monitoring of vegetation. Tagesson et al. (2014 & 2015) analysed ecosystem properties in grazed semi-arid savanna in West Africa and assessed the dynamics in the carbon balance of this vegetation type. They found unusually high peak CO2 fluxes in these grazed ecosystems but the annual total was similar to other semi-arid ecosystem mainly because of a short growing season.

References Bosiö, J., Stiegler, C., Johansson, M. et al. 2014. Increased photosynthesis compensates for shorter growing season in subarctic tundra – 8 years of snow accumulation manipulations. Climate Change 127:321-334. Jamali, S., Jonsson, P., Eklundh, L, Ardo, J, Seaquist, J. 2015 Detecting changes in vegetation trends using time series segmentation. REMOTE SENSING OF ENVIRONMENT 156:182-195. Jamali, S., Seaquist, J., Eklundh, L., Ardo, J. 2014. Automated mapping of vegetation trends with polynomials using NDVI imagery over the Sahel. REMOTE SENSING OF ENVIRONMENT 141:79-89. Jin, HX and Eklundh, L. 2015. In Situ Calibration of Light Sensors for Long-Term Monitoring of Vegetation. TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 53:3405-3416

IEEE

Jin, HX and Eklundh, L. 2014. A physically based vegetation index for improved monitoring of plant phenology. REMOTE SENSING OF ENVIRONMENT 152:512-525. Kasurinen, V., Alfredsen, K., Kolari, P., Mammarella, I., Alekseychik, P., Rinne, J., Vesala, T., Bernier, P., Boike, J., Langer, M., Belelli Marchesini, L., van Huissteden, K., Dolman, H., Sachs, T., Ohta, T., Varlagin, A., Rocha, A., Arain, A., Oechel, W., Lund, M., Grelle, A., Lindroth, A., Black, A., Aurela, M., Laurila, T., Lohila A. and Berninger, F. 2014. Latent heat exchange in the boreal and arctic biomes. Global Change Biology 20: 3439-56. Lin, Y-S., Medlyn, B.E., Duursma, R.A. et al. 2015. Optimal stomatal behaviour around the world. Nature Climate Change 5:459-464. Lund, M., Hansen, B.U., Pedersen, S.H., Stiegler, C. and Tamstorf, M.P. 2015. Characteristics of summer-time energy exchange in a high Arctic tundra heath 2000-1010. Tellus B 66 doi:10.3402/tellusb.v66.21631. Peichl, M., Öquist, M., Lofvenius, M.O., Ilstedt, U., Sagerfors, J., Grelle, A., Lindroth, A. and Nilsson, M.B. 2014. A 12-year record reveals pre-growing season temperature and water table level threshold effects on the net carbon dioxide exchange in a boreal fen. Environmental Research Letters 9: Peltola, O., Hensen, A., Helfter, C., Marchesini, L.B., Bosveld, F.C., van den Bulk, W.C.M., Elbers, J.A., Haapanala, S., Holst, J., Laurila, T., Lindroth, A., Nemitz, E., Rockmann, T., Vermeulen A.T. and Mammarella, I. 2014. Evaluating the performance of commonly used gas analysers for methane eddy covariance flux measurements: the INGOS inter-comparison field experiment. Biogeosciences 11: 3163-3186. Tagesson, T, Fensholt, R, Cropley, F, Guiro, I, Horion, S, Ehammer, A, Ardo, J. 2015. Dynamics in carbon exchange fluxes for a grazed semi-arid savanna ecosystem in West Africa. AGRICULTURE ECOSYSTEMS & ENVIRONMENT 205:15-24. Tagesson, T, Fensholt, R., Guiro, I. et al. 2015. Ecosystem properties of semiarid savanna grassland in West Africa and its relationship with environmental variability. GLOBAL CHANGE BIOLOGY 21:250-264. Podgrajsek, E., Sahlée, E., Bastviken, D., Holst, J., Lindroth, A., Tranvik, L. and Rutgersson, A. 2014. Comparison of floating chamber and eddy covariance measurements of lake greenhouse gas fluxes. Biogeosciences, 11:4225-4233. Schurgers, G., Lagergren, F., Mölder, M. and Lindroth, A. 2015. The importance of micrometeorological variations for photosynthesis and transpiration in a boreal coniferous forest. Biogeosciences 12:237-256. Sundqvist, E., Vestin, P., Crill., Persson, T. and Lindroth, A. 2014. Short-term effects of thinning, clear-cutting and stump harvesting on methane exchange in a boreal forest. Biogeosciences 11: 6095-6105. Sundqvist, E., Persson, A., Kljun, N., Vestin, P., Chasmer, L., Hopkinson, C. and Lindroth, A. 2015a. Upscaling of methane exchange in a boreal forest using soil chamber measurements and high-resolution LiDAR elevation data. Agricultural and Forest Meteorology (in review). Sundqvist, E., Mölder, M., Crill, P., Kljun, N. and Lindroth, A. 2015b. Methane exchange in a boreal forest estimated by gradient method. Tellus B (in review). Yuan, W., Liu, S., Dong, W., Liang, S., Zhao, S., Chen, J., Xu, W., Li, X., Barr, A., Black, A., Yan, W., Goulden, M.L., Kulmala, L., Lindroth, A., Margolis, H.A., Matsuura, Y., Moors, E., van der Molen, M., Ohta, T., Pilegaard, K., Varlagin, A. and Vesala, T. 2014. Differentiating moss from higher plants is critical in studying the carbon cycle of the boreal biome. Nature Communications, doi: 10.1038/ncomms 5270. Öqvist, M.G., Bishop, K., Grelle, A., Köhler, S.J., Laudon, H., Lindroth, A., Ottosson Löfvenius, M., Wallin, M.B. and Nilsson, M.B. 2014. The Full Annual Carbon Balance of Boreal Forests Is Highly Sensitive to Precipitation. Environmental Science & Technology Letters, dx.doi.org/10.1021/ez500169j.

23

LUCCI Annual Report – work package 1

The response of fungal communities to nitrogen and phosphorus fertilization in a spruce forest Juan Pablo Almeida (a), Nicholas Rosenstock (a) & Hakan Wallander (a) (a) Dept of Microbial Ecology, Lund University

Importance of mycorrhizal communities in nutrient uptake Plant direct nutritional dependence on soils is rather uncommon in many natural ecosystems. For that reason, some plant species rely on mutualistic associations to obtain nutrient from soils [1]. Mycorrhizal associations (symbiosis between filamentous fungal species and plant roots) for instance, play an important role in the biogeochemical cycles because key nutrients like phosphorus and nitrogen are taken by the fungus from the soil and interchanged for photosynthetic carbon [1]. Different ectomycorrhizal species may differ in their abilities to mineralize nitrogen and phosphorus. Therefore, the amount of minerals transfer from the fungus to the plant may vary depending on the fungal community composition forming association with the roots [2]. Moreover, previous studies have shown that ectomycorrhizal fungal diversity can enhance the host productivity by the combination of each species physiological attributes to produce key enzymes and compounds for nutrient mineralization [3].

Experiment A community composition analysis was performed in an experimental Norway spruce forests in Tönnersjöheden. Meshbags to capture ingrowth of fungal mycelia were placed in nitrogen and nitrogen plus phosphorus fertilization sites to evaluate fungal community composition changes in different fertilization regimes. Additionally, to test the response to local additions of phosphorus, some meshbags were filled with apatite (calcium phosphate) and were compared with meshbags filled with sand (quartz). The fungal community composition was estimated using 454 pyrosequencing. The statistical and diversity analysis were performed in the R package Vegan.

Findings The ectomycorrhiza Boletus badius was more abundant in the nitrogen fertilization treatment than in the control and in the nitrogen plus phosphorus treatment. Furthermore, Boletus badius was more abundant in meshbags filled with apatite than in meshbags with quartz in all fertilization treatments. See Figure 1. Previous studies have shown that Boletus badius improves phosphorus and other macronutrients uptake and storage [4]. Its increase in abundance in apatite meshbags may represent an increase in apatite weathering to mine phosphorus in response to phosphorus rich substrates. Moreover, the abundance of Boletus badius in the nitrogen fertilization treatment may be explained by a possible switch in the system from nitrogen limitation to phosphorus limitation.

24

LUCCI Annual Report – work package 1

Figure 1. Fungal abundance among the fertilization treatments in Tönnersjöheden. The different bars represent the relative abundances of the species present in the community. The blue bar represents Boletus badius.

The statistical analysis showed that the ectomycorrhizal communities in apatite meshbags were significantly different from the ectomycorrhizal communities in quartz meshbags in the control and in the nitrogen fertilization treatment (Permanova test: p=0.07 and p=0.034 respectively). In the nitrogen plus phosphorus fertilization treatment there were not significantly differences between apatite and quartz (Permanova test: p=0.95). See Figure 2. Phosphorus fertilization in this treatment may have alleviated the phosphorus limitation. These findings may suggest that the community differences between the fertilization treatments may be driven by the limitation of phosphorus in the system.

a

b

c

Figure 2. Ordination analysis in the fertilization treatments in Tönnersjöheden. a) Control b) Nitrogen fertilization. c) Nitrogen plus phosphorus fertilization. Black communities represent apatite. Red communities represent quartz. The only significant differences between apatite and quartz were found in Nitrogen fertilization and the control.

Ectomycorrhizal diversity did not show any clear differences between the fertilization treatments or between apatite and quartz meshbags. See fig 3. The ectomycorrhizal fungal community composition changes did not result in differences in diversity between the treatments. Fungal communities may change their composition towards more specialized species without altering diversity in such communities.

25

LUCCI Annual Report – work package 1

Sequence reads accumulated Figure 3. Rarefaction curves representing ectomycorrhizal diversity. Control is colored with red tones. Nitrogen is colored with green tones. Nitrogen plus phosphorus is colored with blue tones.

References [1] Gaad, G.M.2006. Fungi in biogeochemical cycles. Cambridge University Press. pp 29-30 [2] Passard, C., et al. 2011. Diversity in phosphorus mobilisation and uptake in ectomycorrhizal fungi. Annals Forests of Science 68: 33–43 [3] Baxter, J.M & Dighton, J. 2005. Phosphorus source alters host plant response to ectomycorrhizal diversity. Mycorrhiza 5: 513–523 [4] Kottke, I., et al. 1998. Xerocomus badius – Picea abies, an ectomycorrhiza of high activity and element storage capacity in acidic soil. Mycorrhiza 7: 267–275

26

LUCCI Annual Report – work package 1

Responses of soil microbial community to warming across a natural geothermal gradient in grasslands of Iceland Zhanfeng Liu (a, b), Bjarni Sigurdsson (c) & Håkan Wallander (a) (a) (b) (c)

Section of Microbial Ecology, Departemnt of Biology, Lund University Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, South China Botanical Garden, Chinese Academy of Sciences Agricultural University of Iceland, Hvanneyri, 311 Borgarnes, Iceland

Understanding and predicting how global warming affects the structure and functioning of natural ecosystems is a key challenge of the 21st century. Over the past few decades, scientist have used a variety of methods including heating cables in the soil, small greenhouse-like structures and heating above plants to simulate the warmer temperatures in the field. Such manipulative experiments have provided useful information on community and ecosystem responses to warming at smaller spatial and temporal scales [1]. However, community and ecosystem dynamics often change both over longer timeframes and across larger spatial scales, manipulative warming experiments may underestimate some community- and ecosystem-scale responses to warming [2]. For this reason, natural thermal gradients, such as geothermal ecosystem, can act as powerful study systems for understanding longer-term, larger-scale community and ecosystem responses to warming in a manner that cannot be achieved by empirical methods [3]. Soil microbial community structure and function is directly regulated by temperature and indirectly by temperature effects on the aboveground plant community, thus global warming may rapidly and dramatically alter the structure and function of soil microbial community. Previous manipulative warming studies showed that warming directly affected microbial community physiology, biomass and composition [4]. Microbial responses to experimental warming were idiosyncractic and highly depedent on warming magnitude and duration. However, most manipulative warming studies were conducted at relative smaller warming magnititude and shorter temporal scale [5]. Consequnetly, the information about how warming magnititude and duration affect soil microbial communities is still scarce, which can be easily disentangled in geothermal ecosystem. In 2014, soils were sampled along a natural geothermal gradients (Ambient, +1℃, +3℃, +5℃, + 10℃,+15℃) at newwarmed grassland site (about 10 years) and old-warmed grassland site (about 700 years) in Iceland (the Forhot gradient). Phospholipid fatty acids (PLFA) method was used to characterize soil microbial community composition. We aimed to examine how warming magnititude and duration affects soil microbial community composition along natural geothermal gardients in Icelandic grasslands. We hypothesized that (1) Bacteria are more adaptive to the new environment, growing faster, higher turnover rate and species replacement rate than fungi across the warming gradient; (2) With warming magnitude increasing, selection pressure will increase, thermophiles or thermotolerant species will be dominant and saturated and long-chained PLFAs will be abundant at higher temperature; (3)Soil microbial community composition at the long-term warmed site will be different from that at the short-term warmed site due to selection pressure and substrate availability. By addressing above questions, our study can provide valuable information and improve our understanding the transient and equilibrial responses of soil microbial communities to warming in natural ecosystems. (continues on the next page)

27

New warmed site

20

Old warmed site (A)

a

a c

10

a

a

a

c

5

b d

e

c

0 300

(B)

(E)

a

250

a

a

200

ab ab

150

c bc

100

bc

c

cd d

50

d

0 .10

Fungal : bacterial PLFAs ratio

(D)

b

15

-1

Bacterial PLFAs (nmol g dry soil)

Fungal PLFAs (nmol g-1 dry soil)

LUCCI Annual Report – work package 1

(C)

(F)

.08

.06

.04

.02

0.00 Ambient +1

+3

+5

+10

+15

Ambient +1

+3

+5

+10

+15

Warming gradient

Figure 1 Responses of fungal PLFAs (A, D), bacterial PLFAs (B, E) and fungal to bacterial PLFAs ratio (C, F) across natural geothermal gradients at new-warmed and old-warmed natural grasslands in Iceland. Within each panel, bars topped with the same letters are not significantly different at P < 0.05 (Duncan’s test).

We found that soil microbial communities were strongly structured by warming magnititude, but not by warming duration. PLFA abundance of bacteria and fungi showed a similar pattern to warming and were significantly decreased without significant changes of fungal to bacterial ratio across the warming gradients both at the newwarmed site and old-warmed site, see Figure 1. PLFA pattern was observed to show a clear shift across the warming gradient. Saturated, branched, long-chain PLFAs, such as Gram positive bacteria were more abundant under higher temperature, while unsaturated and short-chain PLFAs, such as Gram negative bacteria were more dominant under lower temperature, see Figure 2. No significant effect of warming duration on PLFAs patterns were observed suggesting that substrate can be rapidly depleted in a relative short period (less than 10 years) in the studied systems. Our results provided direct empirical evidences for a shift in PLFA patterns related to warming magnititude and warmig duration in natural grasslands. Such information is quite valuable and helpful for disentangling the underling mechanism about ecosystem responses to the future global warming.

28

LUCCI Annual Report – work package 1

Figure 2. Principal component analysis (PCA) of PLFA data across natural geothermal gradients at new-warmed and oldwarmed grassland sites. (A) Biplot of sample scores. Samples were labeled by warming magnititude: circle = Ambient, square = + 1℃, diamond = +3℃, box = +5 ℃, star = +10 ℃, up triangle = +15℃. The samples labeled by filled symbols were from old-warmed grassland site; the samples labeled by open symbols were from new warmed grassland site. (B) Biplot of individual PLFAs loadings.

References [1] Wu Z, Dijkstra P, Koch GW, et al. 2011. Responses of terrestrial ecosystems to temperature and precipitation change: a meta-analysis of experimental manipulation. Global Change Biology 17: 927–42. [2] Wolkovich EM, Cook BI, Allen JM, et al. 2012. Warming experiments underpredict plant phenological responses to climate change. Nature 485: 494–497. [3] O'Gorman EJ, Benstead JP, Cross WF, et al. 2014. Climate change and geothermal ecosystems: natural laboratories, sentinel systems, and future refugia. Global Change Biology 20: 3291–3299. [4] Frey SD, Lee J, Melillo JM, et al. 2013. The temperature response of soil microbial efficiency and its feedback to climate. Nature Climate Change 3: 395–398. [5] Stewart RIA, Dossena M, Bohan DA et al. 2013. Mesocosm experiments as a tool for ecological climate-change research. Advances in Ecological Research 48: 71–181.

29

LUCCI Annual Report – work package 1

δ13C – A useful tool of source characterization of the carbonaceous aerosol at rural measurement sites? Johan Martinsson (a), Kristina Eriksson Stenström (a) Erik Swietlicki (a) and Pål-Axel Olsson (b) (a) Dept of Nuclear Physics, Lund University (b) Dept of Biology, Lund University

Introduction The ambient air contains various amounts of particles, aerosols. These can be of anthropogenic or natural origin. The carbonaceous fraction of the atmospheric aerosol has considerable effects on climate and human health. This fraction can be divided into organic carbon (OC) and elemental carbon (EC). EC or black carbon (BC) is formed during incomplete combustion and is strongly light absorbing and thus warms the climate. OC is emitted from various sources (fossil fuel combustion, biomass combustion and biogenic emissions) and was formerly believed to scatter solar radiation and thereby to have a cooling effect on the climate. However, recent studies [1] have shown that fractions of OC can be strongly light. Due to the different climate effects of OC and EC, and of different types of OC, large uncertainties are associated with the net effect of carbonaceous aerosols on the climate. If the sources of the atmospheric aerosols are known, both qualitatively and quantitatively, the uncertainty of the climate effect of carbonaceous aerosols can be lowered. There are many techniques that are used to characterize the sources of atmospheric carbonaceous aerosols. Two of the most common are radiocarbon (14C) and levoglucosan [5]. These are tracers for biomass combustion and fossil fuel combustion, respectively. Stable carbon isotopes have widely been used to characterize the sources of atmospheric aerosol since the 1980’s [2,3,4]. In principle, depending on the sources of particles they have specific isotopic signatures. Several studies has been carried out were δ13C has been measured in various environments [6,7,8,9]. When measuring δ13C next to the source the measured δ13C value reflects the δ13C value of the source. For example, if δ13C are measured in aerosols on a highly trafficated highway, the δ13C value will most probably reflect the δ13C signature of gasoline or diesel. However, many aerosol measurement stations are rural and thus influenced by a large number of various anthropogenic and biogenic activities. A problem is that many aerosol sources overlap in δ13C signatures (Figure 1). The case of overlapping values of δ13C for different sources can be difficult to handle when applying source characterization of the aerosol. One way to treat these difficulties is to measure δ13C on specific fractions of the carbonaceous aerosol. For instance, separate the OC from the EC and measure δ13C on both fractions, separately. However, at present, there are no robust standardized methods for separating OC from EC. Due to the lack of a robust separation method it is still desirable to measure δ13C on the total carbon (TC). The applicability of δ13C for source apportionment studies can thus be questioned at these rural sites. Filters containing atmospheric aerosols were analyzed for levoglucosan and radiocarbon. Remaining filter material (TC) was analyzed for δ13C for comparison with the above mentioned tracers to see if the δ13C value is susceptible for influences of biomass and fossil fuel combustion and to evaluate if δ13C contributes to an improved source characterization at a rural sampling site.

Methods Sampling was conducted on the EUSAAR and EMEP background station Vavihill in southern Sweden (56°01’ N, 13°09’ E, 172 m.a.s.l.). The station is located on a pasture which is occasionally visited by grazing cattle. The surrounding land is mainly deciduous forest. The closest large pollution sources are the cities of Copenhagen, Malmö and Helsingborg which are located west and southwest of the station with distances of 50, 45 and 25 km, respectively. Particles were collected weekly (168 h) from May 2008 to April 2009 on 47 mm quartz-fiber filters (n=25). After sampling, filters were placed in petri dishes, wrapped in aluminum foil and stored in a refrigerator at +5°C until analysis.

30

LUCCI Annual Report – work package 1 OC/EC analysis was conducted on a 0.5 cm2 punch of the original filter using a DRI Model 2001 OC/EC Carbon Analyzer (Atmoslytic, Calabasas, CA, USA). Stable isotopes of carbon were measured on a filter punch using an elemental analyzer connected to isotope ratio mass spectrometry (EA-IRMS) at the newly installed IRMS facility at the Department of Biology, Lund University.

Results and Discussion Figure 1 and 2 shows that the δ13C varied between -26.73 ‰ and -25.54 ‰ with a mean of -26.16 ‰ during the sampling period. These values overlap with several sources, such as combustion of fossil fuels like gasoline, diesel, natural gas and fuel oil (figure 1). Within this interval we also find charcoal, C3 plants and pollen. It is thus difficult to conclude any sources from such a narrow interval that overlap with the values for several potential sources. ? 13 C(‰) -40 Source

? 13 C(‰)

Natural gas 1,2

-27.0 - -23.4

Gasoline 1,2,3

-26.0 - -20.6

Diesel 1,2,3

-28.0 - -24.2

Fuel oil 1,2

-31.0 - -24.0

Charcoal 3,4

-27.4 - -21.7

Coal 1,2,3,5 Waste incinerators

-36

-32

-28

-24

-20

-16

-12

-8

-25.0 - -21.0 2

-39.0 - -29.0

C 4 plants 3,5,6,7

-19.9 - -11.5

C 3 plants 3,5,6,7

-34.7 - -23.8

Pollen 8

-26.2 - -23.5

Animal -dung 4

-34.9

Figure 1. Data for particulate δ13C from various sources from the litterature. 1[6]; 2[7]; 3[8]; 4[9]; 5[10]; 6[11]; 7[12]; 8[13]. The yellow area illustrate the measured δ13C range in this study.

Table 1 shows that the δ13C correlates poorly (0.012
13

δ C (‰)

The relationship between δ13C and the other parameters (levoglucosan, OC and EC) show very low R2-values (0.012
Figure 2. δ13C values for aerosols when the measurement periods were divided into seasons.

31

LUCCI Annual Report – work package 1 Table 1. R2-values for measured parameters. 13

δ C

14

f C

Levoglucosan

OC

EC

13

δ C 14

f C Levoglucosan OC EC

0.133 0.047

0.210

0.012

0.079

0.344

0.063

0.193

0.518

0.413

The overlap between sources with respect to δ13C signatures toghether with the poor correlation to other strong source tracers (14C and levoglucosan) makes source identification and characterization of the rural aerosol TC fraction difficult using this method. Future work will aim to measure δ13C on the separated fractions of TC (OC and EC) for improved source charazterization of the carbonaceous aerosol.

References [1] Kirchstetter et al. (2004). Journal of Geophysical Research. 109, D21208. [2] Cachier, H. (1989). Aerosol Science and Technology, 10:2, 379-385. [3] Cachier et al. (1985). Journal of Atmospheric Chemistry. 3, 469-489. [4] Chesselet et al. (1981). Geophysical research letters. 8, 345-348. [5] Genberg et al. (2011). Atmos. Chem. Phys. 11, 11387-11400. [6] Widory et al. (2004). Atmospheric Environment, 38, 953-961. [7] Widory, D. (2006). Combustion Theory and Modelling, 10, 831-841. [8] Kawashima et al. (2012). Atmospheric Environment. 46, 568-579. [9] Agnihotri et al. (2011). Atmospheric Environment. 45, 2828-2835. [10] Cao et al. (2011). Atmospheric Environment. 45, 1359-1363. [11] Das et al. (2010). Organic Geochemistry, 41, 263-269. [12] Turekian et al. (1998). Chemical Geology, 152, 181-192. [13] Jung et al. (2011). Atmospheric Chemistry and Physics, 11, 10911-10928.

32

LUCCI Annual Report – work package 1

BVOC emission variation caused by intra-genotypic variation, canopy height and change in time Ylva Persson (a), Guy Schurgers (b), Min Wang (a) & Thomas Holst (a) (a) Dept of Physical Geography and Ecosystem Sciences, Lund University (b) Dept of Geoscience and Natural Resource Management, Copenhagen University

Introduction Biogenic Volatile Organic Compounds (BVOCs) emitted from the vegetation into the atmosphere consist of a range of trace gases other than CO2 and CH4 [1]. Studies have been performed on a range of plant species in order to quantify their emission rates, but there exists high uncertainties regarding the influence of local growing conditions, seasonality and height within the canopy at which the measurements were performed [1-6]. Furthermore, it has also been found that significant differences in emission patterns can occur due to genetic variation within a species, which produce high uncertainties in atmospheric models [7]. Here we report the BVOC emission patterns from individuals of three common adult European tree species grown under natural conditions, and the change in emission patterns in regards to height within the canopy and the time measurements were performed during the growing season.

Methods The field study was carried out at Højbakkegård Experimental Station in Taastrup, Denmark (55°40´ N, 12°18´ E), which is involved in a network called International Phenological Gardens (IPG). This network studies long-term phenological patterns for the most common European tree and bush species throughout Europe and how phenology is influenced by the change in climate. All plants used by the network are genetically identical, with the advantage that it restricts the genetic variation between sites [8]. The study was performed between 4th of June and 15th of August in 2013 on seven adult trees; two English oaks, four Norway spruces and one European beech. The Norway spruces from two different provenances are classified into early spruce and late spruce due to a distinct difference in time of budburst. Photosynthesis and intercellular CO2 concentration in relation to net C assimilation were made with a portable infra-red gas analyzer (LI-6400, LI-COR, Lincoln, NE, USA) and sampling of chamber air for the BVOC emission concentration was done for three height levels (2m, 5.5m and 12.5m) within the canopy for each of the measured trees. In order to measure possible changes with time in the BVOC emission patterns, measurements were performed twice for the oaks and early spruces; in June and August for oak and in July and August for early spruce. Finally, all emission rates were standardized in order to make better comparisons between individuals with the help of an algorithm produced by Guenther et al. [9].

Results and discussion Intra-genotypic variation There was little difference in emission patterns between the individuals of the same species (figure 1). Oak was mainly emitting isoprene, whilst spruce and beech are mainly monoterpene emitters. There was little difference in the total amount of BVOC emissions between the two subspecies of Norway spruce. However, there was a difference in the main compounds released, where early spruce emitted more limonene and late spruce emitted more α-pinene (data not shown). This suggests that the emission patterns may vary with genotypes, which has also been shown in previous studies [7,10].

(continues on the next page)

33

LUCCI Annual Report – work package 1

Figure 1. Standardized emission rate per gram of dry weight (gdw) for English oak, Norway spruce with an early and late May shoot (called early spruce and late spruce) and European beech between June and August in 2013.

Canopy height Figure 2 shows the total standardized emission pattern for the different species and with different heights within the canopy. The biggest difference in emission patterns in regards to canopy height was for Beech, where the top of the canopy released 7-9 times more in comparison to lower levels. The change in emission patterns in relation to height of the canopy was not as distinct for the other measured individuals. This lack of dependence can be related to a wide spacing of approximately 10 m in between rows of plants, which likely results in a reduced difference to light adaptation between different heights within the tree.

Figure 2. Total standardized emission rate per gram of dry weight (gdw) for English oak, Norway spruce with an early and late May shoot (called early spruce and late spruce) and European beech at heights of 2m, 5.5m and 12.5m.

Change with time of season The BVOC emission patterns for both oak and early spruce changed with the progression of time. Both oak trees not only increased their emission rates from June to August, but they also changed from a more mixed release of different

34

LUCCI Annual Report – work package 1 compounds to releasing approximately 97% of only one compound, namely isoprene. The change is believed to be caused both by an insect attack in June, which has been shown to induce the emission of other compounds than isoprene [1,11] and due to leaf development as the season progressed. For the early spruce trees, their total emission decreased over the time period between July and August, but without a loss in the number of released compounds. During this time, there was a prolonged period without precipitation (data not shown) of three weeks and one of the trees started to drop its needles on the lower levels by the time the second measurement period was performed. The drop in emissions could therefore be an indication of plant water stress.

Figure 3. Variation in standardized emission rate of the most common BVOCs per gram of dry weight (gdw) for English oak and Norway spruce with an early May shoot. The English oaks were measured in June and in August, whilst the Norway spruces were measured in July and August.

The results from Taastrup show there was no clear intra-genotypic variation between individuals and highlight the importance of measurements performed at different canopy heights and time in the growing seasons. Therefore, more studies of a similar setup are needed in order to evaluate how the emission patterns of plants are influenced by the above mentioned variables. In a future study, different sites from the IPG network will be sampled to study the effects of growing conditions on BVOC patterns.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]

J. Kesselmeier et al. 1999. J Atmos Chem 33, 23-88. R. Steinbrecher et al. 2009. Atmos Environ 43, 1380-1391 H. Hakola et al. 2001. Boreal Environ Res 6, 237-249. S.M. Owen et al. 2001. Atmos Environ 35, 5393-5409. H. Hakola et al. 2012. Atmos Chem Phys 12, 11665-11678. M. Šimpraga et al. 2013. Atmos Environ 80, 85-95. J. Bäck et al. 2012. Biogeoscienses 6, 2709-2718. F.-M. Chmielewski et al. 2013. In: M.D. Schwartz (Eds): Phenology: An integrative Environmental Science, 137-153. A. Guenther et al. 1993. J Geophys Res 98, 12609-12617. M. Staudt et al. 2001. Can J Forest Res 31, 174-180. M. Lerdau et al. 1997. BioScience 47, 373-383.

35

LUCCI Annual Report – work package 1

Salt sensitivity of soil microbial processes Kristin M Rath (a), Arpita Maheshwari (a), Per Bengtson (a) & Johannes Rousk (a) (a) Dept of Biology, Microbial Ecology, Lund University

The amount of carbon stored as soil organic matter (SOM) constitutes a pool more than double the size of the atmospheric carbon pool. Soil respiration represents one of the largest fluxes of carbon between terrestrial ecosystems and the atmosphere. A large fraction of the CO2 released by soils is produced by the microbial decomposition of SOM. The microbial carbon budget is characterized by their carbon use efficiency, i.e. the partitioning of substrate into growth and respiration. This will shape the role of the soil as a net source or sink for carbon. In addition to their role in soil C dynamics, microbes are also important for making nutrients bound in organic matter available in inorganic form through organic matter breakdown. One of the canonical factors known to influence microbial processes in soil is pH. In aquatic systems salinity has been found to have a comparably strong influence as pH. However salinity remains understudied in soil, despite its growing relevance due to land use change and agricultural practices. The aim of this study was to understand how microbial carbon dynamics and processes related to soil nitrogen cycling respond to disturbance by changing environmental conditions, using salinity as a reversible stressor. First, we compiled a comparative analysis of the sensitivity of different microbial processes to increasing salt concentrations. Second, we compared different salts to determine whether salt toxicity depended on the identity of the salt. The sensitivity of microbial processes to salt was investigated by establishing inhibition curves in order to estimate EC50 values (the concentration resulting in 50% inhibition). These EC50 values were used to compare bacterial and fungal growth responses, as well as catabolic processes such as respiration and nitrogen mineralisation. Our results suggest that growth related measures are more sensitive to salinity than catabolic processes (Fig 1). This could be an indication that at higher salt concentrations, the microbial community allocates less carbon towards growth, resulting in reduced carbon use efficiency. In addition, fungi appeared to be less affected by salinity than bacteria. We also found that microbial processes show different sensitivities depending on the kind of salt used in the experiment. Chloride salts tended to be more toxic than sulphate salts.

(continues on the next page)

36

LUCCI Annual Report – work package 1

A

norm. rel. process rate

1.00 0.75

fungal growth respiration

0.50 0.25 0.00

bacterial growth -5

-4

-3

-2

-1

0

NaCl (log (mmol/g)

rel. normalised process rate

B

1.0

+

NH production 4

0.8 +

NH consumption 4

0.6 0.4 0.2

-

NO production 3

0.0

-4

-2

-1

0

1

-1

NaCl (log (mmol g )) Figure 1. Inhibition curves of different microbial processes in response to acute salt exposure: (A) bacterial growth rate measured as incorporation of 3H-labelled leucine, fungal growth rate measured as incorporation of 14C-labelled acetate into ergosterol, and soil respiration; (B) gross nitrogen transformation rates. Process rates were normalized to the value measured in the control.

37

LUCCI Annual Report – work package 1

Using pollution-induced community tolerance (PICT) to identify direct salt effects on the soil microbial community Kristin Rath (a), Daniel V. Murphy (b), Francisco Dini-Andreote (c) & Johannes Rousk (a) (a) Dept of Biology, Microbial Ecology, Lund University (b) Soil Biology and Molecular Ecology Group, School of Earth and Environment, Institute of Agriculture, The University of Western Australia

(c) Department of Microbial Ecology, Centre for Ecological and Evolutionary Studies, University of Groningen Soil salinization poses a major threat to agricultural productivity in many areas of the world. While negative impacts of salinity on plant growth have been exhaustively assessed, a much smaller number of studies have focused on the effect of soil salinity on the soil microbial community. Naturally saline soils have been found to show lower microbial acitivity than non-saline soils. However, one problem affecting studies looking at salt effects in naturally saline soils is that direct salt affects on the soil microbial community are difficult to distinguish from indirect effects due to other confounding factors, such as pH or organic matter content. In order to overcome this problem we employed the ecotoxicological concept of pollution-induced community tolerance (PICT). The basic idea behind the PICT method is that if a community is exposed to a toxicant at an ecologically relevant concentration, the average tolerance of the community to the toxicant should increase. This increase in community tolerance is brought about by two main mechanisms: (1) physiological adapations, and (2) selection for more tolerant strains and species. Therefore a preexposed community should have developed a higher community tolerance to that toxicant than an unexposed community (Fig.1). We employed this concept to identify direct salt effects on microbial communities by linking community salt tolerance to in situ salt concentration along natural salinity gradients. We studied soil samples coming from three replicated natural salinity gradients encompassing a wide range of salinity levels (from 0.1 dS m-1 to >10 dS m-1). Two of the salinity gradients are located at the shore of Lake O’Connor, a former salt lake in Western Australia, the third salinity gradient is situated on Schiermonnikoog, an island off the Dutch North Sea coast. After a two week long pre-incubation with ground plant material we first measured bacterial and fungal growth as well as respiration along the Australian salinity gradients. We found that both bacterial growth and respiration declined with increasing in situ salinity, while fungal growth did not show a clear negative trend. In the second part of the experiment we wanted to assess whether high salt concentrations have exerted a selective pressure on the community. We estimated community tolerance to salt by measuring the bacterial growth response to added NaCl in a soil suspension. If salinity was an ecologically relevant factor it should have selected for a more salttolerant community, indicated by lower inhibition by added salt (Fig. 1). We observed a clear increase of bacterial community tolerance along both the Australian (Fig. 2) and the Dutch salt gradients (Fig.3) with in situ soil salinity. In samples from the low-saline end of the gradient, bacterial growth was inhibited by NaCl. Bacterial tolerance indicated by the concentration inhibiting 50% of the growth rate (IC50) increased with in situ salinity (Fig. 3). In soil samples with in situ salt concentrations of more than 2 dS m-1 bacterial growth was no longer inhibited by adding high concentrations of NaCl to the bacterial soil suspension. In fact, adding NaCl promoted bacterial growth rates. Our result suggests that salt has selected for a community composed of more tolerant members at the highly saline end of the gradients.

38

LUCCI Annual Report – work package 1

1.2

relative bacterial growth

1.0

0.8 PICT

0.6

0.4

0.2

0.0 0

1

2

3

4

5

log salt concentration Figure 1. Example of a PICT-assay of salt tolerance of the bacterial community. The black curve shows the tolerance of a sensitive community (the control treatment) and the blue curve that of a more salt tolerant community (as induced by salt exposure). The difference between the curves is a measure of the Pollution Induced Community Tolerance (PICT).

Figure 2. Representative examples of dose-response curves of bacterial growth rates to additions of NaCl from the Australian salinity gradients. Bacterial growth rates have been normalized to the values measured in the control treatment without added NaCl. Blue colours indicate samples from the low salinity end of the gradients, green and yellow colours represent samples from the middle of the gradient, and red curves are from samples from the highly saline end of the gradient. In situ soil salinity is given on the right site of the picture as electrical conducitivy measured in a 1:5 soil:water suspension.

39

LUCCI Annual Report – work package 1

Figure 3. IC50 values (concentration of added NaCl at which bacterial growth rate is inhibited by 50%) measured along the Dutch salinity gradients plotted against in situ Na+ concentration in the soils. Samples were taken along a chronosequence ranging from 0-8 year old sites. Samples were taken four times (in May, July, September and November). (n=3)

40

LUCCI Annual Report – work package 1

Soil microbial decomposer responses to salt: A review Kristin Rath (a) & Johannes Rousk (a) (a) Dept of Biology, Microbial Ecology, Lund University

Salinization of soil is recognised as one of the most pressing environmental challenges to resolve for the next century. We conducted a synoptic review of the available research on how salt affect decomposer microbial communities and carbon (C) cycling in soil [1]. After summarizing known physiological responses of microorganisms to salinity, we went on to provide a brief overview and qualification of a selection of widely applied methods to assess microorganisms in soil to date. The dominant approaches to characterise microbial responses to salt exposure have so far been microbial biomass and respiration measurements. We compiled datasets from a selection of studies and found that (1) microbial biomasscarbon (C) per C held in soil organic matter shows no consistent pattern with long-term (field gradients) or short-term (laboratory additions) soil salinity level (Fig. 1A), and (2) respiration per soil organic C shows clear patterns for a substantial inhibition of respiration with higher salt concentrations in soil, that are consistent for both short-term and long-term salinity levels (Fig. 1B). Patterns that emerge from extra-cellular enzyme assessments are more difficult to generalize, and appear to vary with the enzyme studied, and its context. Growth based assessments of microbial responses to salinization are largely lacking. Relating the established responses of microbial respiration to that of growth could provide an estimate for how the microbial C-use efficiency would be affected by salt exposure. This would be a valuable predictor for changes in soil C sequestration as well as for plant nutrition. Tolerance assays for microbial communities from environmental samples have been successfully used to link microbial community responses to specific chemical factors (including e.g. metals, antibiotics and phenolics). A few studies have investigated the connection between microbial tolerance to salt and the soil salinity levels, but so far results have not been conclusive. However, we predict that more systematic inquiries including comprehensive ranges of soil salinities will substantiate the predicted connection, which would also confirm that salinity has a direct effect on the composition of microbial communities. While salt has been identified as one of the most powerful environmental factors to structure microbial communities in aquatic environments, no up-to-date sequence based assessments currently exist from soil. Filling this gap should be a research priority. Moreover, linking sequencing based assessments of microbial communities to their tolerance to salt would have the potential to yield biomarker sets of microbial sequences. This could provide predictive power for, e.g., the sensitivity of agricultural soils to salt exposure, and, as such, a useful tool for soil resource management. We conclude that salt exposure has a powerful influence on soil microorganisms. In addition to being one of the most pressing agricultural problems to solve, this influence could also be used as an experimental probe to better understand how microorganisms control the biogeochemistry in soil.

(continues on the next page)

41

LUCCI Annual Report – work package 1

A

Biomass (normalised to 1)

2.5

Batra and Manna (1997)

2.0

Chowdhury et al. (2011c) Egamberdieva et al. (2010) Iwai et al. (2012)

1.5

Mavi and Marschner (2012) Muhammad et al. (2006) Muhammad et al. (2008)

1.0

Ramirez-Fuentes et al. (2012) Sardinha et al. (2003) Tripathi et al. (2006)

0.5

Wong et al. (2008) Yuan et al. (2006)

0.0

0

1

2

-1

Electrical conductivity (log (dS m ))

Respiration (normalised to 1)

B

Baumann and Marschner (2013) Chowdhury et al. (2011c) Garcia and Hernandez (1996) Iwai et al. (2012) Laura (1974) Mavi and Marschner (2012) Mavi et al. (2012) Mavi and Marschner (2013) Muhammad et al. (2006) Muhammad et al. (2008) Pathak and Rao (1998) Ramirez-Fuentes et al. (2002) Sardinha et al. (2003) Saviozzi et al. (2011) Setia et al. (2011a) Tripathi et al. (2006) Wong et al. (2008) Yuan et al. (2007)

1.5

1.0

0.5

0.0

0

1

2 -1

Electrical conductivity (log (dS m ))

Figure 1. Literature data compilation of (A) microbial biomass C per C in soil organic matter, and (B) soil respiration. in soils of different salinities. Biomass and respiration values have been normalized to the value measured in the control level or the soil with the lowest salinity level included in the study.

References [1] Rath, Kristin M and Rousk, Johannes. Salt effects on the soil microbial decomposer community and their role in organic carbon cycling: A review. Soil Biology and Biochemistry 81 (2015), pp. 108-123

42

LUCCI Annual Report – work package 1

Aging soil organic matter: concentration dependence and microbial control of the priming effect Johannes Rousk (a), Paul W. Hill (b), & Davey L. Jones (b) (a) Microbial Ecology, Dept of Biology, Lund University, Sweden. (b) School of the Environment, Natural Resources and Geography, Bangor University, UK

Abstract

(a) Respiration

1000 800

250 200 150

600

100 50 0

400

0

5

10 15 20 25

200 0

0

50

100

150

1.28

(b)

14

C-SOM mineralisation

500 400 300

1500200 100 0 0

1000

5

10 15 20 25

500 0

0

50

-1

Fungal biomass (µg C g )

40

4 2

30 0 0

5

10 15 20 25

20 10 0

0

50

100

Time (h)

100

150

Time (h)

(c) Bacterial growth

50

160

2000

Time (h)

-1

2

Cumul leu inc in extr bacteria (nmol g ) Cumul respiration (µg CO -C g-1)

13 months, September

Cumul SOM respiration (dpm 14C)

The amount of carbon stored in soil is a regulator for the global climate and soil fertility, and is the balance between formation and decomposition of soil organic matter. Decomposition of soil organic matter can be affected by high quality carbon supplements, in the form of e.g. sugars or vitamins that can occur close to plant roots. A stimulation of the decomposition of soil organic matter induced by high quality carbon additions is termed ‘priming’ and the mechanisms for this phenomenon remain elusive. The most common explanation assigns priming to successional growth responses of groups within the microbial community specialised at processing different forms of carbon. These specialisations include those adept at degrading complex organic materials, or high quality carbon; a division that has also been connected with fungal (complex carbon) and bacterial (high quality carbon) decomposers. Organic matter relatively freshly formed from plant carbon input to soil has also been found to be particularly sensitive to priming. We investigated how the concentration of added high quality carbon influenced priming, if the age of the soil organic matter affected priming, and if priming was related to bacterial or fungal growth responses triggered by the high quality carbon additions [1]. To create an age gradient of traceable soil carbon, we spiked a pasture soil using sugar with a radioactive marker (14C), and subsampled plots during 13 months after application. A range of concentrations of sugar was then added in subsequent laboratory experiments, and respiration, soil carbon decomposition (14C tracing), bacterial growth rates and fungal biomass were tracked (Fig. 1).

150

(d) Fungal biomass

250

60

200

40 20

150

0 0

100

5

10 15 20 25

50 0

0

50

100

150

Time (h)

0.0 0.26 1.28

6.4 32 160

800 4000

Figure 1. Cumulative respiration (panel a), cumulative SOM mineralisation (panel b), cumulative bacterial growth (panel c) and the fungal biomass concentration (panel d) during the ca. 1 week study period for the 13-months field aged SOM. The legend denotes the glucose concentrations (0-4000 µg C g-1). Values are the mean and error bars ± 1 SE.

43

LUCCI Annual Report – work package 1

14

C-SOM minerlisation (log(dpm g -1 ))

The decomposition of soil carbon aged 2-13 months showed similar concentration dependencies to added sugar, and priming increased the decomposition the soil carbon by up to 350%. We found no connection between the successional growth responses of microbial groups specialised at differently complex carbon and the priming of soil carbon decomposition. It has been suggested that enzymes that the microbial community degrade soil carbon with could remain active after microbial death. This could explain the lack of connection to the microbial community growth responses but clear concentration dependence to high quality carbon additions (Fig. 2).

5-Mo, 23h 5-Mo, 144h 13-Mo, 28h 13-Mo, 173h

1-Day, 22h 1-Day, 69h 2-Mo, 22h 2-Mo, 148h

(a)

5

(b)

(c)

4 3 2 1 0

1

2

3

4 -1

Respiration (log(µg CO -C g )) 2

3

4

5 0 -1

Bacterial growth (log(pmol Leu g ))

1

2

3 -1

Fungal biomass (log(µg C g ))

Figure 2. The dependency of SOM mineralisation on respiration (panel a), bacterial growth (panel b) and fungal biomass (panel c). A type-II major axis linear regression was used to test the relationships between logarithmically transformed values. The relationship was tested overall, and for the 1-day and 7-day study periods for each SOM-age timepoint, separately (see legend for symbol definitions). Only statistically significant relationships (P<0.05) are presented.

Alternative mechanisms for the priming of SOM mineralisation The rate-limiting step for SOM mineralisation in soil is thought to be the activity of oxidative enzymes including peroxidases that e.g. depolymerise SOM macromolecules, and produce soluble substrates for microbial assimilation. In a secondary step, thought to be exclusively intracellular, the labile C substrates are metabolised, and CO2 is released. The availability of H2O2 will govern the activity of the rate-limiting SOM oxidising enzymes, and the H2O2 supply, in turn, is regulated by a range of different oxidases that use simple labile C sources, including e.g. sugars such as glucose, as substrate. Thus, it has been forwarded, that the underlying mechanism for the priming of SOM mineralisation by labile C addition could be due to the stimulation of H2O2 production, inducing a increased activity of e.g. peroxidases that trigger the solubilisation (via e.g. depolymerisation with peroxidases or cell wall deconstruction with endoglucanase) and subsequent respiration of SOM. Further, it has recently been suggested that intracellular soil enzymes can be stabilized in soil, and that an intact reaction chain from SOM to CO2 consequently can be maintained in soil, contributing to a substantial fraction of the CO2 evolved, even in the absence of metabolising cells. Extracellular enzymes are abundant in the soil matrix and numbers are relatively stable over time. Taken together, it is possible that H2O2 would show a glucose concentration dependence; if this indeed is the rate limiting step for the transformation of SOM to monomer C substrates that immediately are oxidised to CO2 by the resident microbial community or by stabilized intracellular enzymes in the soil matrix, the lack of a link to simultaneous microbial growth rates could be explained. Rather, the historical conditions that have produced the stock of extracellular enzymes, could determine the ability of glucose to prime the mineralisation of SOM. While this putative mechanism is consistent with the obtained results, it has yet to be experimentally tested. It should be noted that the hypothesized enzymatic stability of peroxidases in soil can be questioned, and their dependence on soil physicochemistry, including e.g. mineralogy and pH, will need consideration. So far, we can conclude is that microbial growth dynamics could not explain the priming of SOM aged 2-13 months in the studied soil.

44

LUCCI Annual Report – work package 1 References [1] Rousk J, Hill PW, Jones DL (2015). Priming of the decomposition of agein soil organic matter: concentration dependence and microbial control. Functional Ecology doi: 10.1111/1365-2435.12377

45

LUCCI Annual Report – work package 1

Does the fungal-to-bacterial dominance characterize the turnover of soil organic matter and nutrients? Johannes Rousk (a), & Serita D. Frey (b) (a) Microbial Ecology, Department of Biology, Lund University, Sweden (b) Department of Natural Resources & the Environment, University of New Hampshire, NH, USA

Soil biogeochemical and fungal and bacterial responses to the Detrital Input and Removal Field experiment Resolving fungal and bacterial groups within the microbial decomposer community is thought to capture disparate life strategies for soil microbial decomposers, associating bacteria with an r-selected strategy for carbon (C) and nutrient use, and fungi with a K-selected strategy. Additionally, food-web model based work has established a widely held belief that the bacterial decomposer pathway in soil supports high turnover rates of easily available substrates, while the slower fungal-dominated decomposition pathway supports the decomposition of more complex organic material, thus characterising the biogeochemistry of the ecosystem. We used a field experiment, the Detritus Input and Removal Treatments – DIRT – experiment (Harvard Forest Long-term Ecological Research site, USA) where litter and root inputs (control, no litter, double litter, or no tree roots) have been experimentally manipulated during 23 years, generating differences in soil C quality [1]. We hypothesized (i) that δ13C enrichment would decrease with higher soil C quality and that a higher C quality would favour bacterial decomposers, (ii) that the C mineralised in fungal dominated treatments would be of lower quality and also depleted in δ13C relative to bacterial dominated high quality soil C treatments, and (iii) that higher C mineralisation along with higher gross N mineralisation rates would occur in bacterial dominated treatments compared with more fungal dominated ones. The DIRT treatments resulted in a gradient of soil C-quality, as shown by up to 4.5-fold differences between the respiration per soil C between treatments (Fig. 1). High quality C benefited fungal dominance (Fig. 2), in direct contrast with our hypothesis. Further, there was no difference between the δ13CO2 produced by a fungal compared with a bacterial dominated decomposer community (Fig. 1b). There were differences in C and N mineralisation between DIRT treatments (Fig. 1), but these were not related to the relative dominance of fungal and bacterial decomposers.

Ecosystem implications of fungal-to-bacterial dominance revised To summarize and to evaluate our set of hypotheses together, we found no support for a positive relationship between the relative dominance of bacteria and high SOC-quality (H1) (Fig. 2). The SOC decomposed in relatively more fungal dominated soils is not different from that degraded in relatively more bacterial dominated systems (H2) (Fig. 2). The C and N mineralisation, and thus the potential for C and N leakage, in relatively fungal dominated systems was not lower than that occurring in relatively bacterial dominated ones; and fungal dominated systems were not better than bacterial dominated systems at retaining mineral N (H3). These results suggest structural differences between soil microbial characteristics driven by the age and quality of C input, that also carry implications of system biogeochemistry. Specifically, increased fresh plant C influx will promote a shift toward a fungal dominance of SOM turnover. Such changes could be induced by increased forest productivity (e.g. in warmer conditions) or induced by biomass harvest and subsequent re-colonisation (as indicated by the OA treatment). A shift toward fungal dominance will also lead to a faster mineralisation of C and N, and a somewhat lower microbial retention of mineral N.

46

LUCCI Annual Report – work package 1

Figure 1. The dependence of carbon (C) quality (as indicated by respiration per SOC) on the ∂13C in soil organic C (panel a) or ∂13C in respired CO2 (panel b), and the dependence of nitrogen (N) quality (as indicated by NH4+ production per SOC) on the C quality (panel c) and the ∂13C in soil organic C (panel d). The fitted relationship is a linear type II major axis regression. The O and M horizons are coded in blue and red, respectively. Values represent the mean ± 1 SE of 6 (Co) or 3 (other treatments) field replicates.

These findings contrast with widely held beliefs within soil microbial ecology. Detrital food webs dominated by bacteria are thought to support high turnover rates of easily available substrates, while slower fungal-dominated decomposition pathways support the decomposition of more complex organic material. The crude phylogenetic resolution of fungal and bacterial decomposers has been thought to capture disparate life strategies for soil decomposers, associating bacteria with an r-selected strategy for C and nutrient use, and fungi with a K-selected strategy. These assumptions have been incorporated into food-web models, where the different properties of fungi and bacteria with regard to C sequestration, nutrient turnover and ecosystem stability have been concluded. This influential work has manifested a widely held belief that bacterial decomposer pathways in soil support high turnover rates of easily available substrates, while slower fungal-dominated decomposition pathways support the decomposition of more complex organic. We note that these assumptions about the microbial processing of SOM rarely have been tested, or, when they have, either (i) the microbial use of SOM has been inferred from the colonization of plant litter (e.g. de Graaff et al. 2010) rather than SOM or (ii) biomass concentrations of fungi and bacteria have been assumed to reflect their contribution to SOM turnover; an assumption that is not met in soil.

47

LUCCI Annual Report – work package 1

Type-II major axis regression: Y = 8.1X + 0.34 2

R =0.38, P=0.034, f=6.01, df=11

300

OA

OA 2X

-1

-1

(pmol h g SOC)

Ac inc into ergosterol

(a)

T0X

200

Co

T 0X

100

T T0X 0X

2X

Co

0

(b) OA

(pmol h g SOC)

2000 T0X

-1 -1

Leu inc in extr bact

2500

1500 0X

0X OA

1000 2X

T

500

Co

2X

T Co

T0X

(log((pmol Ac)(pmol Leu) ))

-1

Fungal-to-bacterial growth ratio

0

(c) T

-0.40 T0X

Co 2X

T OA

-0.60

Co

2X

-0.80

0X

OA

T0X 0X

-1.0 -1.2 0

10

20

30

40

Figure 2. The dependence of fungal growth (panel a), bacterial growth (panel b) and the fungal-tobacteria growth ratio (panel c) on carbon (C) quality (as indicated by respiration per SOC). The fitted relationship is a linear type II major axis regression, and is only drawn when significant. The O and M horizons are coded in blue and red, respectively, and treatments are defined abbreviations. Values represent the mean ± 1 SE of 6 (Co) or 3 (other treatments) field replicates.

Respiration per SOC -1

(µg C g SOC)

We find no evidence in favour of the supposition that bacterial dominated food webs cycle faster and leak more C and N than fungal dominated ones, and convincing suggestions for the opposite, in the studied forest soil. Our results highlight the fact that many of the important dogma held as true in soil microbial ecology have limited empirical support. In one of the few empirical validations of another dogma, that fungi have a higher C-use efficiency than bacteria, previous studies have found that in fact the C-use efficiency of bacterial and fungal dominated decomposer systems could not be distinguished. Taken together with the results we report here, the lack of support for central assumptions about the properties of the soil decomposer food web can only be regarded as humbling. Assumptions about the soil decomposer community (e.g. C-use efficiency, nutrient retention, etc.) influence predictions of ecosystem models and the empirical basis for these assumptions is still weak.

References [1] Rousk J, Frey SD (2015). Revisiting the hypothesis that fungal-to-bacterial dominance characterises turnover of soil organic matter and nutrients. Ecology (in press).

48

LUCCI Annual Report – work package 1

Primary producer-release of labile carbon enhances litter degradation in aquatic experimental systems A. Margarida P. M. Soares (a), Emma S. Kritzberg (b), Johannes Rousk (a) (a) Section of Microbial Ecology, Department of Biology, Lund University, Sweden (b) Section of Aquatic Ecology, Department of Biology, Lund University, Sweden

The quality of organic matter is highly variable and covers a large gradient of resistance to degradation. Recalcitrant organic matter can be ‘activated’ and thus involved in carbon and nutrient cycling in aquatic ecosystems through the priming effect, which is the increased mineralization of recalcitrant organic matter triggered by inputs of labile organic matter. To understand the decomposition of organic matter in general, and the priming effect in particular, it is important to investigate how microbial communities can interconnect organic matter pools of different recalcitrance. Most previous assessments of priming effect have used pulse additions of single substrates at high concentrations. This may not represent a natural environment well. Therefore, to achieve an assessment of the priming effect it is important to simulate a realistic delivery of labile organic matter. One way to accomplish this is by introducing labile carbon through primary producer activity. In laboratory experiments, we used pond water (aquatic) microbial communities to which dried C4 plant litter was added [1]. We monitored the successional dynamics of fungal and bacterial growth, primary production activity, and respiration on litter under dark and light conditions. We used Photosynthetically Active Radiation (PAR) light in order to minimally impact photodegradation of organic matter by UV light. We observed an increased fungal production and abundance in light treatments. The fungal growth response coincided with an increase in algal primary productivity occurring at day 7. Dark treatments showed a low fungal growth and no primary production. Bacterial production increased rapidly in the first days but decreased with no differences between light and dark systems. We conclude that primary production can stimulate fungal growth and that the presence of labile carbon consequently can ‘prime’ the decomposition of recalcitrant organic matter.

A 18

25

L… D…

20 15 10

5 0

1

C150

Time 6(days)

11

Fungi

16

0

pmol ergosterol/litter piece

2

4

D 80

L i…

Time 6 8(days) 10

12

Bacteria

14

16

L i…

60 pmol leucine/litter/h

100

50

0 -4

Respiration

B

µg CO2/litter piece/h

µg C/ litter/h -4

30

L Primary Production i…

16 14 12 10 8 6 4 2 0

1

Time 6(days)

11

16

-4

40 20 0 1

Time 6(days)

11

16

Figure 1. Primary production (A) increases in Light treatmens coinciding with a higher respiration (B) and fungal production (C) in light systems compared to dark. This indicates an influence of primary production in decomposition by fungi. Bacteria production (D) show high numbers in the beginning but rapidly declines in both treatments.

49

LUCCI Annual Report – work package 1 References [1] Soares AMPM, Kritzberg ES, Rousk J (2015) Is litter decomposition ’primed’ by primary producer-release of labile carbon in terrestrial and aquatic experimental systems? Geophysical Research Abstracts
 17: EGU2015-8787-2

50

LUCCI Annual Report – work package 1

A comparable approach to assess variation in surface energy fluxes in northern high-latitude ecosystems Christian Stiegler (a), Anders Lindroth (a, b), Magnus Lund (a) Norbert Pirk (a), Torben R. Christensen (a, b), Mikhail Mastepanov (a) & Frans-Jan Parmentier (a) (a) Dept of Physical Geography and Ecosystem Sciences, Lund University (b) Department of Bioscience, Aarhus University

Introduction The partitioning of energy at the surface is a crucial process which exerts a major control on climatic and hydrologic regimes in northern high-latitude ecosystems [1,2]. High-latitude ecosystems are also known to be very sensitive to climate change [3,4]. With concern made to the potential effects of climate warming, knowledge of the terrestrial surface energy balance of arctic and subarctic environments is therefore important but there is a lack of direct measurements and observations. In this study we assess the variability of arctic terrestrial ecosystems in surface energy partitioning and moisture exchange. We use micrometeorological data provided by the INTERACT-network (International Network for Terrestrial Research and Monitoring in the Arctic). The aim of the study is to: (1) Determine and quantify the controlling factors of the surface energy balance of these arctic terrestrial ecosystems; (2) Examine the effects of differences in regional climate, vegetation, topography and substrate on the surface energy budget and evapotranspiration regime; (3) Assess possible changes in landatmosphere interactions caused by climate change.

Methods We have carried out field measurements of surface energy balance components and meteorological parameters from several types of tundra ecosystems. The study area covers high-arctic (Zackenberg, Northeast Greenland and Adventdalen, Svalbard) and low-arctic (Kobbefjord, West Greenland) heath and fen ecosystems. Data shown in this study represents the growing seasons during the period 2012 to 2014. The growing season was defined as the time period where the surface was free of snow cover and midday net radiation and midday air temperature were positive. Fluxes of sensible heat and latent heat were gap-filled using the moving look-up table approach [5].

Results and discussion At the high-arctic sites in Zackenberg, strong variation in the length of the snow cover season and in the snow cover thickness had a pronounced impact on the length of the growing season and on the partitioning of surface energy fluxes, see Figure 1. In 2013, the onset of the growing season occurred 33 days earlier compared to 2012 and 30 days earlier compared to 2014. In total, the growing season lasted for 94 days in 2012, 120 days in 2013 and 96 days in 2014. Consequently, the lack of meltwater in 2013 caused sensible heat fluxes to clearly dominate over latent heat fluxes at the dry heath site. Average midday Bowen ratio increased with increasing air temperature and peaked at 5.2 when the air temperature was >15°C, see Figure 1. However, in 2014 sensible heat fluxes only dominated over latent heat fluxes when the air temperature was >15°C with a Bowen ratio of 1.3. The wet fen sites in Adventdalen (Svalbard), Zackenberg (Northeast Greenland) and Kobbefjord (West Greenland) showed increased latent heat fluxes with increasing air temperature. Therefore, Bowen ratio generally decreased when air temperature increased and dropped below 1 in the highest air temperature classes. At the wet fen site in Zackenberg, e.g., latent heat fluxes exceeded the decreasing sensible heat fluxes at air temperature >15°C in 2013, with Bowen ratio of 0.9, and >10°C in 2014, with Bowen ratio of 0.8. Ground heat fluxes show similar intensities in all classes of air temperature but the importance of the ground heat fluxes to the surface energy budget was more pronounced at the high-arctic fen sites compared to the dry tundra ecosystems. In comparison to sensible and latent heat fluxes, ground heat fluxes contributed only a minor part to the surface energy budget at the southernmost dry heath site in Kobbefjord. Our results show that interannual variability of the length of the snow cover season and snow layer thickness had profound impact on the subsequent growing season at our monitored ecosystems in Zackenberg. Due to the

51

LUCCI Annual Report – work package 1 importance of the wintertime precipitation in this area, soil moisture content, partitioning of heat fluxes and heat storage in the soil differed substantially during the growing seasons. The summertime surface energy budget and partitioning of heat fluxes is closely connected to the exchange of greenhouse gases and influences the overall carbon balance of the investigated ecosystems. These ongoing measurements provide first insights to the locally determined conditions of the surface energy budget in high-latitude tundra ecosystems.

Wet fen (Adventdalen, Svalbard)

15-20

1.0

0.0 0-5

5-10

1.0

0.0

0.0

>20

0-5

5-10

15-20

>20

(b)

10 0

155

115

-10

275

255

20

135

Air temp. [°C]

DOY

10-15

Air temperature [°C]

(b)

115

275

255

235

215

195

175

155

-10

135

15-20

235

0

15 10 5 0 -5 -10

155

(b)

10

135

Air temp. [°C]

20

115

Air temp. [°C]

10-15

2.0

0.2

Air temperature [°C]

Air temperature [°C]

4.0 3.0

0.3

0.1

0.0

>20

0.4

5.0

Bowen ratio

0.2

215

10-15

2.0

195

5-10

0.3

0.1

0.0 0-5

3.0

6.0

DOY

275

0.0

0.4

0.5

255

1.0

0.1

4.0

0.5

235

0.2

0.6

215

2.0

0.7

5.0

195

0.3

H/Rnet LE/Rnet G/Rnet Bowen ratio

(a)

175

3.0

(a)

Flux ratio

0.4

0.6

175

0.5

Wet fen (Kobbefjord, West Greenland) 0.8

6.0

H/Rnet LE/Rnet G/Rnet Bowen ratio

0.7

Flux ratio

Flux ratio

0.6

2012

0.8

Bowen ratio

0.7

6.0

H/Rnet LE/Rnet 5.0 G/Rnet Bowen ratio 4.0

(a)

Bowen ratio

Dry heath (Zackenberg, Northeast Greenland) 0.8

DOY

Wet fen (Zackenberg, Northeast Greenland)

1.0

0.0 15-20

2.0

0.2 1.0

0.0 0-5

5-10

1.0

0.0

0.0 0-5

5-10

(b)

0 -10

115

275

255

235

215

>20

10

155

0

115

15-20

20

135

Air temp. [°C]

(b)

DOY

10-15

Air temperature [°C]

-10

275

255

235

215

195

175

155

2.0

0.1

>20

10

155

0 -10

20

135

(b)

10

135

15-20

4.0 3.0

0.3

Air temperature [°C]

Air temp. [°C]

20

115

Air temp. [°C]

Air temperature [°C]

10-15

0.4

5.0

0.2

0.0

>20

0.5

Bowen ratio

0.3

195

10-15

3.0

0.1

0.0 5-10

0.4

175

2013

0.1

0.6 4.0

0.5

6.0

275

2.0

0.2

H/Rnet LE/Rnet G/Rnet Bowen ratio

(a)

255

0.3

0.7

5.0

235

3.0

0.8

6.0

215

0.4

(a)

195

0.6 4.0

0.5

0-5

H/Rnet LE/Rnet G/Rnet Bowen ratio

175

0.7

Flux ratio

0.8

5.0

Flux ratio

Flux ratio

0.6

6.0 (a)

Bowen ratio

H/Rnet LE/Rnet G/Rnet Bowen ratio

0.7

Bowen ratio

0.8

DOY

DOY

Dry heath (Kobbefjord, West Greenland)

0-5

5-10

1.0

0.1 0.0

0.0 0-5

5-10

10-15

15-20

>20

0 -10

155

275

(b)

10

135

0

20

115

Air temp. [°C]

Air temperature [°C]

-10

DOY

Bowen ratio

2.0

>20

(b)

155

275

255

235

215

195

175

155

135

DOY

15-20

10

135

0 -10

20

115

Air temp. [°C]

(b)

10

115

Air temp. [°C]

20

10-15

4.0 3.0

0.3

Air temperature [°C]

Air temperature [°C]

5.0

0.2

0.0

>20

0.4

6.0

275

15-20

0.0

255

10-15

0.0

1.0

235

5-10

0.1

215

0-5

0.2

1.0

0.5

255

0.0

2.0

195

2014

0.1

0.6 4.0 3.0

0.3

H/Rnet LE/Rnet G/Rnet Bowen ratio

(a)

235

0.2

0.4

0.7

5.0

215

2.0

0.5

0.8

6.0

195

0.6 4.0 3.0

0.3

H/Rnet LE/Rnet G/Rnet Bowen ratio

(a)

175

0.4

0.7

Flux ratio

0.5

0.8

5.0

175

Flux ratio

0.6

6.0

Bowen ratio

H/Rnet LE/Rnet G/Rnet Bowen ratio

(a)

Flux ratio

0.7

Bowen ratio

0.8

DOY

Figure 1. (a) Flux ratios of growing season sensible heat (H), latent heat (LE), ground heat (G) and the Bowen ratio (H/LE) in relation to the air temperature. The error bars represent the standard error. (b) Daily average air temperature during the growing season.

52

LUCCI Annual Report – work package 1 References [1] S. Westerman et al. 2009. The annual surface energy budget of a high-arctic permafrost site on Svalbard, Norway. The Cryosphere 3, 245-263, doi:10.5194/tc-3-245-2009. [2] J. Beringer et al. 2005. Surface energy exchanges along a tundra-forest transition and feedbacks to climate. Agricultural and Forest Meteorology 131, 143-161, doi: 10.1016/j.agrformet.2005.05.006. [3] T.V. Callaghan et al. 2010. A new climate era in the sub-Arctic: Accelerating climate changes and multiple impacts. Geophysical Research Letters 37(14), L14705, doi: 10.1029/2009GL042064. [4] R. Pearson et al. 2013. Shifts in Arctic vegetation and associated feedbacks under climate change. Nature Climate Change 3, 673-677, doi: 10.1038/nclimate1858. [5] M. Reichenstein et al. 2005. On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Global Change Biology 11(9), 1424-1439, doi: 10.1111/j.1365-2486.2005.001002.x.

53

LUCCI Annual Report – work package 1

Succession of ectomycorrhizal fungi in mesh bags over a three year period in a fertilized and unfertilized Norway spruce forest Håkan Wallander, Nicholas Rosenstock Department of Biology, MEMEG, Lund University, 22362 Lund Sweden

Ingrowth mesh bags are commonly used to estimate ectomycorrhizal production in forest ecosystems [1]. A minor disturbance is caused when inserting such bags, and the bags themselves can be viewed as an empty space in the soil open to fungal colonization. Ingrowth mesh bags are usually incubated for one growing season but in this study we harvested mesh bags annually over a three year period in order to follow the succession of fungi over time. The mesh bag substrate consisted either of quartz sand or quartz sand mixed with 1% maize compost. We used three replicate plots in a Norway spruce forest in south-eastern Sweden, which were either unfertilized or fertilized with a complete fertilizer every second year to maximize forest production. Mesh bags were also placed in trenched plots where the importance of mycorrhizal fungi was reduced. At harvest subsamples was taken for fungal biomass estimates (ergosterol) [2] and DNA extraction. The DNA was sequenced after amplifying the ITS region using the 454pyrosequencing facility at the Department of Biology at Lund University. Fertilization reduced the production of fungal biomass considerably but had only minor effects on the fungal community, although the effect of fertilizer on fungal community composition increased over time. Piloderma olivaceum was the most common fungus in control bags after three years (30%) but totally absent from bags collected from fertilized plots. Sampling time had a major influence on the fungal community (Fig 1.) with many yeastlike fungi (Guheomyces, Cryptococcus, Candida) dominating early in the succession (30%), while ectomycorrhizal fungi (EMF) became much more important as the succession proceeded (80% after 3 years). The most common EMF genera after three years were Tylospora, Piloderma and Amphinema. Both maize amendment and trenching had significant effects on the fungal communities in mesh bags and are importance of different factors for shaping EMF communities in the soil.

Figure 1. Sampling time (yr 1, yr 2, yr 3) and maize compost addition have a major impact on fungal community growing into mesh bags, while Trenching also had an effect, but to a lesser extent.

54

LUCCI Annual Report – work package 1

Figure 2. Fertilization effects on fungal community grow over time and are only significant 3 years after fertilization. Effects of maize amendment and trenching diminish over time.

References [1] Wallander H. Nilsson L-O. Hagerberg D. Bååth E. (2001) Estimation of the biomass and Seasonal Growth of External Mycelium of Ectomycorrhizal Fungi in the Field New Phytologist 151: 753-760 [2] Wallander H, Ekblad A, Bergh J. (2011) Growth and carbon sequestration by ectomycorrhizal fungi in intensively fertilized Norway spruce forests. ForestEcology and Management 262: 999-1007.

55

LUCCI Annual Report – work package 1

Seasonal variation of BVOC emissions from Norway spruce Min Wang (a), Guy Schurgers (b), Anna Ekberg (c), Ylva Persson (a) & Thomas Holst (a) (a) Dept of Physical Geography and Ecosystem Sciences, Lund University (b) Dept of Geoscience and Natural Resource Management, Copenhagen University (c) Centre for Environmental and Climate research, Lund University

Introduction Biogenic volatile organic compounds (BVOCs) are a diverse group of reactive compounds produced by more than 90 plants families, which are involved in plant physiological processes, reproduction and self-defense [1-3]. The estimated averaged global annual BVOC emission was 760 Tg (C) yr-1 in the past 30 years [4]. Temperature and light are the main environmental controls on synthesis and emissions of BVOCs [5-7]. Norway spruce (Picea abies) is one of the dominant conifer species widely distributed in northern and central Europe [8]. It emits different terpenes, such as isoprene (C5H8), monoterpenes (MT, a group of compounds with a common chemical formula, C10H16), and sesquiterpene (SQT, C15H24), where MTs are the dominant terpenes released from spruce [9-11]. There are limited studies about characterizing MT and SQT emissions from Norway spruce. Here we measured BVOC emission rates from a Norway spruce in central Sweden over a period of 4 months to study the seasonal development of these emissions.

Study site and methods The study site is a boreal forest located at the ICOS (Integrated Carbon Observation System) site Norunda (60°05'N, 17°29'E), Sweden. This boreal forest is dominated by Norway spruce (Picea abies) and Scots pine (Pinus sylvestris) with a canopy height of about 25 m. 5 campaigns were carried out from June to September in 2013. All the measurements were conducted on a 120 years old Norway spruce at 20 m height. The branches were enclosed in a cylindrical, transparent Teflon chamber with a constant inflow of VOC-free air. The air samples were collected hourly from the chamber with Tenax-TA adsorbent tubes for 30 minutes. The samples were analyzed by gas chromatography and mass selective detector (GC-MS) to quantify each trapped terpene. The emission rates of terpenes were calculated based on the dry weight (dw) of needles enclosed in the chamber.

Results and discussion

Distribution of measured BVOC emission rates Total measured terpene emissions were in the range of 0.05 to 332.5 µg g-1dw h-1 from June to September 2013. Isoprene, MT and SQT had their peak emissions in August, and the highest MT emission was around 3 times larger than the peak of isoprene or SQT emission (Fig. 1b-d). High isoprene emissions were found at the end of July and August with an average of 13.3 µg g-1dw h-1.The MT emission rates here were much higher than other studies. A previous study of a 50-year old Norway spruce growing at the same site reported that MT emissions were in the range of 0.1 to 5.7 µg g-1dw h-1 [10], which in comparison was only 3-4% of our measurements. SQT emissions were comparable with isoprene emissions in July and August campaigns, and SQT had much higher emissions in early June when isoprene was not detected in most of samples (Fig. 1b&d).

56

LUCCI Annual Report – work package 1

Figure 1. a. Measured temperature inside the chamber and photosynthetic active radiation (PAR) next to chamber. b-d. Measured emission rate of isoprene, MT and SQT respectively for each campaign in 2013.

Variation of emission composition The composition of emission spectra changed slightly between months. The fraction of isoprene of total terpenes was below 1% in June, increased to 3% in the beginning of July and reached over 10% in the end of July, August and September (Fig. 2). The dominant MTs were α-pinene, β-pinene, limonene, and camphene, which collectively contributed 48% to 64% of the total terpene emission during summer time (Fig. 2). β-farnesene was the main SQT from June to August and its fraction went down from 25% to 7% with time. The dominant SQTs were α-farnesene and β-farnesene, which are common induced compounds for self-defense against insects attack. (E)-β-farnesene was reported to be one of the most significantly induced emissions from spruce after attacked by beetles [12]. The spruce in this study probably had an insect infestation before the measurements.

57

LUCCI Annual Report – work package 1

Figure 2 Composition of emission spectra for each campaign from June to September based on averaged emission rate.

References [1] Holopainen, J.K., Can forest trees compensate for stress-generated growth losses by induced production of volatile compounds? Tree Physiology, 2011. 31: p. 1356-1377. [2] Loreto, F. and J.-P. Schnitzler, Abiotic stresses and induced BVOCs. Trends in Plant Science, 2010. 15(3): p. 154-166. [3] Penuelas, J. and M. Staudt, BVOCs and global change. Trends in Plant Sciene, 2009. 15(3). [4] Sindelarova, K., et al., Global data set of biogenic VOC emissions calculated by the MEGAN model over the last 30 years. Atmospheric Chemistry and Physics, 2014. 14(17): p. 9317-9341. [5] Monson, R.K., et al., Relationships among Isoprene Emission Rate, Photosynthesis, and Isoprene Synthase Activity as Influenced by Temperature. Plant Physiology, 1992(98): p. 1175-1180. [6] Laothawornkitkul, J., et al., Biogenic volatile organic compounds in the Earth system: Tansley review. New Phytologist, 2009. 183(1): p. 27-51. [7] Llusià, J., et al., Seasonal contrasting changes of foliar concentrations of terpenes and other volatile organic compound in four dominant species of a Mediterranean shrubland submitted to a field experimental drought and warming. Physiologia Plantarum, 2006. 127(4): p. 632-649. [8] Grabmer, W., et al., VOC emissions from Norway spruce (Picea abies L. [Karst]) twigs in the field—Results of a dynamic enclosure study. Atmospheric Environment, 2006. 40(Supplement 1): p. 128-137. [9] Janson, R.W., Monoterpene emissions from Scots pine and Norwegian spruce. Journal of Geophysical Research, 1993. 98(D2): p. 2839-2850. [10] Janson, R., C. De Serves, and R. Romero, Emission of isoprene and carbonyl compounds from a boreal forest and wetland in Sweden. Agricultural and Forest Meteorology, 1999. 98-99: p. 671-681. [11] Hakola, H., et al., The ambient concentrations of biogenic hydrocarbons at a northern European, boreal site. Atmospheric Environment, 2000. 34: p. 4971-4982. [12] Blande, J.D., K. Turunen, and J.K. Holopainen, Pine weevil feeding on Norway spruce bark has a stronger impact on needle VOC emissions than enhanced ultraviolet-B radiation. Environmental Pollution, 2009. 157(1): p. 174-180.

58

LUCCI Annual Report – work package 1

Hygroscopicity of photochemically processed particles from three different sources Cerina Wittbom (a), Joakim Pagels (b), Jenny Rissler (b), Axel Eriksson (a), Erik Nordin (b), Patrik Nilsson (b), Erik Swietlicki (a) & Birgitta Svenningsson (a) (a) Dept of Nuclear Physics, Lund University (b) Ergonomics and Aerosol Technology, Lund University

Introduction Anthropogenic emissions from vehicles can affect both human health and the climate. Freshly emitted soot particles (particles containing black carbon, BC) are known to have a limited water vapour uptake [1]. However, when residing in the atmosphere the soot particles undergo chemical and physical processes, resulting in condensation of inorganic and organic material onto the particles also causing a restructuring of the soot aggregate structure. These processes will alter the water vapour uptake of the soot particles, and hence also the life cycle of the particles in the atmosphere as well as the deposition in the human respiratory tract. The water uptake depence on the abundance of water vapour in the surrounding air, i.e. the relative humidity (RH).

Theory The hygroscopicity of the particles in a subsaturated environment (RH <100%) can be described by the hygroscopic growth factor (HGF), defined as:

Here, d(RH) and d0 is the mobility diameter of the humidified and dry particle, respectively. The particle water uptake or the evaporation of water, in the water vapour subsaturated environment, continuous until the water activity of the particle is in equilibrium with the surroundings. Another way of describing the hygroscopicity of the particles is to use a single parameter approach, i.e. to calculate the κ–value [2]. This parameter describes the number of ions or non-dissociating molecules per unit volume of the dry particle, and can be derived either from measured HGF (κHGF):

where σw, Mw and ρw are the surface tension, molecular mass and density of water, respectively. In the soot experiments, the diameter of the dry particle (dp) is represented by the volume equivalent diameter, to account for the non-sphericity of the agglomerates, otherwise dp=d0. The κ–value can also be calculated with knowledge of the particle chemical composition (κSOA), using:

Here Mi and ρi are the molecular mass and density of a specific component, and ii is the van’t Hoff factor for that component. The ii represents the dissociation and ion interactions. The input parameters for the κSOA are independently derived, describing the organic material coating the particles. A detailed description can be found in [1].

59

LUCCI Annual Report – work package 1 Another way of deriving the κ–value is from measurements of the particles in a supersaturated condition (RH>100%) in combination with knowledge of the dry particle diameter:

σsol is the surface tension of the droplet solution and sc is the critical supersaturation at which the particle is assumed to activate into a cloud droplet.

Measurements In this study laboratory measurements of the HGF of photochemically processed soot have been performed using a Hygroscopic Tandem Differentail Mobility Analyzer (H-TDMA) [3], in the LU smog chamber. For measurements of the particles in the supersaturated regim, a Cloud Condensation Nuclei Counter (CCNC-100, DMT) has been used. To determine the composition of the cores and coatings a soot particle aerosol mass spectrometer (SP-AMS, Aerodyne research) was used. In addition an Aerosol particle Mass Analyzer was placed after a Differential Mobility Analyzer (DMA-APM, Kanomax Japan 3600) [4] for characterization of the particle mass-mobility relationship. The experiments are described in detail elsewhere [1][5]. The three primary sources used are: (1) a Diesel passenger Vehicle (Euro II), (2) a Flame Soot generator, and (3) three gasoline-powered passenger Vehicles (Euro II, III, IV). The primary aerosol was diluted before entering the smogchamber. In the soot experiments ((1) - DEP & (2) - FSP) the aerosol was then exposed to selected amounts of the anthropogenic secondary organic aerosol (SOA) precursors toluene and m-xylene, and the NO concentration was titrated down to about 50 ppb by adding O3, before the onset of UV. In the gasoline experiments ((3) – I) ammonium sulfate (AS) were injected to a concentration of 15-20 µgm-3 and utilized as condensation seeds for the low vapour pressure reaction products formed from the vehicles.

Results No hygroscopic growth of the freshly emitted diesel soot praticles was seen, while for the gasoline experiments, the growth of the AS seeds corresponds to that of AS. As the photochemical processing proceeds in the smog chamber, volatile organic compounds (VOCs) condense onto the particles (both the DEP and FSP (1 &2), as well as the AS seeds (3)) producing secondary organic aerosol (SOA). This condensation alters the chemical properties and the ability to take up water. For the DEP and FSP a restructuring of the soot agglomerates is seen for the dry particles. In addition, an increased RH enhances the morphology change, even for low organic content (organic mass fraction > 0.1). The acquisition of organic material continuous as the photochemical process proceeds, with more spherical like dry soot particles as a consequence. Results show a size dependency, according to the mobility diameter (dm). For the gasoline experiments, results show a decreasing HGF with increasing organic mass, as expected. Preliminary results of the κ–values derived from the three different approaches (subsaturated (κHGF) and supersaturated (κCCN) conditions, as well as the chemical composition (κSOA)) show differing results, Figure 1 (only showing results from the soot experiments). Comparison is only performed for high organic content (organic fraction > 0.7), when the dry particles have restructured to a more compact shape (important for the DEP and FSP) and to get a close to none-interference from the AS-component (important for the I experiments). The discrepancies might be due to small differences in the chemical composition, e.g. how oxidized the material is, which could then explain the higher κHGF for similar κSOA (Figure 1 a). In that case κSOA might not be a representative parameter for the particle hygroscopicity. And the water vapour uptake might be different in sub- and supersaturated conditions, depending on the organic material, hence differences in κHGF and κCCN. Another possibility is that the shape (none-sphericity) of the particles influences the results when deriving κ–values from the measured HGF and sc. Also, there are some instrumentational limitations as can be seen in Figure 1 b (black star), where the results from two CCNC show differences in κCCN for the same κHGF. This is probably due to measurements performed in the outer ranges of the different super saturation spectra. κHGF from the gasoline experiments show higher values (up to κHGF=0.43), probably due to interference from the AS seed.

60

LUCCI Annual Report – work package 1

Figure 1. Comparison of κ–values (a) derived from HGF (κHGF) and calculated κ–values using the chemical components in the particles (κSOA), and (b) between κHGF and κCCN derived from measurements of the critical super saturation, for different dry mobility diameters (identified by markers). Note, the results are preliminary and only include results from the DEP (black) and FSP (turquoise) experiments.

References [1] Wittbom et al., Atmos. Chem. Phys., 14, 9831-9854, 2014. [2] Petter and Kreidenweis, Atmos. Chem. Phys., 7, 1961-1971, 2007. [3] Nilsson et al., Atmos. Meas. Tech., 2, 313-318, 2009. [4] McMurry et al., Aerosol Sci. and Technol. 36: 227-238, 2002. [5] Nordin et al., Atmos. Chem. Phys., 13, 6101-6116, 2013.

61

LUCCI Annual Report – work package 1

62

WORK PACKAGE 2

Climate and carbon cycle variability during historical times and calibration of proxy records to instrumental data

LUCCI Annual Report – work package 2

64

LUCCI Annual Report – work package 2

WP 2: Progress Report Lena Ström & Torben R. Christensen

Report on activities that have seen partial WP2 funding 2014. In general, 2014 has seen continued developments in the LUCCI WP2 area. Joint with WP1 and WP3 the Nordic Center of Excellence, DEFROST, was continued with its now 14 employed PhD and post doc fellows. DEFROST is ending in early 2016 and several PhDs in the project are approaching their finish. The DEFROST activities are all concerned with various aspects of the LUCCI agenda and many are directly feeding into joint outputs. Further collaborative projects include EU-INTERACT that has enabled us to increase in particular our coverage of energy exchange at our measurement sites and funded parts of PhD student Julie Maria Falk’s field-work. Julie Maria Falk successfully finished and defended her PhD in September 2014. Also, the EU project PAGE21 has provided LUCCI WP2 researchers with a new advanced 13CH4 analyzer for flux measurements that has been deployed at Zackenberg to expand on our understanding of temporal dynamics in shifting sources of CH4. Experiments with this analyzer was continued in 2014 and currently the data is being analyzed. We use part of the WP2 funding for improving on efficiency and increasing the speed of operational analysis in the laboratories. The progress on this front has continued becoming more streamlined in 2014 largely through the work of our LUCCI funded permanent laboratory technician, Marcin Jackovicz-Korczynski. We have completed a DEFROST based collaboration with University of Eastern Finland where we incubated permafrost monoliths and manipulate the active layer depth in a controlled environment study. This experiment ran from March through November 2013 and a wealth of data and a series of papers are now being written as part of two joint PhD programs based at University of Eastern Finland.

From our field operations we have the following brief reports: Zackenberg Monitoring of the CO2 and CH4 exchange in a high Arctic fen site (Rylekærene) in the Zackenberg valley, northeast Greenland, was continued in 2014. In the summer 2010 the field setup for the project Plant-soil-Herbivore interactions in the Arctic – Feedback to the carbon cycle, were established. 5 blocks each with an exclosure, to prevent muskoxen from grazing and trampling, a snow-fence and a control were sat up in the Zackenberg valley in northeast Greenland in collaboration with Aarhus University, National Environmental Research Institute (NERI), Department of Arctic Environment, Denmark. Carbon dioxide, methane and nitrous oxide are all monitored inside and outside the exclosures and the planned measurements were completed successfully in 2014. The measurements initiated as part of the above mentioned PhD project by Julie Maria Falk has been continued in 2014. Also in 2014 we conducted a special campaign with a mobile eddy covariance tower looking at the carbon exchanges in the Zackenberg river at the outlet towards Young Sund. This is work connecting between our long time-series of data from the terrestrial environments with similar work on the marine side in the near coastal environments.

Nuuk Monitoring of the CO2 and CH4 exchange in a low Arctic fen site (Kobbefjord) close to Nuuk, the capitol of Greenland, was continued in 2014. During 2013 we have also expanded our measurements in the Kobbefjord area to include a heath site with INTERACT energy exchange and CO2 flux measurements. Some problems with power supply at this site was solved in 2013 through the installation of a windmill. This has ensured a much better data series from the 2014 summer season. Our post-doc Frans-Jan Parmentier out of the GREENCYCLES-II project is working with large scale issues of how seaice and oceans carry and have interactions with the terrestrial realms. A major international meeting on this topic was

65

LUCCI Annual Report – work package 2 organized by us in Nuuk, September 2011. The outcome of this was published in a Nature Climate Change review paper published in 2013.

Svalbard Our work with the LUCCI related GREENCYCLES II with flux measurements at Adventdalen has been continued primarily with support from a recently started Nordic Center of Excellence, eSTICC. This includes a collaboration that has been established with UNIS University Svalbard. There, a study site was instrumented in spring-summer 2011. This site is located in Adventdalen has been subject to Ice-wedge polygon Studies in the past. The experimental design used in the Zackenberg valley in Greenland (74°30’N, 21°00’ W) has been reproduced: automatic chambers and a gradient tower has been installed to report carbon dioxide exchanges and methane emissions. Also an INTERACT energy exchange tower has been installed. The site has seen the addition of a wind mill and the first years of longer time series with data are very promising and will be treated and communicated in the near future. A PhD student, Norbert Pirk, has been responsible for this site as of 1 January 2014. He is moving very fast and two papers on the basis of the data from this site (and other sites) have been submitted.

Fäjemyr Year-round gas flux measurements using eddy covariance (CO2) and automated chamber techniques (CH4 and CO2) in a temperate peatland site in southern Sweden (Fäjemyr) were continued during 2014. We obtained an infrastructure grant from the Faculty that has enabled us to replace a number of old and worn out instruments on the mire. This was implemented during the autumn of 2013. A paleo project studying the peat stratigraphy in relation to modern day carbon fluxes in collaboration with WP3 (Mats Rundgren and Nathalie van der Putten) was also initiated in the autumn of 2013 and continued in 2014.

Chosen literature from WP2, accepted for publication in 2014

[1] Christensen, T.R. 2014. Climate Science: Understand Arctic methane variability. Nature, 509, 279-281. [2] Bosiö J., C. Stiegler, M. Johansson, H. N. Mbufong, T. R. Christensen. 2014. Increased photosynthesis compensates for shorter growing season in subarctic tundra—8 years of snow accumulation manipulations. Climatic Change 127:321–334 DOI 10.1007/s10584-014-1247-4. [3] Mbufong, H. N., Lund, M., Aurela, M., Christensen, T. R., Eugster, W., Friborg, T., ... & Tamstorf, M. P. (2014). Assessing the spatial variability in peak season CO 2 exchange characteristics across the Arctic tundra using a light response curve parameterization. Biogeosciences, 11(17), 4897-4912. [4] Petrescu, A. M. R., A. Lohila, J.-P. Tuovinen, D. Baldocchi, A. R. Desai, N. T. Roulet, T. Vesala, A. J. Dolman, W. C. Oechel, B. Marcolla, T. Friborg, J. Rinne, J. H. Matthes, L. Merbold, A. Meijide, G. Kiely, M. Sottocornola, T. Sachs, D. Zona, A. Varlagin, D. Y. F. Lai, E. Veenendaal, F.-J. W. Parmentier, U. Skiba, M. Lund,, A. Hensen, J. van Huissteden, L. B. Flanagan, N. Shurpali, T. Grünwald, E. Humphreys, M. Jackowicz-Korczynski, M. Aurela, T. Laurila, C. Grüning, C. A. R. Corradi, A. P. Schrier-Uijl, T. R. Christensen, M. P. Tamstorf, M. Mastepanov, P. J. Martikainen, S. B. Verma, C. Bernhofer & A. Cescatti (2015) The uncertain climate footprint of wetlands under human pressure. Proceedings of the National Academy of Science, 201416267. [5] Ström, L., J. M. Falk, K. Skov, M. Jackowicz-Korczynski, M. Mastepanov, T. R. Christensen, M. Lund & N. M. Schmidt (2015) Controls of spatial and temporal variability in CH4 flux in a high arctic fen over three years. Biogeochemistry, in press. [6] Tang, J; Miller, PA; Persson, A; Olefeldt, D; Pilesjö, P; Heliasz, M; Jackowicz-Korczynski, M; Yang, Z; Smith, B; Callaghan, TV, Christensen, T.R. (2015) Carbon budget estimation of a subarctic catchment using a dynamic ecosystem model at high spatial resolution. Biogeosciences Discussions,12,2,933-980. Copernicus GmbH.

66

LUCCI Annual Report – work package 2

A Modern Automatic Chamber Technique As A Powerful Tool For CH4 And CO2 Flux Monitoring. M. Mastepanov (a,b) and T.R. Christensen (a,b). (a) Department of Earth and Ecosystem Sciences, Lund University, Sweden (b) Arctic Research Centre, Aarhus University, Denmark

A number of similar systems were used for monitoring of CH4 and CO2 exchange by the automatic chamber method in a range of different ecosystems. The measurements were carried out in northern Sweden (mountain birch forest near Abisko, 68°22´N 18°48´E, CO2, 2004-2010), southern Sweden (forest bog Fäjemyr near Hässleholm, 56°15´N 13°33´E, CH4 and CO2, 2007-2014), northeastern Greenland (arctic fen in Zackenberg valley, 74°30´N 21°00´W, CH4 and CO2, 2005-2014), southwestern Greenland (fen in Kobbefjord drainage basin near Nuuk, 64°08´N 51°23´W, CH4 and CO2, 2007-2014), central Svalbard (arctic fen in Adventdalen valley near Longyearbyen, 78°13´N 15°45´E, CH4 and CO2, 2011-2014). Those in total 37 seasons of measurements delivered not only a large amount of valuable flux data, including a few novel findings (Mastepanov et al., Nature, 2008; Mastepanov et al., Biogeosciences, 2013), but also valuable experience with the implementation of the automatic chamber technique using modern analytical instruments and computer technologies. A range of high resolution CH4 analysers (DLT-100 fast methane analyzer, RMT-200 fast methane analyzer, FGGA fast greenhouse gas analyzer – Los Gatos Research, USA), CO2 analyzers (EGM-4, SBA-4 – PP Systems, UK; Li-820 – Li-Cor Biosciences, USA; FGGA fast greenhouse gas analyzer – Los Gatos Research, USA) and specific gas analyzers (photoacoustic multigas analyzer Innova 1312 – Innova Airtech Instruments, Denmark; Methane Carbon Isotope Analyzer – Los Gatos Research, USA) has shown to be suitable for precise measurements of fluxes, from as low as 0.1 mg CH4 m-1 d-1 (wintertime measurements at Zackenberg, unpublished) to as high as 2.4 g CH4 m-1 d-1 (maximum of autumn burst 2007 at Zackenberg, Mastepanov et al., Nature, 2008). Some of these instruments had to be customized to accommodate 24/7 operation in harsh arctic conditions. In this presentation we will explain some of these customizations. High frequency of concentration measurements (1 Hz in most cases) provides a unique opportunity for quality control of flux calculations; on the other hand, this enormous amount of data can be analyzed only using highly automated algorithms. A specialized software package was developed and improved through the years of measurements and data processing. This software automates the data flow from raw concentration data of different instruments and sensors, automatic and manual status records, through a single database with all recorded parameters, through to the visualized flux calculation module, which suggests the optimized flux calculation while allowing for manual correction of all parameters. In this presentation we will communicate the most recent versions of this software package and demonstrate it with different kinds of sample data.

67

LUCCI Annual Report – work package 2

High Arctic tundra greenhouse gas emissions: new insights from Adventdalen, Svalbard Norbert Pirk (a) (a) Department of Earth and Ecosystem Sciences, Lund University, Sweden,

Data on greenhouse gas (GHG) exchange in high Arctic environments is rare but badly needed, as the environment in these places will likely respond early and most rapidly to climate warming. Also, the high Arctic holds carbon stocks in permafrost that is most likely to be made vulnerable to decomposition as ecosystems get warmer. In 2011 we established a long-term GHG flux-monitoring site in the Adventdalen valley on the Svalbard archipelago. The site is characterised by little precipitation (about 200 mm/year), a strong marine influence (summer air temperatures predominantly around 5 degrees Celsius), and a continuous permafrost ground with low centered ice wedge polygons at the surface. The measurement equipment features an eddy-covariance tower and an automatic chamber system, both set up in accordance with already existing monitoring programs (INTERACT, ICOS and the Greenland Ecosystem Monitoring program). Here, we present the first results of this multi-year campaign, such as the pattern of carbon dioxide and methane fluxes during the growing and shoulder seasons, including the recently discovered autumn burst, as well as carbon dioxide fluxes during wintertime. Carbon dioxide fluxes measured with eddy-covariance agree well with the automatic chamber data, and indicate that the site acts as a sink for carbon dioxide. Methane fluxes measured by the automatic chamber system indicate a significant methane source, which is, however, weaker than at comparable sites in Greenland. There is a high degree of inter-annual variation in methane emissions, which seems to be driven by the amount of precipitation in summer. On top of that, there is a strong spatial variability resembling the polygonal ground pattern. This variability is also reflected in the occurrence of the methane autumn burst, which could be detected and is spatially and temporally distributed at a few locations and years.

68

LUCCI Annual Report – work package 2

Controls of spatial and temporal variability in CH4 flux in a high arctic fen over three years Lena Ström (a), Julie Maria Falk (a), Kristine Skov (b), Marcin Jackowicz-Korczynski (a), Mikhail Mastepanov (a), Magnus Lund (a, b), Torben Røjle Christensen (a, b) , and Niels Martin Schmidt (b) (a) Department of Physical Geography and Ecosystem Science, Lund University, Sölvegatan 12 S-223 62 Lund, Sweden (b) Arctic Research Centre, Department of Bioscience, Aarhus University, Frederiksborgvej 399, 4000 Roskilde, Denmark

*Correspondence author. E-mail: [email protected] Telephone: +46 46 2223746

Arctic wetlands are an important source of CH4 to the atmosphere and store large amounts of carbon as peat. The aim of this study was to determine the main driving forces of the spatial variability in CH4 flux in a high arctic fen situated in Zackenberg, NE Greenland. The study was conducted over three years (2011-2013) and included 16-20 control plots and in 2013 also 29 “treated” plots, e.g., excluded muskoxen grazing (10), snow fence (9) and automated chamber (10). The plots were distributed over a 0.25 km2 area. From 1 July to 10 August we measured (all variables only in 2013) fluxes of CH4 and CO2 (NEE, Reco and GPP), temperature, water table and active layer depth, PAR, substrate conc. for CH4 production in pore-water (i.e. acetate, AA) and the species composition and density of sedges. We found significant treatment effects, a 1.8-times difference in CH4 flux between the years and a high spatial variability, e.g., 9- and 35-times difference between max and min plots depending on year and treatment. GPP was consistently a strong driver of the variability in CH4 flux. In 2013 several plant productivity related variables were singled out as the strongest drivers of the variability in CH4 flux, e.g., (in order of strength) NEE, GPP and AA. The driver of NEE, GPP and AA was the density of Eriophorum scheuchzeri. The drivers were the same and followed the same patterns irrespective of treatment. In conclusion, the results show a strong dependence of the spatial variability in CH4 flux on productivity and carbon input to vegetation and pore-water. The main driver of this input is the vegetation composition and density. The results indicate that future environmental changes in wet arctic ecosystems that affect the vegetation composition and productivity will have large impacts on their carbon balance and CH4 flux, irrespective of whether these changes are driven directly by climate change or by indirect effects on for instance grazing pressure.

69

WORK PACKAGE 3

Late Quaternary climate and carbon cycle variability

LUCCI Annual Report – work package 3

72

LUCCI Annual Report – work package 3

WP 3: Progress Report Helena Filipsson & Raimund Muscheler (a) Dept of Geology, Lund University

Scientists involved: Helena L. Filipsson and Raimund Muscheler (co-leaders), Helena Alexandersson, Belinda Alvarez, Ivar Benediktsson, Svante Björck, Anna Broström, Wim Clymans, Daniel Conley, Markus Czymzik, Carolina Funkey, Dan Hammarlund, Sofia Holmgren, Nadine Krupinski, Inga Labuhn, Hans Linderson, Karl Ljung, Anne Birgitte Nielsen, Mats Rundgren, Petra Schoon, Jesper Sjolte, Göran Skog, Johanna Stadmark, Nathalie Van der Putten.

Ph.D. students involved: Florian Adolphi, Hanna Alfredsson, Martin Bernhardson, Laurie Charrieau, Guillaume Fontorbe, Patrick Frings, Anton Hansson, Thanh Le, Conny Lenz, Bryan Lougheed, Claire McKay, Florian Mekhaldi, Anette Mellström, Wenxin Ning, Maja Reinholdsson, Thorbjorg Sigfusdottir

Technical personnel involved: Git Klintvik Ahlberg, Mattias Olsson, Åsa Wallin

Apart from the administrative support to LUCCI, our work package has financed research infrastructure (part time technical staff), part-time post-doc research (Nathalie van der Putten) and two PhD positions (Florian Adolphi and Claire McKay) during the last year. WP3 has also financed a number of smaller research projects. Several of the Department’s laboratories are used by the WP3 scientists, but the Radiocarbon Dating Laboratory holds a special position in relation to the WP3 research efforts: it gives detailed chronological information as well as important clues regarding past carbon cycle dynamics. The research in work package 3 focuses on the last glacial-interglacial cycle, but we also work with older Pleistocene time scales as well as modern environmental changes in the carbon cycle. We utilize a diverse set of archives, e.g. lake and marine sediments, peat, tree rings and ice cores, and an array of different methodologies to investigate interactions between climate variability and carbon cycling. Some of the projects were initiated at or even before the start of the LUCCI project, but are still ongoing, some have evolved into new exciting directions and some are totally new. The last progress in the Southern Ocean part of the Atlantis project concerns a recent submission of a manuscript on the development of the zonal marine and atmospheric circulation in the central South Atlantic (37°S) at Nightingale Island/Tristan da Cunha during the Last Termination, 16.3-11.0 ka BP [1], and a new Aeolian (ASI) data set from Isla de los Estados prolonging the Southern Hemisphere Westerlies (SHW) record in Tierra del Fuego back to 16.3 ka BP [2]. In the former study, a combination of a large set of proxies from Nightingale Island (NI) and climate simulations of a northern hemisphere fresh-water forcing at 17 ka BP shows that the ITCZ may have moved far south during this period of the Heinrich 1 event (H1). The study also shows several other significant meridional shifts of the zonal circulation belts during the Last Termination, all of great importance for the global carbon cycle and the CO2 outgassing/ventilation from the Southern Ocean, and especially the South Atlantic. Apart from the H1 shift, the most southerly change of the circulation belts occurs in early to mid-Younger Dryas, giving clear evidence for the fact that the bipolar seesaw effect reached at least as far north as 37°S in the South Atlantic and with a time lag between the north and the south of a few hundred years, as also recently shown by the WAIS Divide Project Members [3], indicating a northern oceanic “push” for the bipolar seesaw effects. The new ASI record from Tierra del Fuego (55°S) displays high Aeolian variability between 16.3-14.3 ka BP, which, if combined with the NI record, imply that the SHW may have varied considerably in its latitudinal extent. This indicates an expanding and contracting wind belt probably in phase with changes of the oceanic fronts, during a period when CO2 was rising noticeably but with a few clear set backs [4]. Intriguing questions are: Are all the zonal circulation shifts in the south (delayed?) responses to freshwater forcings in the north and are the major changes of the southern circulation belt reflected by changes in atmospheric CO2 content? In the Indian sector of the Southern Ocean a second field campaign has been completed within the PALATIO-program (Palaeoclimate Latitudinal Transect in the Indian Ocean, IPEV, France) in November-December 2014. Field work was

73

LUCCI Annual Report – work package 3 conducted on Ile Amsterdam, a volcanic island at 37° S, with at high altitudes (c. 800 m asl) Sphagnum-dominated peat bogs. Two peat sequences of about 5.3 m have been sampled together with 100 surface samples for testate amoebae analysis and an extensive plant reference collection for future pollen and macrofossil analyses. The aim of PALATIO is to reconstruct past climate variability with a focus on potential intensification/latitudinal shifts of the southern Hemisphere westerly belt (SHW) in the Indian Ocean for the Holocene period. To do so we now obtained peat records from Iles Kerguelen, Iles Crozet and Ile Amsterdam at 49°, 46° and 37° S, respectively, covering a meridional transect from the core to the northern edge of the modern SHW. Researchers from LU (N. Van der Putten, S. Björck) as well as from LSCE (France, E. Michel, B. Klinck) participated in the field work. Secondly, a study on the Last Glacial-Interglacial Transition of Iles Kerguelen (49°S), located in the core of the SHW, is completed [5]. The onset of peat growth at the Estacade site c. 16 kyr BP coincides with the post-LGM warming in Antarctica, which already started c. 18 kyr BP. At 13.6 kyr BP, a shift to very humid and windy conditions occurred as a result of intensified SHW influence on the archipelago. The timing coincides with the onset of the so-called Oceanic Cold Reversal (OCR), based on the deuterium excess data in the EPICA Dome C ice core record [6]. Kerguelen Islands are located in the moisture source area of Dome C and a change in atmospheric circulation at that time could explain both records. Around 12.9 kyr BP, at the end of the Antarctic Cold Reversal (ACR), the proxies suggest slightly drier, less windy and probably warmer conditions. Kerguelen Islands became less influenced by the SHW and these conditions were amplified during the early Holocene climate optimum as found in Antarctic ice core records. A conceptual model has been presented by Toggweiler et al. [7] suggesting a pole-ward shift of the SHW from its equator-ward LGM position, favouring ventilation of CO2 to the atmosphere at the Antarctic divergence. Our results support the Toggweiler hypothesis with a position south of Kerguelen until 13.6 kyr BP, creating the relatively warm and dry conditions for peat growth; an equator-ward shift of the SHW at the onset of the OCR with Kerguelen situated in the core of the wind belt and subsequently, a southward shift bringing less intense SHW influence to the archipelago at end of the ACR culminating in the early Holocene climate optimum. Our quantitative landscape reconstruction group has published new pollen-based quantitative reconstructions of vegetation composition changes and landscape openness during the Holocene at regional scales for Blekinge in SE Sweden [8] and the Transylvanian Plain (NW Romania) [9], and on more local scale for the last Millenium in Estonia [10]. We have also contributed to European scale land cover reconstructions within the VR funded LANDCLIM project [11,12,13]. In collaboration with colleagues from China, we have published new estimates of pollen productivity, fall speed and pollen source area for major tree taxa in northeastern China [14]. These factors are of key importance for applying the models developed for quantitative pollen based land cover reconstructions also outside a European context. Within a project based on high-resolution stratigraphic analyses of lake sediments we have focused on the recent increase in dissolved organic carbon content of lake waters in the framework of natural and anthropogenic forcings during the last 800 years in southern Sweden [15]. This lead to increased insights into the role of land-use dynamics (changes in forest cover and agricultural land-use intensity) in interaction with sulfur deposition and climate variations for humic matter export to lakes. Another ongoing project focuses on the aeolian record in Sweden and Norway (a largely unused palaeoenvironmental resource) that can be deciphered to reveal information on humidity, forest fires, land use and storminess (including wind directions). In this project we are studying aeolian deposits, particularly inland dunes, in south and central Sweden and Norway. Most of the dunes were formed right after deglaciation but we also find evidence of younger, Holocene sand drift events that may be related to climatic change or human impact [16,17]. Sampling of submerged subfossil wood and lacustrine deposits is carried out in an ongoing PhD project (Anton Hansson), which focusses on increased understanding of the Baltic shore-displacement history, early Holocene climate dynamics and human interaction with the coastal landscape of south-eastern Sweden. Preliminary results of this project will be published next year as part of a EU Cost Action proceedings volume, and additional data are currently being prepared for publication. The project relies partly on dendroclimatic analysis of pine tree-ring series carried out at the Swedish National Laboratory for Wood Anatomy and Dendrochronology, which is hosted at the Department of Geology. As part of previous work at the Dendro Laboratory, a study of growth responses of South Swedish peatland trees to Holocene climate variability has been published [18]. Our involvement in peat stratigraphic analyses in northernmost Sweden has also contributed to a major synthesis of carbon and nitrogen accumulation rates in

74

LUCCI Annual Report – work package 3 peatland soils worldwide [19]. A study of late Holocene peatland expansion in southern Sweden and associated impacts on peatland forest dynamics has also been published [20]. In collaboration with researchers from CNRS, Gif-Sur-Yvette, a VR-supported project has recently been initiated. Holocene sediment sequences from two lakes near Östersund in mid-central Sweden were collected from the lake ice in the spring of 2014. The project aims at improving our knowledge of centennial- to decadal-scale climate dynamics across central and northern Europe during the Holocene, with a specific focus on the last 2000 years. We will obtain stable oxygen-isotope records from ostracod valves in well-dated sediments from large and deep lakes. Integration of such data with corresponding records from speleothems and tree-ring sequences, which were collected in the autumn of 2014, will form the basis for reconstructions of the oxygen-isotope composition of precipitation ( 18OP), providing information on changes in atmospheric circulation patterns and precipitation seasonality. Some of the most significant challenges in palaeooceanographic research arise from the need to both understand and reduce the uncertainty associated with proxy methods. This is especially important for shelf and coastal environments where increasing numbers of high-resolution palaeorecords are being generated that are likely to be characterized by greater environmental variability. We work with both marine proxy calibration through a combination of field- and culture-based approaches to provide better reconstructions of salinity, temperature and oxygen status [21,22,23,24] and applying a suite of proxy methods down core to better understand climate and environmental variability over time [25,26,27]. Recently we have worked towards developing the use of Mn/Ca as a proxy for hypoxic conditions in bottom water [23,24]. We conduct research on marine sediment cores off NW Africa spanning the last 35 000 years, where we have been able to link changes in aeolian dust input with increased coastal upwelling and changes in primary productivity and a resulting response on the benthic environment [25]. Furthermore we have expanded the project to cover the Namibian upwelling system, SW Africa as well [26]. Here we document environmental changes covering the last 70 000 and again note the importance of the strength of the coastal upwelling system but also morphological changes of the area due to sea-level changes. These two systems, off NW and SW Africa, are the two strongest coastal upwelling systems in the world and it is of vital importance to better understand how the biological productivity varies over time and under different climate regimes (McKay’s PhD project). In WP 3 research is also conducted which is strongly linked to WP 1 and 2. One example is a study of coastal ocean acidification in the Skagerrak and Kattegat carried out in collaboration with colleagues from Japan. We note substantial changes, documented by micro Computed Tomography, in the calcite shells of bottom-dwelling microorganisms (Foraminifera). These significant shell changes have occurred over the last 20 years and we tentatively link the changes to a decrease in pH and in saturation state (L. Charrieau’s PhD project). Within this project we have also noted that pH and saturation state in the Baltic Sea are so low that calcite foraminifera dissolve their shell, however, they are still alive [28,29]. A new project “The Baltic Sea - a marginal marine environment as a potential recorder and player in global climate” was initiated during 2014, where we work on IODP samples from the Baltic Sea (Quintana Krupinski, Filipsson, Groeneveld et al). Initial results, based on oxygen and carbon isotopes, suggest that the Kattegat area experienced substantial freshening during the early part of the deglaciation and also that the bottom waters were stagnant [27]. We have reconstructed salinity in the southern Baltic during the mid Holocene using a combination of 87/86Sr isotopes and the process length of dinoflagellate cysts [30]. We noted maximum salinity was 13 compared to the modern salinity of 7. The period with the highest salinity (6.4- 3.9 ka BP) concur with the highest productivity in the record [30]. Our research on the history of hypoxia in the Baltic Sea and the mechanisms bringing us into hypoxia and those mechanisms creating conditions that allow for the termination of hypoxia continue to receive attention [31,32]. We have recently shown that glacio-isostatic uplift and reconfiguration of the Baltic Sea basin was an important component for terminating hypoxia in the mid-Holocene [33]. Our attention is currently focused on the role of the coastal zone in processing nutrients and organic matter with field work at different sites in the Baltic Sea (Öre River estuary and Stockholm Archipelago, Sweden). The role of chemical weathering, which ultimately consumes CO2 from the atmospheric, has been another important theme addressed in WP3. Our research has partly focused on how the products of chemical weathering are accumulating on the continents [34,35], but also the changes that occur in Si transport from the continents to the

75

LUCCI Annual Report – work package 3 oceans [36]. We are also addressing very long time scales and how the Earth’s long-term silica cycle is intimately linked to weathering rates and biogenic uptake. Changes in weathering rates and the retention of silica on land have altered silica availability in the oceans for hundreds of millions of years [37]. WP3’s research group “Late Quaternary geomagnetic field variability, cosmogenic radionuclide production and reconstructions of solar activity” (http://www.geol.lu.se/kvg/eng/grp_latequa.htm) could celebrate two successful PhD defenses by Anette Mellström and Florian Adolphi. Florian’s work has led to the first solar activity reconstruction for the end of the last ice age (20000 to 10000 years BP) and the identification of sun-climate linkages during the last glacial maximum. These results have been published in Nature Geoscience [38] and received significant attention in the public media. Anette’s results have recently been published and they include detailed investigations of a climate shift around 2750 yrs BP likely connected to solar forcing [39]. In addition, her results regarding the acquisition process of natural remanent magnetization [40] can prove valuable for better solar activity reconstructions and using paleomagnetism as a dating/correlation tool. Bryan Lougheed identified patterns in geomagnetic field changes that will be valuable for synchronizing climate records to investigate the timing of climate change [41]. Solar-induced 10Be and 14C variations have been used for high-resolution investigations of time scale differences between the IntCal13 14C time scale and the Greenland ice core time scale [42] and geomagnetic induced 10Be and 14C changes revealed major uncertainties in the 14C calibration curve towards the limit of the method around 40000 years ago [43]. Further results include the identification of solar induced changes in the summer sea-surface temperatures in the northern North Atlantic [44], 10Be-based climate reconstructions for Northern Greenland for the EEM interglacial [45] and the contribution of the above-mentioned results to a new data-base that will allow better comparison and integration of time scales for imprived studies regarding the temporal development of climate change [46]. This research branch of LUCCI/WP3 could be strenghtened in 2014 by the employment of a postdoctoral researcher (Markus Czymzik) and two PhD students (Emma Nilsson & Florian Mekhaldi). Markus and Florian will be looking at processes influencing 10Be in varved lake sediments to explore their potential for novel solar activity reconstructions and as a dating/correlation tool. Emma’s project will involve high-resolution solar activity reconstructions from ice cores. Efforts on climate modeling within WP3 showed three major steps forward i) a transient simulation for AD 800-2000 using the COSMOS-wiso coupled atmsophere-ocean climate model with stable water isotopes in the hydrological cycle was completed ii) the ECHAM5-wiso atmospheric model model with stable water isotopes in the hydrological cycle was setup for paleoclimate mode including changes to the ice sheet configuration iii) the SOCOLv3 chemistry climate model was set up and tested. For ii) and iii) the tests were performed at the LUNARC computing center in Lund. The COSMOS-wiso AD 800-2000 run has many applications including reconstruction of circulations patterns by linking to ice cores and tree rings, as well as attribution of climate forcings (solar activity, volcanic aerosols) in palaoclimate archives. ii) will in 2015 lead to expansion of the modelling efforts within the Atlantis project mentioned above and iii) will enable us to test hypothesis for climate effects of solar activity and geomagnetic field changes. Two papers on modeling of stable water isotopes in precipitation during the Eemian were published in 2014. Both papers deal with the climate response to orbital changes and the climatic imprint of these changes in stable water isotopes, with one paper focused on the Greenland temperature signal [47] and the other focused on the precipitation amount in the tropics [48].

References [1] Ljung, K., Holmgren, S., Kylander, M., Sjolte, J., Van der Putten, N., Kageyama, M. & Björck, S., submitted. The Last Termination in the central South Atlantic. Quaternary Science Reviews. [2] Björck, S., in prep. A Last Termination Aeolian record of changes of the Southern Hemisphere Westerlies in Tierra del Fuego. [3] WAIS Divide Project Members, 2015. Precise interpolar phasing of abrupt climate change during the last ice age. Nature 520, 661-665. [4] Schmitt, J., Schneider, R., Elsig, J., Leuenberger, D., Lourantou, A., Chapellaz, J., Köhler, P., Joos, F., Stocker, T.F., Leuenberger, M., & Fischer, H., 2012. Carbon isotope constrains on the deglacial CO2 rise from the ice cores. Science 336, 711-714. [5] Van der Putten, N., Verbruggen, C., Björck, S., Michel, E., Disnar, J.R., Chapron, E., Moine, B.N., de Beaulieu, J.L., in press. The Last Termination in the South Indian Ocean: a unique terrestrial record from Kerguelen Islands (49°S) situated within the Southern Hemisphere westerly belt. Quaternary Science Reviews.

76

LUCCI Annual Report – work package 3 [6] Stenni, B., Masson-Delmotte, V., Johnsen, S., Jouzel, J., Longinelli, A., Monnin, E., Röthlisberger, R., Selmo, E., 2001. An oceanic cold reversal during the last deglaciation. Science 293, 2074-2077. [7] Toggweiler, J.R., Russell, J.L., Carson, S.R., 2006. Midlatitude westerlies, atmospheric CO2, and climate change during the ice ages. Paleoceanography 21, PA2005. [8] Åkesson, C., Nielsen, A.B., Broström, A., Persson, T., Gaillard, M.-J. and Berglund, B. (2015): From landscape description to quantification: a new generation of reconstructions provides novel perspectives on Holocene regional landscapes of SE Sweden. The Holocene. 25, 78-93. [9] Feurdean, A., Marinova, E., Nielsen, A. B., Liakka, J., Veres, D., Hutchinson, S. M., Braun, M., Timar-Gabor, A., Astalos, C., Mosburgger, V. and Hickler, T. 2015. Origin of the forest steppe and exceptional grassland diversity in Transylvania (central-eastern Europe). Journal of Biogeography, 42, 951–963. [10]Poska, A., Saarse, L., Koppel, K., Nielsen, A.B., Avel, E., Vassiljev, J., Väli, V. (2014) The Verijärv area, South Estonia over the last millennium: A high resolution quantitative land-cover reconstruction based on pollen and historical data. Review of Palaeobotany and Palynology 207, 5-17. [11]Pirzamanbein, B., Lindström, J., Poska, A., Sugita, S., Trondman, A.-K., Fyfe, R., Mazier, F., Nielsen, A.B., Kaplan, J.O., Bjune, A.E. (2014) Creating spatially continuous maps of past land cover from point estimates: A new statistical approach applied to pollen data. Ecological Complexity 20, 127-141 [12]Trondman A.K., Gaillard M.J., Mazier F., Sugita S., Fyfe R., Nielsen A.B., Twiddle C. et. Al. (2014). Pollen‐based quantitative reconstructions of Holocene regional vegetation cover (plant functional types and land cover types) in Europe suitable for climate modeling. Global Change Biology 21, 676–69. [13]Marquer, L., Gaillard, M.-J., Sugita, S., Trondman, A.-K., Mazier, F., Nielsen, A.B., Fyfe, R.M., Odgaard, B.V, Alenius, T., Birks, H.J.B, Bjune, A.E, Christiansen, J., Dodson, J., Edwards, K.J., Giesecke, T., Herzschuh, U., Kangur, M., Lorenz, S., Poska, A., Schult, M., Seppä, H. (2014): Holocene changes in vegetation composition in northern Europe: why pollen-based quantitative reconstructions matter. Quaternary Science Reviews 90, 199-216. [14]Li, Y., Nielsen, A. B., Zhao, X., Shan, L., Wang, S., Wu, J. and Zhou, L. (2015): Pollen production estimates (PPEs) and fall speeds for major tree taxa and relevant source areas of pollen (RSAP) in Changbai Mountain, northeastern China. Review of Palaeobotany and Palynology, 216, 92-100. [15]Bragée, P., Mazier, F., Nielsen, A.B., Rosén, P., Fredh, D., Broström, A., Granéli, W. and Hammarlund, D. (2015): Historical TOC concentration minima during peak sulfur deposition in two Swedish lakes. Biogeosciences, 12, 307– 322, 2015. [16]Alexanderson, H. & Fabel, D. 2015: Holocene chronology of the Brattforsheden delta and inland dune field, SW Sweden. Geochronometria 42, 1-16. [17]Alexanderson, H. & Henriksen, M. in press. A short-lived aeolian event during the Early Holocene in southeastern Norway. Quaternary Geochronology. [18]Edvardsson, J., Edwards, T.W.D., Linderson, H. and Hammarlund, D. 2014. Exploring climate forcing of growth depression in subfossil South Swedish bog pines using stable isotopes. Dendrochronologia 32, 55-61. [19]Loisel, J., Yu, Z., Beilman, D.W., Camill, P., Alm, J., Amesbury, M.J., Anderson, D., Andersson, S., Bochicchio, C., Barber, K., Belyea, L.R., Bunbury, J., Chambers, F.M., Charman, D.J., De Vleeschouwer, F., Fialkiewicz-Koziel, B., Finkelstein, S.A., Galka, M., Garneau, M., Hammarlund, D., Hinchcliffe, W., Holmquist, J., Hughes, P., Jones, M.C., Klein, E.S., Kokfelt, U., Korhola, A., Kuhry, P., Lamarre, A., Lamentowicz, M., Large, D., Lavoie, M., MacDonald, G., Magnan, G., Makila, M., Mallon, G., Mathijssen, P., Mauquoy, D., McCarroll, J., Moore, T. R., Nichols, J., O'Reilly, B., Oksanen, P., Packalen, M., Peteet, D., Richard, P.J.H., Robinson, S., Ronkainen, T., Rundgren, M., Sannel, A.B., Tarnocai, C., Thom, T., Tuittila, E.-S., Turetsky, M., Väliranta, M., van der Linden, M., van Geel, B., van Bellen, S., Vitt, D. and Zhao, Y. 2014. A database and synthesis of northern peatland soil properties and Holocene carbon and nitrogen accumulation. The Holocene 24, 1028-1042. [20]Edvardsson, J., Poska, A., Van der Putten, N., Rundgren, M., Linderson, H. and Hammarlund, D. 2014. LateHolocene expansion of a south Swedish peatland and its impact on marginal ecosystems: Evidence from dendrochronology, peat stratigraphy and palaeobotanical data. The Holocene 24, 466-476. [21]Filipsson HL, Austin WEN, Darling K , Groeneveld J and the CONTEMPORARY team. 2014 Baltic benthic foraminifera cultured over a large salinity gradient: first results and comparison with field data. The Micropalaeontological Society Foraminifera and Nannofossil Groups Spring Meeting, NIOZ, Texel the Netherlands, June 2014 [22]Groeneveld J, Filipsson,HL Austin, WEN Darling, K., Quintana Krupinski N., 2015. Benthic foraminifera cultured over a large salinity gradient: first results and comparison with field data from the Baltic Sea. European Geophysical Union General Assembly, EGU2015-11429. [23]Groeneveld J & Filipsson HL. 2013. Mg/Ca and Mn/Ca ratios in benthic foraminifera: The potential to reconstruct past variations in temperature and hypoxia in shelf regions. Biogeosciences 10, 5125–5138, doi:10.5194/bg-105125-2013.

77

LUCCI Annual Report – work package 3 [24]McKay CL, Groeneveld J, Filipsson HL, Gallego-Torres D, Whitehouse M., Toyofuku T., Romero O.E. (submitted) A comparison of benthic foraminiferal Mn/Ca and sedimentary Mn/Al as proxies of relative bottom water oxygenation in the low latitude NE Atlantic upwelling system. [25]McKay CL, Filipsson HL, Romero OE, Stuut JBW, & Donner B. 2014. Pelagic-benthic coupling within an upwelling system of the subtropical northeast Atlantic over the last 35 ka BP. Quaternary Science Reviews, 106: 299-315 http://dx.doi.org/10.1016/j.quascirev.2014.04.027 [26]McKay CL, Filipsson HL, Romero OE, Stuut JBW., Björck S, & Donner B (submitted) The interplay between the surface and bottom water environment within the Benguela Upwelling System over the last 70 ka. [27]Quintana Krupinski, N.B.; Filipsson, HL.; Bokhari-Friberg, Y; Knudsen, K. L; Mackensen, A.; Groeneveld, J; Austin, W. E. N., IODP Expedition 347 Scientists. 2015. Foraminiferal stable isotope constraints on salinity changes in the deglacial and early Holocene Baltic Sea region. European Geophysical Union General Assembly, EGU2015-7196. [28]Charrieau L., Schoon P., et al., 2014 Zombie foraminifera reveal impacts of ocean acidification in the Baltic Sea. The Micropalaeontological Society Foraminifera and Nannofossil Groups Spring Meeting. The Royal Netherlands Institute for Sea Research (NIOZ), Texel, the Netherlands, June 23-25 2014. [29]Filipsson H.L., Charrieau L., et al., 2014. Microfossils and microbial biomarkers as recorders of coastal ocean acidification in the Skagerrak-Baltic Sea region. Gordon Research Conference in Ocean Global Change Biology, Waterville, USA, July 4-12 2014. [30]Ning W., Andersson P., Ghosh A., Khan M., Filipsson, H.L (submitted) Quantitative salinity reconstruction of the Baltic Sea during the Mid-Holocene. [31]Carstensen, J., J. Andersen, B.G. Gustafsson, D.J. Conley. 2014. Deoxygenation of the Baltic Sea during the last century. Proc. Natl. Acad. Sci. 111:5628-5633. [32]Lenz, C., T. Jilbert, D.J. Conley, M. Wolthers and C.P. Slomp. 2014. Are recent changes in sediment manganese sequestration in the euxinic basins of the Baltic Sea linked to the expansion of hypoxia? Biogeosciences In revision. [33]Jilbert, T., D.J. Conley, B.G. Gustafsson, C.P. Funkey and C.P. Slomp. 2015. Glacio-isostatic control on the Holocene distribution of hypoxia in a high-latitude shelf basin. Geology 43: 427–430. [34]Frings, P.J., W. Clymans, E. Jeppesen, T.L. Lauridsen, E. Struyf, and D.J. Conley. 2014. Perspectives: Lack of steadystate in the global biogeochemical Si cycle: emerging evidence from lake Si sequestration. Biogeochemistry 117:255-277. [35]Barão, L., W. Clymans, F. Vandevenne, P. Meire, D.J. Conley, E. Struyf. 2014. Pedogenic and biogenic amorphous Si distribution along a temperate land use gradient. Eur. J. Soil Sci. 65, 693–705. [36]Frings P., W. Clymans, G. Fontorbe, C. De La Rocha and D.J. Conley. Submitted. Linking the terrestrial and oceanic Si cycles on millennial-plus timescales. Chemical Geology [37]Conley, D. J. and J. C. Carey. 2015. News & Views: Silica cycling over geologic time. Nature Geo. 8: 431-432. [38]Adolphi F., Muscheler R., Svensson A., Aldahan A., Possnert G., Beer J., Sjolte J., Björck S., Matthes K., and Thiéblemont, R., 2014. Persistent link between solar activity and Greenland climate during the Last Glacial Maximum, Nature Geoscience, 7, 662-666. [39]Mellström, A., Van der Putten, N., Muscheler, R., De Jong, R., and Björck, S., 2015, A shift towards wetter and windier conditions in southern Sweden around the prominent solar minimum 2750 cal a BP: Journal of Quaternary Science, v. 30, no. 3, p. 235-244 [40]Mellström, A., Nilsson, A., Stanton, T., Muscheler, R., Snowball, I.,Suttie N., Post-depositional remanent magnetization lock-in depth in precisely dated varved sediments assessed by archaeomagnetic field models, Earth and Planetary Science Letters, 410, 186-196, 2015. [41]Lougheed, B.C, Nilsson, A., Björck, S., Muscheler, R, Snowball, I, A deglacial palaeomagnetic master curve for Fennoscandia – providing a dating template and supporting millennial-scale geomagnetic field patterns for the past 14 ka, Quaternary Science Reviews, 106, 155-166, 2014. [42]Muscheler, R., Adolphi, F. and Knudsen, M., Assessing the differences between the IntCal and Greenland ice-core time scales for the last 14,000 years via the common cosmogenic radionuclide variations, Quaternary Science Reviews, 106, 81-87, 2014. [43]Muscheler, R., Adolphi, F. and Svensson, A., Challenges in 14C dating towards the limit of the method inferred from anchoring a floating tree ring radiocarbon chronology to ice core records around the Laschamp geomagnetic field minimum, Earth and Planetary Science Letters, 394, 209-215, 2014. ] [44] Jiang, H., Muscheler, R., Björck, S., Seidenkrantz, M.-S., Olsen, J., Sha, L., Sjolte, J., Eiríksson, J., Ran, L., and Knudsen, K.-L., 2015, Solar forcing of Holocene summer sea-surface temperatures in the northern North Atlantic: Geology, v. 43, no. 3, p. 203-206.

78

LUCCI Annual Report – work package 3 [45]Sturevik-Storm A., Aldahan A., Possnert G., Berggren A.-M., Muscheler R., Dahl-Jensen D., Vinther B.M. and Usoskin I., 10Be climate fingerprints during the Eemian in the NEEM ice core, Greenland, Scientific reports, 4, doi:10.1038/srep06408, 2014. [46]Ramsey, C.B., Albert, P., Blockley, S., Hardiman, M., Lane, C., Macleod, A., Matthews, I.P., Muscheler, R., Palmer, A., Staff, R.A., Integrating timescales with time-transfer functions: a practical approach for an INTIMATE database, Quaternary Science Reviews, 106, 67-80, 2014. [47]Sjolte et al., 2014 Tellus B 2014, 66, 22872, http://dx.doi.org/10.3402/tellusb.v66.22872 [48]Sjolte J, Hoffmann G, 2014. Modelling stable water isotopes in monsoon precipitation during the previous interglacial. Quaternary Science Reviews, 85: 119–135. doi:10.1016/j.quascirev.2013.12.006

79

LUCCI Annual Report – work package 3

Persistent solar influence on Greenland climate during the last Glacial Maximum Florian Adolphi (a), Raimund Muscheler (a), Anders Svensson (b), Ala Aldahan (c,d), Göran Possnert (e), Jürg Beer (f), Jesper Sjolte (a), Svante Björck (a), Katja Matthes (g) and Rémi Thiéblemont (g) (a) (b) (c) (d) (e) (f) (g)

Department of Geology – Quaternary Sciences, Lund University, Sweden Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen, Denmark Department of Earth Sciences, Uppsala University, Uppsala, Sweden Department of Geology, United Arab Emirates University, Al Ain, UAE Tandem Laboratory, Uppsala University, Uppsala, Sweden Swiss Federal Institute of Aquatic Science and Technology, Eawag, Dübendorf, Switzerland Division of Ocean Circulation and Climate, GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany

Introduction The Sun is by far Earth’s most important source of energy. Yet, it remains controversial whether and how changes in solar activity may affect the climate system. Paleoclimatic records offer the opportunity to study climate-forcing mechanisms during periods unperturbed by anthropogenic influences and thus, contribute to constraining climate sensitivity. In the absence of direct solar observations, reconstructions of past solar activity changes are crucial for gaining a longer-term perspective on solar influences on climate. The most reliable proxies for solar activity reconstructions are cosmogenic radionuclides such as 10Be and 14C. Their atmospheric production rates depend on the flux of galactic cosmic rays (gcr) which is modulated by the geo- and heliomagnetic fields. Hence, weak (strong) geo- and/or solar magnetic activity lead to enhanced (reduced) production rates of 10Be and 14C. Therefore, paleoarchives of 10Be and 14 C, such as ice cores and tree-rings, respectively, allow inferences of past solar activity variations. So far solar activity reconstructions have been limited to the last ~10,000 years. Here, we extend these efforts back to 22,500 years ago providing the first solar activity reconstructions for the end of the last ice age [1]. We then use this solar activity reconstruction to test whether there was a solar influence on Greenland climate at the termination of the last ice age.

Extending solar activity reconstructions back into the last Ice Age To reconstruct solar activity during the end of the last ice age we measured 10Be on the GRIP ice core and combined it with previously published 10Be data from the GRIP [2, 3] and GISP2 [4] ice cores. To test for the robustness of the ice core 10Be record, we compare it to 14C data based on tree rings [5] and the Hulu Cave speleothem [6] which have been converted into 14C production rates using carbon cycle modelling [7]. We normalize all records to their slow (>500 years) variations. This isolates the solar modulation of production rates by minimizing the geomagnetic modulation which is largely limited to millennial time scales [8, 9]. The good agreement of the 10Be record with high quality 14C data from tree-rings (figure 1, top) and consistency within error with lower resolution speleothem 14C data (figure 1, bottom) lends support to our 10Be based solar activity reconstruction. We find that glacial solar activity variations have been of comparable amplitude as seen during the Holocene (figure 1, horizontal dashed lines). Furthermore, the spectral characteristics of glacial solar activity variations have been similar to Holocene reconstructions indicating the presence of typical solar cycles (e.g. 88 and 210 yr cycles).

80

LUCCI Annual Report – work package 3

Figure 1. Past solar activity variations between 10,000 and 22,500 years BP (before present, AD 1950). Top: Normalized GRIP/GISP2 10Be fluxes (orange) in comparison to tree-ring based 14C production rates (blue). Bottom: Normalized GRIP/GISP2 10 Be fluxes (orange) in comparison to speleothem based 14C production rates (black). All data have been normalized to their slow (>500 yrs) variations and low pass filtered (cut off = 150 yrs) to isolate the solar modulation of cosmogenic radionuclide production rates as opposed to geomagnetic influences. The dashed grey horizontal lines indicate production rate changes expected for typical centennial solar activity changes seen during the Holocene [8].

Evidence for solar forcing of Greenland climate during the Last Glacial Maximum Comparing the 10Be-based solar activity reconstruction to climate records (δ18O) from the same ice core reveals a significant solar influence on Greenland climate during Greenland Stadial 2 (14,642 – 22,850 yrs BP). Additional aerosol records [10] indicate that periods of low solar activity (high 10Be) were associated with i) reduced input of terrestrial aerosols, ii) enhanced deposition of marine aerosols, as well as iii) increased snow accumulation rates at the ice core site. We interpret this as a more meridional atmospheric circulation during solar minima winters advecting moist Atlantic air masses to Greenland, leading to more winter precipitation. This surplus in winter precipitation leads to low δ18O values in the ice, which is otherwise dominated by a summer precipitation signal during the last glacial maximum [11]. We find this mechanism of enhanced meridional atmospheric circulation (i.e. a higher likelihood of high-latitude blocking situations) during solar minima winters consistent with high-top chemistry-climate model experiments run under pre-industrial conditions (CESM-WACCM) [12, 13] as well as reanalysis data [14]. This may indicate that a similar solar forcing mechanism was operating during the last glacial maximum and modern day despite otherwise vastly different boundary conditions.

Figure 2. Proxy evidence for a solar modulation of Greenland climate during Greenland Stadial 2. Sub-millennial (150-500 yr) GRIP (blue) and GISP2 (green) δ18O anomalies, their mean (dark blue, thick), and 10Be based solar activity variations (orange). The 10Be axis is reversed (note that low 10Be indicates high solar activity). Both individual δ18O anomaly records and their mean are significantly (p<0.01) correlated to solar activity (r2 = 0.2-0.3).

81

LUCCI Annual Report – work package 3 References [1] Adolphi, F., et al., Persistent link between solar activity and Greenland climate during the Last Glacial Maximum. Nature Geoscience, 2014. 7(9): p. 662-666. [2]

Muscheler, R., et al., Changes in the carbon cycle during the last deglaciation as indicated by the comparison of 10Be and 14C records. Earth and Planetary Science Letters, 2004. 219(3-4): p. 325-340.

[3]

Yiou, F., et al., Beryllium 10 in the Greenland Ice Core Project ice core at Summit, Greenland. Journal of Geophysical Research, 1997. 102(C12): p. 26783.

[4]

Finkel, R.C. and K. Nishiizumi, Beryllium 10 concentrations in the Greenland Ice Sheet Project 2 ice core from 3–40 ka. Journal of Geophysical Research, 1997. 102(C12): p. 26699.

[5]

Reimer, P.J., et al., IntCal13 and Marine13 Radiocarbon Age Calibration Curves 0–50,000 Years cal BP. Radiocarbon, 2013. 55(4): p. 1869-1887.

[6]

Southon, J., et al., A high-resolution record of atmospheric 14C based on Hulu Cave speleothem H82. Quaternary Science Reviews, 2012. 33: p. 32-41.

[7]

Siegenthaler, U., M. Heimann, and H. Oeschger, 14C variations caused by changes in the global carbon cycle. Radiocarbon, 1980. 22(2): p. 177-191.

[8]

Muscheler, R. and U. Heikkilä, Constraints on long-term changes in solar activity from the range of variability of cosmogenic radionuclide records. Astrophysics and Space Sciences Transactions, 2011. 7(3): p. 355-364.

[9]

Snowball, I. and R. Muscheler, Palaeomagnetic intensity data: an Achilles heel of solar activity reconstructions. The Holocene, 2007. 17(6): p. 851-859.

[10] Mayewski, P.A., et al., Major features and forcing of high-latitude northern hemisphere atmospheric circulation using a 110,000-year-long glaciochemical series. Journal of Geophysical Research, 1997. 102(C12): p. 26345. [11] Werner, M., et al., Borehole versus isotope temperatures on Greenland: Seasonality does matter. Geophysical Research Letters, 2000. 27(5): p. 723-726. [12] Gent, P.R., et al., The Community Climate System Model Version 4. Journal of Climate, 2011. 24(19): p. 49734991. [13] Martin-Puertas, C., et al., Regional atmospheric circulation shifts induced by a grand solar minimum. Nature Geoscience, 2012. 5(6): p. 397-401. [14] Compo, G.P., et al., The Twentieth Century Reanalysis Project. Quarterly Journal of the Royal Meteorological Society, 2011. 137(654): p. 1-28.

82

LUCCI Annual Report – work package 3

A comparison of size fractions in faunal assemblages of deepwater benthic foraminifera – a case study from the Benguela Upwelling System Claire McKay (a), Helena Filipsson (a), Oliver Björnfors (a) (a) Dept of Geology, Lund University

Upwelling Systems Coastal upwelling zones along the eastern boundary current systems (EBCS) of the ocean are currently of integral significance when considering the ecological, climatic and socioeconomic issues. Upwelling; a climatically driven process which results in the ascendancy of cool, nutrient enriched deeper waters to the ocean surface, infuses the surface layers with fresh nutrients [1]. This process influences marine primary productivity and sustains upper trophic levels, complex ecosystems and marine resources. Upwelling is also a key component of climate-active biogeochemical cycles including the carbon cycle [2]. A definitive challenge facing marine palaeoecologists is understanding how the EBCS have responded to past climatic changes which has implications of how the benthic environment will change in the future. To reconstruct the benthic environment of the past, we utilize the marine sediment record. Sediment cores from regions characterised by high sedimentation rates and well preserved microfossils; in this case off the coast of southwest Africa, provide a unique opportunity to study the sensitivity of the benthic environment. In particular, we analyse the community composition of benthic foraminifera which are a valuable proxy to infer past environmental conditions since certain species have specifically defined environmental preferences and hence indicate the palaeoenvironment. The two major factors which play an integral role in controlling the benthic microhabitat and subsequently, foraminifera populations are the flux and temporal fluctuations of organic matter (food) to the sea floor and bottom and pore water oxygen content [3]. Foraminifera play a primary role in the initial breakdown and cycling of carbon [4] and also form a link between lower and higher trophic levels [5].

Benthic foraminiferal faunal analyses Benthic foraminiferal assemblages are fundamental in palaeoceanographic research and are dependent on quantitative data by manually counting species in washed sediment residues. Assemblage composition has been of integral significance within benthic foraminiferal based research; however, the comparability of results from different investigations is often hampered by the analysis of different size fractions. Investigations into the various size fractions of benthic foraminiferal assemblages have been carried out over the years [6], however no standardised consensus has been reached to determine the optimum size fraction that provides the most comprehensive information on faunal composition. In our study we analyse samples from a marine sediment core retrieved from the Benguela Upwelling System (BUS) to address this issue by comparing faunal data generated when using >63 µm and >125 µm fractions. The results highlight that even between the >63 and >125 µm fractions, relative abundances can vary from one minimum size threshold to another. Preparation of samples for benthic foraminiferal faunal analyses has previously been reported and analyses of 300 specimens were determined to the species level from each of the three >125 µm fractions separately; 125-250 µm, 250-500 µm and >500 µm [7]. Here we extend this dataset by accounting for the 63-125 µm fraction. In order to potentially obtain a wider scope of the faunal composition changes in the 63-125 µm size fraction, samples occurring under different productivity regimes were selected where the greatest shifts in faunal composition have been noted [7]. The species found in the 63-125 µm fraction are Bulimina aculeata, Bulimina pseudopunctata, Bulimina mexicana, Cassidella bradyi, Chilostomella oolina, Epistominella exigua, Eubuliminella exilis and Nonionella iridea. Their concentrations and relative abundances are plotted alongside the >125 µm fraction in figures 1 and 2.

83

LUCCI Annual Report – work package 3

Figure 1. Benthic foraminiferal species concentrations (specimens cm3), overall concentration and benthic foraminiferal accumulation rate analysed from the >63 µm and >125 µm size fractions.

Figure 2. Relative abundances (%) of benthic foraminiferal species analysed from the >63 µm and >125 µm size fractions.

Benthic foraminiferal concentrations are lower when only the >125 µm is considered, with a maximum difference of 1755 specimens/cm3 (Figure 1). Similarly, the benthic foraminiferal accumulation rate (BFAR) shows differences ranging from 740-52690 cm-2 ka-1. Perhaps the most noteworthy paleoecological finding is the underrepresentation of Epistominella exigua. Whilst the overrepresentation of certain species, in particular Eubuliminella exilis is relatively magnified when the finer fraction is not considered, the palaeoenvironmental interpretation of our data does not change significantly. In conclusion, we underline the need for consideration of the effect of specimen size on benthic foraminiferal concentrations and species dominance, drawing further attention to this issue and the need for a consistent protocol when comparing different faunal datasets. Whilst a minimum size threshold of 125 µm is representative from this

84

LUCCI Annual Report – work package 3 deep sea, upwelling environment, the 63 µm should still be considered by occasional spot checks. Overall however, whilst the inclusion of the >63 µm fraction can significantly change the concentration and accumulation rate of the benthic foraminiferal specimens along with the concentrations of the species, relative abundances (%) do not alter greatly and therefore has no impact on our palaeoenvironmental interpretation.

References [1] Bakun, A., 1990. Global climate change and intensification of coastal upwelling. Science, 247, 198-201. [2] Mariotti, V., Bopp, L., Tagliabue, A., Kageyama, M. & Swingedouw, D., 2012. Marine productivity response to Heinrich events: a model-data comparison, Climate of the Past Discussions, 8, 557-594. [3] Mackensen, A., Schmiedl, G., Harloff, J. & Giese, M., 1995. Deep-sea Foraminifera in the South Atlantic Ocean: ecology and assemblage generation: Micropaleontology, 41, 342 - 358. [4] Woulds, C., Cowie, G.L., Levin, L.A., Andersson, J.H., Middelburg, J.J., Vandewiele, S., Lamont, P.A., Larkin, K.E., Gooday, A.J., Schumacher, S., Whitcraft, C., Jeffreys, R.M., & Schwartz, M., 2007. Oxygen as a control on seafloor biological communities and their roles in sedimentary carbon cycling, Limnology and Oceanography, 52, 1698-1709 [5] Gooday, A.J., Levin, L.A., Linke, P. & Heeger, T., 1992. The role of benthic foraminifera in deep-sea food webs and carbon cycling. In: Rowe, G.T. & Pariente, V. (ed’s) Deep-sea food chains and the global carbon cycle. Kluver Academic Publishers, Dordrecht. [6] Schröder, C.J., Scott, D.B. & Medioli, F.S., 1987. Can smaller benthic foraminfiera be ignored in paleoenvironmental analyses?, Journal of Foraminiferal Research, 17, 101-105. [7] McKay, C.L., Filipsson, H.L., Romero, O.E., Stuut, J.-B.W. & Donner, B., submitted. The interplay between the surface and bottom water environment within the Benguela Upwelling System over the last 70 ka. Submitted to Paleoceanography.

85

LUCCI Annual Report – work package 3

Reconstructing the Holocene landscape development of SE Blekinge using pollen and the REVEALS model Anne Birgitte Nielsen (a,b), Christine Åkesson (b,c), Anna Broström (d), Thomas Persson (b), Marie-José Gaillard (e) & Björn E. Berglund (b) (a) (b) (c) (d) (e)

Dept of Physical Geography and Ecosystem Sciences, Lund University Dept of Geology, Lund University Florida Institute of Technology National Historical Museums of Sweden Linnaeus University

Introduction The landscape of southern Sweden, like the rest of Europe, has changed over millennia due to human impact in interaction with climate change and plant species migrations. Agriculture and pastoralism has lead to an opening of the landscape, which has had secondary effects including changes in carbon storage and transport, increased soil erosion as well as changes in biodiversity. To study landscape changes at a regional scale, we have in this study analysed fossil pollen assemblages from a large lake, Färskesjön in Blekinge [1, 2] and applied a quantitative landscape reconstruction model, REVEALS [3] to obtain quantitative estimates of the past vegetation composition.

Methods Study site Lake Färskesjön on the Torhamn Peninsula in southeastern Blekinge is a classical site for the study of the landscape development and human impact during the late glacial and Holocene periods[1]. It is a large lake of 50 ha, with a water depth in the central part of 3–4 m. The lake is located 14 m.a.s.l. and is drained by a brook running southwest towards the sea, which is ca. 3 km away.

Sediment sampling and analysis A sediment core was first collected in 1956, and a detailed pollen analysis was carried out [1]. However, at that time, it was not possible to collect the topmost, water rich lake sediments, so the latest few centuries of the Holocene vegetation development could not be described. To obtain data for this crucial period of human impact on the landscape (including the near half century that has gone by since the first sampling), the site was cored again in 2013 [2]. We used a HON-Kajak corer [4] to obtain the surface sediments, and a Russian corer to retrieve a full Holocene sediment sequence. By dating the new sediment sequence based on four AMS 14C-dates of terrestrial macrofossils and 210Pb measurements, it was possible to establish an agedepth relationship for the last 3000 years (Figure 1). By correlating the new and old high resolution pollen sequences, the top age of the 1956 core could be estimated to 350 cal. BP, and its age-depth model adjusted accordingly.

Landscape reconstruction To obtain quantitative estimates of past landcover, and thereby provide a new perspective on the Holocene landscape development, we applied the REVEALS model (Regional Vegetaton EstimAtes from Large Sites) [3] to both the old and new pollen records. REVEALS is developed to account for differences between species in terms of relative pollen productivity and pollen dispersal [3], and thus “translate” pollen percentages into estimates of the composition of vegetation cover at a regional (50-100 km) scale [6]. We used existing and well tested pollen productivity estimates from Southern Scandinavia [7].

86

LUCCI Annual Report – work package 3

Figure 1. Age–depth model for core 2013 based on four 14C dates and 210Pb measurements. The IntCal09 calibration curve and smoothing spline was applied using CLAM [5]. For more details, see [2]

Results and discussion The REVEALS based estimates (which are in line with historical landscape maps) indicate a much higher degree of landscape openness, created by long-term farming and grazing than is apparent from the uncorrected pollen assemblages (Figure 2). Importantly, the degree to which the raw pollen percentages underestimate openness varies greatly over time (Figure 2), depending on the species composition of both the forest and the open-land plant communities, underlining the importance of model application. The effect, on landscape openness, of applying the REVEALS model is thus largest during the early Holocene, where the estimated cover of grass- and heatland is around five times greater than the corresponding pollen percentages. This is probably due to the high abundance of trees speices with high pollen productivities, especially Betula (Birch) and Pinus (Pine). The application of REVEALS is therefore highly relevant to understand the vegetation development, not only due to human impact in the late Holocene, but also the natural landscape openness in the pre-agricultural period, which was much higher that previously understood. This openness may be due to natural disturbances by e.g. wind, fire or the effect of wild grazing animals. The thin soils overlaying the bedrock in large parts of the catchment may also have contributed to a relatively open vegetation structure, with enough light for grasses, herbs and Calluna (Heather) to thrive. In regard to the cultural landscape development, the analysis shows that humans began to noticeably open up the landscape at around 3200 cal. BP. There was an expansion 2300-1500 cal. BP, during the Roman Iron Age, followed by some reforestation. There was a new expansion in the Early Medieval period, followed by the Late Medieval decline at 600-700 cal. BP, where the cropland area was reduced by half. The highest degree of landscape openness was reached after 350 cal. BP (and was therefore not seen in the original 1956 core). After 200 cal. BP, openness has been reduced due to plantation, especially of coniferous forest. For the late Holocene, the REVEALS based quantification of past land-use changes can be related to changes in erosion in the catchment, as reflected by periods of low loss-on-ignition (LOI) in the lake sediments [2]. At a regional scale, this erosion and consequent transfer of nutrients [8] and organic carbon [9] from the land to the aquatic environment may have affected the nutrient loading to the coastal areas of the Baltic Sea [10; 11].

87

360

0

320 340

0

1000

360

1000

380

60

80

100

20

C a

3000

20

20

0

40

60

80

100

300

300 1000

350

1956 core, adjusted age scale

2000

400

Y e ar s

40

lB D P ec id uo us w oo dl an d

0

20

C on C ife al r l o P un us as a w v C ture ulg ood ro pl an aris lan d D an d e d m pt ea h, do cm w s

Y ea rC al

3000

B D P ec id uo us w oo dl an d

380

2000 400

400

1000

2000

2000

3000

3000

4000

4000

5000

5000

6000

6000

7000

7000

700

8000

8000

750 800

9000

9000

10000

10000

11000

11000

450

450

20 0 0

1000

350 400

2000

B et ul a

0 500 1000 1500 2000 2500 3000

C on i C fero al lu us na w P he oo as a dla tu nd C re th ro a pl n an d d me ad ow s

320 340

D ep th ,c m

2013 core

D ep th ,c m Y ea rs C al B P D ec id uo us w oo C dl on an d C ife al ro lu us Pa n s a h wo C tur ea od ro e th la p a nd A lan nd ln d m ea D us ep do w Be th, s tu cm Y la ea rC al BP D ec id uo us C w on oo C ife dl al ro an lu us d na w he oo P at dl as h an C tur d ro e pl an an d d m ea A do ln us w s

LUCCI Annual Report – work package 3

3000

2

4

6

2

4

6

500

500 4000

550

550

5000 600

600 6000

650

650

7000

700

8000

750 800

9000 10000

850

850

11000 900

12000

950

13000

20

40

60

80

900

12000

100 950

13000

12000 20

40

60

80

100

13000 0

Uncorrected pollen %

REVEALS estimated cover %

Ratio openness REVEALS/Pollen

Figure 2. Summary pollen percentages and REVEALS-based land-cover categories from lake Färskesjön. The ratio between the sum of REVEALS estimates and pollen percentages for the open-land taxa (Calluna heathland, pasture and meadow, and cropland) in each sample is also shown.

References [1] Berglund BE (1966) Late-Quaternary Vegetation in Eastern Blekinge, South-Eastern Sweden: A Pollen-Analytical Study (Opera Botanica 12). Stockholm: Almqvist & Wiksel.

[2] Åkesson C, Nielsen AB, Broström A., Persson T, Gaillard M-J and Berglund BE (2015) From landscape description to quantification: A new generation of reconstructions provides new perspectives on Holocene regional landscapes of SE Sweden. The Holocene 25: 178-193.

[3] Sugita S (2007) Theory of quantitative reconstruction of vegetation I. Pollen from large sites REVEALS regional vegetation composition. The Holocene 17: 229–241.

[4] Renberg I (1991) The HON-Kajak sediment corer. Journal of Paleolimnology 6: 167–170. [5] Blaauw M (2010) Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quaternary Geochronology 5: 512–518.

[6] Hellman SEV, Gaillard MJ, Broström A and Sugita S (2008) The REVEALS model, a new tool to estimate past regional plant abundance from pollen data in large lakes: Validation in southern Sweden. Journal of Quaternary Science 23: 21–42.

[7] Mazier F, Broström A, Bragée P, Fredh D, Stenberg L, Thiere G, Sugita S and Hammarlund D (2015) Two hundred years of land-use change in the South Swedish Uplands: comparison of historical map-based estimates with a pollen-based reconstruction using the landscape reconstruction algorithm. Vegetation History and Archaeobotany, online first.

[8] Bradshaw EG, Rasmussen P and Odgaard BV (2005) Mid- to late-Holocene land-use changes and lake development at Dallund Sø, Denmark: Synthesis of multiproxy data, linking land and lake. The Holocene 15: 1152–1162.

88

LUCCI Annual Report – work package 3 [9] Bragée P, Mazier F, Nielsen AB, Rosén P, Fredh D, Broström A, Granéli W and Hammarlund D (2015) Historical TOC concentration minima during peak sulfur deposition in two Swedish lakes. Biogeosciences 12: 307-322.

[10] Zillén L and Conley DJ (2010) Hypoxia and cyanobacteria blooms – Are they really natural features of the late Holocene history of the Baltic Sea? Earth-Science Reviews 91: 877–889.

[11] Ghosh A, Ning W and Filipsson HL (2012) Microfossil assemblages response to anthropogenic influence over the last 2000 years in coastal Baltic Sea: Initial results. In: International Paleolimnology Symposium, Glasgow, 21–24 August 2012. IPS-2012, p. 153.

89

LUCCI Annual Report – work package 3

Atmospheric circulation patterns in the Southern Ocean during the Holocene - a terrestrial view from Iles Kerguelen (49°S, Indian Ocean) on Southern Hemisphere Westerly belt variability Nathalie Van der Putten (a), Cyriel Verbruggen (b) & Svante Björck (a) (a) Dept of Geology, Quaternary Sciences, Lund University (b) Dept of Geology and Soil Sciences, Ghent University, Belgium

Introduction and study area The Southern Hemisphere Westerly wind belt (SHW) dominates the Southern Hemisphere mid to high latitude climate system. It has been suggested that changes in the strength and position of the SHW influenced global climate variability in the past through wind-induced upwelling in the Southern Ocean and subsequent increases in atmospheric CO2 concentrations [1]. However, proxy-based knowledge about orbital to centennial/millennial scale changes in the strengthening and/or latitudinal shifts of the wind belt is still scarce and contradictory. For the midHolocene, for instance, some studies show a northward shift and/or increased zonal flow for the SHW while others show a southward displacement and/or weakening [2].

Figure 1. Synthesis of global climate change for the preindustrial period (c. 1700 AD) compared to the mid Holocene (c. 6 kyr BP). The question mark shows the blank spot in proxy- records in the Indian Ocean [after 3].

Southern Ocean marine records reflect changes in ocean circulation, partly connected to atmospheric circulation changes but with no unambiguously proxy-link to wind intensity. Furthermore, a precise chronology of marine records is hampered by the presence of a varying 14C depletion of ocean surface waters compared to the atmosphere and many marine records lack the possibility for high temporal resolution studies. Most terrestrial records originate from South America and from New Zealand for obvious geographical reasons, resulting in substantial gaps in the proxybased Holocene climate history in general and SHW variability in particular. The Southern Indian Ocean is a blank spot in this respect (Fig. 1). Well-dated and high-resolution terrestrial proxy-records from this area are highly needed to obtain a more complete Southern Hemispheric view on this circulation pattern. A prerequisite for reconstructing changes in zonal winds, currently situated between 35° to 60° S, is the availability of terrestrial records that (i) mainly reflect atmospheric conditions and (ii) are situated on a latitudinal transect covering the wind belt.

90

LUCCI Annual Report – work package 3

Figure 2. Map of the Southern Hemisphere high and mid-latitudes with the location of Kerguelen Islands (green star), Iles Crozet and Ile Amsterdam (white stars), the oceanic fronts and the mean annual zonal winds (m/s) at 850mb averaged for the period 19792009 based on Re-analysis data (NCEP CFSR).

Iles Kerguelen, Iles Crozet and Ile Amsterdam at 49°, 46° and 37°S in Indian Ocean respectively (Fig. 2) are ideally located to reconstruct palaeo-intensity and position of the SHW with Kerguelen in the core and Amsterdam at the northern edge. During three field campaigns, in collaboration with the French polar institute (IPEV) in 2006, 2013 and 2014, peat sequences were collected from the three islands/archipelagos, resulting in a unique set of terrestrial records [e.g. 4]. Radiocarbon dating of peat archives usually results in reliable age-depth models and their proxies mainly reflect atmospheric circulation changes. Moreover, the South Indian Ocean is the moisture source area for the EPICA Dome C (EDC) ice core record [5], i.e. an ideal position for understanding mechanisms behind variations in climate parameters reconstructed from Antarctic ice core data.

Preliminary results and discussion In a peat record from Iles Kerguelen (49°S, Indian Ocean, Fig. 1, green star), we found multi-centennial variability in the late Holocene (after 4 kyr BP). The record has been investigated using a multi-proxy approach (pollen and plant macrofossils, XRF core scanning, magnetic susceptibility, biogenic silica content (BSi), total organic carbon content and humification degree of the peat (Van der Putten et al., in prep). In fig. 3A we show a selection of pollen spectra for the last 5500 years. The yellow bars show periods with increased Azorella selago pollen (higher wind intensity, green curve), increased Ranunculus sp. (increased humidity, blue curve) suggesting cyclic SHW influence in Iles Kerguelen at 49°S. Galium antarctica, a plant species growing in more sheltered conditions, is present in between the wet-windy periods (black curve). Comparison with a precipitation record off Chile at 41° S (Fig. 3B) [6], reveals opposite centennial scale variations of SHW influence: strengthening (weakening) at 41° S coincides with weakening (strengthening) in Iles Kerguelen suggesting an equator-ward (pole-ward) position or expansion (contraction) of the wind belt. However, both sites are situated far apart in different ocean basins. To test the opposite centennial scale variability in the Indian Ocean, however, a latitudinal transect is needed using well dated terrestrial records from Iles Kerguelen, Iles Crozet and Ile Amsterdam.

91

LUCCI Annual Report – work package 3

Figure 3. (A) Proxy-record of a peat sequence from Iles Kerguelen for the last 5500 yrs showing pollen (%) plotted against (B) a precipitation record taken from [6].

References [1] Toggweiler, J.R., Russell, J.L., Carson, S.R. (2006). Midlatitude westerlies, atmospheric CO2, and climate change during the ice ages. Paleoceanography 21, PA2005. [2] Varma, V., Prange, M., Merkel, U., Kleinen, T., Lohmann, G., Pfeiffer, M., Renssen, H., Wagner, A., Wagner, S., Schulz, M., 2012. Holocene evolution of the Southern Hemisphere westerly winds in transient simulations with global climate models. Clim. Past 8, 391-402. [3] Wanner, H., Beer, J., Bütikofer, J., Crowley, T.J., Cubasch, U., Flückiger, J., Goosse, H., Grosjean, M., Joos, F., Kaplan, J.O., Küttel, M., Müller, S.A., Prentice, I.C., Solomina, O., Stocker, T.F., Tarasov, P., Wagner, M., Widmann, M., 2008. Mid- to Late Holocene climate change: an overview. Quaternary Science Reviews 27, 1791-1828. [4] Van der Putten, N., Hébrard, J.-P., Verbruggen, C., Van de Vijver, B., Disnar, J.-R., Spassov, S., Keravis, D., de Beaulieu, J.-L., De Dapper, M., Hus, J., Thouveny, N., Frenot, Y., 2008. An integrated palaeoenvironmental investigation of a 6200 year old peat sequence from Île de la Possession, Îles Crozet, sub-Antarctica. Palaeogeography, Palaeoclimatology, Palaeoecology 270, 179-195. [5] Stenni, B., Masson-Delmotte, V., Johnsen, S., Jouzel, J., Longinelli, A., Monnin, E., Röthlisberger, R. and Selmo, E. (2001). An oceanic cold reversal during the last deglaciation. Science 293, 2074-2077. [6] Lamy, F., Hebbeln, D., Rohl, U., Wefer, G., 2001. Holocene rainfall variability in southern Chile: a marine record of latitudinal shifts of the Southern Westerlies. Earth and Planetary Science Letters 185, 369-382.

92

WORK PACKAGE 4

From a greenhouse to icehouse world – The climatic evolution during the past 70 million years

LUCCI Annual Report – work package 4

94

LUCCI Annual Report – work package 4

WP 4: Progress Report Vivi Vajda (a) & Birger Schmitz (b) (a) Dept of Geology, Lund University (b) Div of Nuclear Physics, Lund University

Scientists involved in LUCCI- WP4 2008 - 2014:

WP-leaders Schmitz Vajda Senior scientists Ferrow Lindgren Terfelt Cronholm Alwmark Larsson Postdocs Bercovici Willumsen Meier Steinthorsdottir Visiting professor Ocampo Hori Monechi Schweitzer Rubinstein Phd students Holm Lindskog Mehlqvist Einarsson Bjärnborg Ashley Johan Samuele Martin Holstein

Gender Birger Vivi

M F

Embaie Johan Fredrik Anders Carl Linda

M M M M M F

Antoine Pi Mattias Margarete

M F M F

Adriana Rie Simonetta Mary Claudia

F F F F F

Sanna Anders Kristina Elisabeth Carolina Gumsley Gren Boschi Ellinor Jim

F M F F F M M M F M

Technical personnel KlintvikAhlberg Iqbal Master students Younes Badawy Kumar Santasaalo Afridi

Gender Git Faisal

F M

Hani Ahmed Pardeep Liina Saeed

M M M F M

95

LUCCI Annual Report – work package 4 Background Performed research (WP4) includes studies on the evolution of Earth through the perspective of deep geological time targeting the causes and patterns of major global environmental perturbations. During most of the last 500 million years the polar regions on Earth have been ice free and latitudinal temperature gradients were substantially smaller than today. The amount of greenhouse gases in Earth’s atmosphere has changed considerably over this period. In a geological perspective the present Earth is an icehouse world, with temperatures and atmospheric carbon dioxide levels significantly lower than normal. As late as in the early Paleogene, 65-40 million years ago, palm trees grew as far north as Kamchatka and alligators thrived at Ellesmere Island near the North Pole. During the last 40 million years the climate has gradually cooled culminating with the onset of the Quaternary period ca. 2 million years ago, when the polar regions of both hemispheres became covered with ice. During the last 100 kyr the northern hemisphere ice has extended as far south as southern Germany, and has now retreated to Greenland, but in a geological perspective we still live in an ice age. Our research is focusing on understanding of the most fundamental aspects of the global climate system: Why do we have an icehouse and not a greenhouse world today? If we understand this we will have a greater ability to distinguish natural from anthropogenic changes in climate. A question we deal with is establishing a robust empirically derived information on what actually occurs on an Earth-like planet in connection with dramatic changes in the composition of the atmosphere. This empirically derived information can be used to calibrate model scenarios.

Figure 1. Analysis of the temperature oscillations in the Geological Eras ( after Scotese 2002 and Ruddiman 2001.

Our group is unique internationally since we investigate forcing and feedback mechanisms of the global carbon cycle by applying a multidisciplinary and integrated approach from a deep time perspective. We utilize not only a broad range of geochemical methods, including stable isotope measurements, but also micro-paleontological studies covering depositional environments from terrestrial to deep oceanic systems globally. We develop new methods to reconstruct changes in astronomical parameters, for example, by extracting meteoritic spinel mineral grains from condensed sediments. During the course of the LUCCI-project our group has reached several significant results and breakthroughs. Our research within WP4 continues to focus also on the signatures and mechanisms of the long-term carbon budget (figs. 1-2) and the significance of different carbon sinks and sources through geological time as it is a prerequisite for understanding present carbon cycling. Changes in the Earth’s climate and of life on Earth are preserved in the rock

96

LUCCI Annual Report – work package 4 record and the geological record provides an invaluable archive of information on global environmental change, both rapid and long-term.

Figure 2. Temperature assessments based on oxygen and carbon isotopes from the end Cretaceous to present (from Zachos et al., 2001, Science 292, 686-693)

What we achieved in 2014 The research performed during 2014 has been very successful and productive. We have provided new information within a broad range of fields related to climate, spanning over 100 millions of years, in marine and in terrestrial environments but also by providing additional knowledge on extra-terrestrial causes behind climate patterns. We have disseminated the results on various major international conferences including the International Paleontological Congress in Mendoza, Argentina (Fig. 3), The Goldschmidt meeting, EGU and the paleobotanical meeting in Padua, Italy to mention some. We have continued our efforts focussing on the signatures and mechanisms of the long-term carbon budget and the significance of different carbon sinks and sources through geological time. Highlights include the finding of an exquisitely preserved Jurassic ecosystem in southern Sweden where Jurassic lava has preserved entire vegetation landscapes allowing for climate reconstructions. The preservation is such that the cell nuclei even with chromosomes are preserved, 160 million year old cells. The finding, published in Science in March 2014 (Bomfleur et al. 2014) rendered major attention from the scientific community and beyond. New data on the earliest land plants and their invasion of land have likewise provided new information documenting the timing of land plant appearances and its impact on atmospheric CO2 and O2 levels and global climate (Mehlqvist et al. 2014), and most importantly the oldest traces of land plants from the geological record of our paleocontinent – Baltica.

97

LUCCI Annual Report – work package 4

Figure 3. Central Argentina; the Late Triassic deposits of the Ischigualasto Formation represent arid conditions with extreme green-house warming. It also contains some of the oldest known dinosaur remains.

We have continued with the studies of the Paleocene-Eocene thermal maximum (Figs, 2 and 4) with focus on two issues: what is the triggering mechanism for this greenhouse catastrophe, and at what rate did the composition of the atmosphere change with dramatic climatic effects (Pujalte et al., 2014 and Pujalte and Schmitz, 2014). This research is now being very actively pursued in the Tremp Basin in northern Spain. We have made considerable progress in our attempts to develop new methods to link environmental changes on Earth to large-scale astronomical processes. The Astrogeobiology Laboratory at Medicon Village is now fully operable. One hypothesis to be tested is whether the major changes between green house and ice house conditions on Earth are related to instability of the solar system, rather than to changes in the position of the continents or atmospheric CO2 levels. We link the histories of the Earth and the solar system by studying extraterrestrial minerals in Earth's sediments. Much of the work now deals with methodology development, i.e. creating an understanding of what different chemical and isotopic signatures of the extraterrestrial minerals mean (e.g. Meier et al., 2014a,b). We also try to understand the possible relation between the Great Ordovician Biodiversification Event in the mid-Ordovician with dramatic events in the asteroid belt at the same time (Schmitz et al., 2014). Our studies of the relation between the evolution of life and the carbon cycle during the Paleozoic have led to important new insights, for example, about the global anoxia developing in connection with the Cambrian-Ordovician bioevent (Terfelt et al., 2014). The development of ice ages in relation to changes in the carbon cycle in the Hirnantian at the Ordovician-Silurian boundary has also been studied by us (Bergström et al., 2014).

98

LUCCI Annual Report – work package 4

Figure 4. The Paleocene-Eocene boundary in terrestrial environments in the Pyrenees is charac-terized by a change from soils to a widespread conglomerate, reflecting the dramatic change in hydrological regime in connection with the development of super-greenhouse conditions

Selected publications 2014 [1] Badawy, A.S., Mehlqvist, K, Vajda, V., Ahlberg, P., Calner, M. (2014) Late Ordovician (Katian) spores in Sweden – oldest land plant remains from Baltica. GFF a Scandinavian Journal of Earth Sciences 136, 16–21. [2] Bercovici, A., Cui, Y., Forel, M.B., Yu, J., Vajda, V. (2014) Terrestrial paleoenvironment characterization across the Permian–Triassic boundary in South China. Journal of Asian Earth Sciences 98, 225-246. [3] Bergström, S.M., Eriksson, M.E., Young, S., Ahlberg, P., Schmitz, B. (2014) Hirnantian (latest Ordovician) 13C chemostratigraphy in southern Sweden and globally: A refined integration with the graptolite and conodont zone successions. GFF 136:355-386. [4] Bomfleur, B., McLoughlin, S. & Vajda, V. (2014) Fossilized nuclei and chromosomes reveal 180 million years of genomic stasis in Royal Ferns. Science 343(6177), 1376–1377. [5] McLoughlin, S., Jansson, I.-M., Vajda, V. (2014) Megaspore and microfossil assemblages reveal diverse herbaceous lycophytes in the Australian Early Jurassic flora. Grana 53, 22–53. [6] Mehlqvist, K., Larsson, K., Vajda, V. (2014) Linking upper Silurian terrestrial and marine successions-Palynlogical study from Skåne, Sweden. Review of Paleobotany and Palynology 202, 1–14. [7] Mehlqvist, K., Larsson, K., Vajda, V. (2014) Linking upper Silurian terrestrial and marine successions—Palynological study from Skåne, Sweden. Review of Palaeobotany and Palynology 202, 1-14 [8] Meier, M.M., Schmitz, B., Alwmark, C., Trappitsch, R., Maden, C., Wieler, R. (2014) He and Ne in individual chromite grains from the regolith breccia Ghubara (L5): Exploring the history of the L chondrite parent body regolith. Meteoritics & Planetary Science 49:576-594.

99

LUCCI Annual Report – work package 4 [9] Meier, M.M.M., Schmitz, B. Lindskog, A., Maden, C. Wieler, R. (2014) Cosmic-ray exposure ages of fossil micrometeorites from mid-Ordovician sediments at Lynna River, Russia. Geochimica et Cosmochimica Acta 125, 338-350. [10]Pujalte, V., Schmitz, B. (2014) Comment on ”Magnitude and profile of organic carbon isotope records from the Paleocene-Eocene Thermal Maximum: Evidence from northern Spain” by Manners et al. EPSL 376:220-230. Earth and Planetary Science Letters 395:291-293. [11]Pujalte, V.P., Schmitz, B., Baceta, J.I. (2014) Sea-level changes across the Paleocene-Eocene boundary in the Spanish Pyrenees and other basins, and their possible link with the Paleocene-Eocene thermal maximum and North Atlantic magmatism. Palaeogeography, Palaeoecology, Palaeoclimatology 393:45-60. [12]Schmitz, B., Huss, G.R. Meier, M.M.M. Peucker-Ehrenbrink, B. Church, R.P. Cronholm, A. Davies, M.B.. Heck, P.R. Johansen, A Keil, K. Kristiansson, PRavizza, . G. Tassinari, M. Terfelt F. (2014) A fossil winonaite-like meteorite in Ordovician limestone: A piece of the impactor that broke up the L-chondrite parent body? Earth and Planetary Science Letters 400, 145-152. [13]Steinthorsdottir,M., Vajda, V., (2014) Early Jurassic (late Pliensbachian) CO2 concentrations based on stomatal analysis of fossil conifer leaves from eastern Australia, Gondwana Research27, 932-939. [14]Terfelt, T., Eriksson, M.E. Schmitz, B. (2014) The Cambrian-Ordovician transition in dysoxic facies in Baltica – diverse faunas and carbon isotope anomalies. Palaeogeography, Palaeoecology, Palaeoclimatology 394:59-73. [15]Vajda, V., & Bercovici, A., (2014) The global vegetation pattern across the Cretaceous-Paleogene mass-extinction interval: A template for otehr extinction events Global and Planetary Change 122, 29–49. [16]Vajda, V., Ocampo, A., Ferrow, E., Bender Koch (2014) Nano particles as the primary cause for long-term sunlight suppression at high southern latitudes following the Chicxulub impact –evidence from ejecta deposits in Belize and Mexico Gondwana Research, 27, 1079–1088

100

LUCCI Annual Report – work package 4

First evidence of the Cretaceous decapod crustacean Protocallianassa from Sweden and its implications for paleoenvironment and climate Elisabeth Einarsson (a), Aron Praszkier (a) & Vivi Vajda (a) (a) Dept of Geology, Lund University

The Kristianstad Basin The Kristianstad Basin in the southernmost Swedish province of Skåne hosts a richly fossiliferous succession of Upper Cretaceous marine deposits (ca. 67 Mya). During the mid-Late Cretaceous, repeated transgressive episodes occurred in what today represents southern Scandinavia which created islands in the Kristianstad Basin that were surrounded by shallow seas forming paralic environments adjacent to the deeper Tethyan shelf to the southwest (Vajda & Solakius 1999, Larsson et al. 2000). Today these palaeoenvironments are represented by several well-known fossil deposits, such as those exposed at Ignaberga, Ullstorp, Ivö Klack and Åsen (Fig. 1). The successions at Åsen and Ivö Klack and are especially rich in vertebrate remains including sharks, rays, actinopterygian fishes, mosasaurs, plesiosaurs, marine and freshwater turtles, aquatic birds, and rare non-avian dinosaur bones. Coeval invertebrate assemblages comprise bivalves, cephalopods, gastropods, brachiopods, bryozoans, and echinoderms mainly represented by sea-urchin (Einarsson et al. 2010).

Figure 1. Schematic geological map of southern Sweden, Skåne, with focus on the Kristianstad Basin. Studied localities are marked.

101

LUCCI Annual Report – work package 4 The material for this study was recovered from the earliest late Campanian succession at Åsen and the latest early Campanian succession at Ivö Klack localities (Figure 1). For this study eighteen calcified chelipeds of Protocallianassa faujasi were measured based on their characters (Figure 2). They all showed to belong to the burrowing ghost shrimp, Protocallianassa faujasi, providing the first evidence of this decapod species from Sweden. The fossils occur in successions of the Creataceous (Campanian) both at Ivö Klack and Åsen in the Kristianstad Basin of northeastern Skåne. Based on the burrowing lifestyle of modern mud shrimps, we interpret these nodules as infilled burrow chambers. The low abundance of molluscs within the Protocallianassa beds is also consistent with analogous extant communities, indicating that a similar ecologically exclusive relationship ruled within the Late Cretaceous shallow marine ecosystems in a subtropical climate.

Conclusions The present study adds to the distribution of Late Cretaceous Protocallianassa extending its northern range into Scandinavia. At the close of the Cretaceous, Protocallianassa was replaced by representatives of other subfamilies which might be linked to the Cretaceous–Paleogene extinction event, the consequence of an asteroid impact in Yucatan, Mexico 66 Ma (Vajda & Bercovici, 2014). The turnover amongst Protocallianassinae appears to have been more severe in Europe where the youngest record is from the Maastrichtian in the Netherlands and Belgium (Swen et al. 2001). The replacement by Corallianassa and other taxa appears to have been delayed in high-latitude Southern Hemisphere settings which consistent with the extinction patterns seen in the post-impact vegetation, with more severe extinctions in Northern Hemisphere floras (Vajda & Bercovici 2014).

Figure 2. Selected Cheliped of Protocallianassa faujasi from the Ivö klack (Blaksudden) locality, scale bars = 1 cm.

Comparison with modern ecosystems The extant mud shrimp Neotrypaea californiensis excavates burrows up to 24 cm deep, and can incorporate as many as seven chambers. This is consistent with the Protocallianassa fossils from Åsen and Ivö Klack, which comprise isolated chelipeds (often represented by internal casts) but no recognizable cephalothoracic or abdominal components. Moreover, they occur preserved within carbonate-cemented nodules that resemble chambered hollows

102

LUCCI Annual Report – work package 4 of thalassinoidean burrows, presumably infilled and cemented by CaCO2 after decomposition of the remainder of the carapace and soft tissues. In modern ecosystems, Thalassionides species inhabit the littoral zone down to a shelf depth of 100 metres (Müller 1984). This is especially consistent with the shallow marine depositional setting for Åsen. The abundance of decapod remains within the informal B. balsvikensis zone (particularly Balsvikensis yellow) might explain the concomitant dearth of bivalves and other molluscs.

References [1] Einarsson, E., Lindgren, J., Kear, B. & Siverson, M., 2010: Mosasaur bite marks on a plesiosaur propodial from the Campanian (Late Cretaceous) of southern Sweden. GFF 132, 1–6. [2] Larsson, K., Solakius, N. & Vajda, V. 2000: Foraminifera and palynomorphs from the greensand-limestone sequences (Aptian-Coniacian) in southwestern Sweden. Geologische Jahrbuch fur Geologie und Paläontologie 216, 277–312. [3] Müller, P. 1984: Decapod Crustacea of the Badenian. Geologica Hungarica, Series Palaeontologica 42, 3–317. [4] Vajda, V. & Solakius, N. 1999: Palynomorphs, foraminifera and calcisphaeres from the greensand-limestone transition at Arnager, Bornholm: evidence for the late Cenomanian to early Coniacian transgression. GFF 121, 281– 286. [5] Vajda, V. & Bercovici, A. 2014: The global vegetation pattern across the Cretaceous–Paleogene mass-extinction interval – an integrated global perspective. Global and Planetary Change 122, 29–49.

103

LUCCI Annual Report – work package 4

The Claret Conglomerate: Evidence of an abrupt change in the hydrological cycle at the Paleocene- Eocene Boundary Birger Schmitz (a), Victoriano Pujalte (b) (a) Astrogeobiology Laboratory, Dept. of Physics, Lund University (b) Dept. of Stratigraphy and Paleontology, University of the Basque Country, Bilbao, Spain

A prominent increase in atmospheric CO2 at the Paleocene-Eocene boundary, ca. 55 Ma ago, led to the warmest Earth of the Cenozoic for ~100 ka High-resolution studies of continental flood-plain sediment records across this boundary can provide crucial information on how the hydrological cycle responds to rapidly changing CO2. Here we show from continental records across the Paleocene-Eocene boundary in the Spanish Pyrenees, a subtropical paleosetting, that during the early, most intense phase of CO2 rise, normal, semiarid coastal plains with few river channels of 10–200 m width were abruptly replaced by a vast conglomeratic braid plain, covering at least 500 km2 and most likely more than 2000 km2. This event is today represented by the Claret Conglomerate. The vast braid plain is interpreted as the proximal parts of a megafan. Carbonate nodules in the megafan deposits attest to seasonally dry periods and together with megafan development imply a dramatic increase in seasonal rain and an increased intraannual humidity gradient. The megafan formed over a few thousand years to ~10 ka directly after the PaleoceneEocene boundary. Only repeated severe floods and rainstorms could have contributed the water energy required to transport the enormous amounts of large boulders and gravel of the megafan during this short time span. The findings represent evidence for considerable changes in regional hydrological cycles following greenhouse gas emissions.

Figure 1. The Claret Conglomerate (upper right) that formed over vast areas in northern Spain at the onset of the "supergreenhouse" catastrophe at the Paleocene-Eocene boundary. Only a very dramatic and major change in the composition of Earth's atmosphere could have initiated the changes in the hydrological cycle leading to the deposition of the Claret Conglomerate.

We now try to establish new carbon isotopic data for plant remains across the transition into the Claret Conglomerate in order to further estimate how fast the hydrological cycle responded to increasing CO2 levels. The observed changes at the Paleocene-Eocene boundary in the Spanish Pyrenees are similar to the projected future changes in the hydrological cycle at corresponding latitudes in a future greenhouse world.

104

LUCCI Annual Report – work package 4 References

[1] Pujalte, V, Schmitz, B., 2014. Comment on "Magnitude and profile of organic carbon isotope records from the Paleocene-Eocene Thermal Maximum: Evidence from northern Spain". Earth and Planetary Science Letters 395:291293. [2] Pujalte. V., Schmitz, B., Baceta, J.I., 2014. Sea-level changes across the Paleocene-Eocene interval in the Spanish Pyrenees, and their possible relationship with North Atlantic magmatism. Palaeogeography, Palaeoclimatology, Palaeoecology 393:45-60.

105

LUCCI Annual Report – work package 4

A multitype asteroid shower in the Late Eocene: Does this indicate an astronomical trigger of the late Cenozoic ice house world? Birger Schmitz (a,b,c), Samuele Boschi (a), Anders Cronholm (a) Philipp R. Heck (c), Simonetta Monechi (d), Alessandro Montanari (e), Fredrik Terfelt (a) (a) (b) (c) (d) (e)

Astrogeobiology Laboratory, Dept. of Physics, Lund University Hawai'i Institute of Geophysics and Planetology, Univ. of Hawai'i at Manoa, USA Robert A. Pritzker Center for Meteoritics and Polar Studies, The Field Museum of Natural History, Chicago, USA Dept of Earth Sciences, Florence University, Italy Geological Observatory of Coldigioco, Frontale di Apiro, Macerata, Italy

The onset of Earth's present icehouse climate in the Late Eocene coincides with astronomical events of enigmatic causation. At ~36 Ma ago the 90-100 km large Popigai and Chesapeake Bay impact structures formed within ~10-20 ka. Enrichments of 3He in coeval sediments also indicate high fluxes of interplanetary dust to Earth for ~2 Ma (Farley et al., 1998). Additionally, several medium-sized impact structures are known from the Late Eocene. Here we report from sediments in Italy the presence of abundant ordinary chondritic chromite grains (63-250 m) associated with the ejecta from the Popigai impactor. The grains occur in the ~40 cm interval immediately above the ejecta layer. Element analyses show that grains in the lower half of this interval have an apparent H-chondritic composition, whereas grains in the upper half are of L-chondritic origin (Schmitz et al., 2015). The grains most likely originate from the regoliths of the Popigai and the Chesapeake Bay impactors, respectively. These asteroids may have approached Earth at comparatively low speeds, and regolith was shed off from their surfaces after they passed the Roche limit. The regolith grains then settled on Earth some 100 to 1000 years after the respective impacts. Further neon and oxygen isotopic analyses of the grains can be used to test this hypothesis. If the Popigai and Chesapeake Bay impactors represent two different types of asteroids one can rule out previous explanations of the Late Eocene extraterrestrial signatures invoking an asteroid shower from a single parent-body breakup. Instead a multi-type asteroid shower may have been triggered by changes of planetary orbital elements. This could have happened due to chaos-related transitions in motions of the inner planets or through the interplay of chaos between the outer and inner planets. Asteroids in a region of the asteroid belt where many ordinary chondritic bodies reside, were rapidly perturbed into orbital resonances. This led to an increase in small to medium-sized collisional breakup events over a 2-5 Ma period. This would explain the simultaneous delivery of excess dust and asteroids to the inner solar system. Independent evidence for our scenario are the common cosmic-ray exposure ages in the range of ca. 33-40 Ma for recently fallen H and L chondrites. The temporal coincidence of gravity disturbances in the asteroid belt with the termination of ice-free conditions on Earth after 250 Ma is compelling. We speculate that this coincidence and a general correlation during the past 2 Ga between K-Ar breakup ages of parent bodies of the ordinary chondrites and ice ages on Earth suggest that there may exist an astronomical process that disturbs both regions of the inner asteroid belt and Earth's orbit with a potential impact on Earth's climate.

106

LUCCI Annual Report – work package 4

Figure 1. Profiles for the Massignano section of extraterrestrial 3He concentrations (Farley et al., 1998) and the total number of recovered EC grains per kg sediment. Indicated in numbers on the latter curve are the masses in kilogram of the samples searched for EC grains.

References [1] Farley, K.A., Montanari, A., Shoemaker, E.M., Shoemaker, C.S., 1998. Geochemical evidence for a comet shower in the Late Eocene. Science 280, 1250-1253. [2] Schmitz, B. et al., 2015. Fragments of Late Eocene Earth-impacting asteroids linked to disturbance of asteroid belt. Earth and Planetary Science Letters, in press.

107

LUCCI Annual Report – work package 4

The carbon cycle and its relation to astronomical signals during the Triassic-Jurassic mass extinction event (201 Mya) – studies from the Junggar Basin, China Vivi Vajda (a), Jingeng Sha (b), Yanhong Pan (b), Paul E. Olsen (c) (a) Department of Geology, Lund University, 223 62 Lund, Sweden (b) State Key Laboratory of Palaeobiology and Stratigraphy, Nanjing Institute of Geology and Paleontology, Nanjing 210008, China

(c) Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY 10968

Our understanding of Triassic and Early Jurassic (ca. 210-180 Mya) high-latitude climate, biotic evolution, mass extinction, and geochronology is very limited in contrast to the knowledge on modern climate systems. This relatively poor resolution impairs an understanding of the basic patterns of Earth system function during the early Mesozoic. Besides, integral to the long-term chaotic behavior of the Solar System are the secular resonances of the planets, particularly for the inner Solar System. Accurate and detailed dating of sedimentary succession is crucial for any paleoenvironmental study especially concerning short term catastrophic events in Earth history (Sha et al., 2015). Here we study a succession from the Junggar Basin in North Eastern China (Figure 1.). The thick, non-marine, early Mesozoic sequence of the Junggar Basin of Western China comprises >11 km of largely non-marine Late Paleozoic to Cenozoic strata deposited in the northwestern-most of the “walled basins” of China. In contrast to the present, Late Triassic–Early Jurassic high latitude regions had warm, humid climates according to the presence of broad-leaf gymnosperm macrofossil assemblages from both formations and the presence of coal, consistent with many early Mesozoic northern hemisphere high-latitude sedimentary basins (Sha et al., 2011).

Figure 1. Map of the Junggar and Tarim basins, Northwest China

108

LUCCI Annual Report – work package 4 LITH Index Cyclicity is visually evident at multiple scales in the studied succession as alternations of organic-rich mudstones with sandstone and conglomerate beds. To quantify this cyclicity and examine its possible periodicity in thickness, we constructed a scale of lithologies based largely on grain size (Figure 2), termed the LITH index, which, in a broad way, can be interpreted as a proxy of the degree of flooding of the land surface during deposition.

Temporal assessments of the latest Triassic and earliest Jurassic sediments within the Junggar Basin The dating of the successions was performed by palynology (pollen analysis) based on forty samples spanning the boundary of the Haojiagou and Badaowan formations. Our analysis focuses on the End Triassic Event (ETE). Wellpreserved pollen and spore assemblages of medium diversity were identified including 60 species of fossil pollen and spore taxa were identified in this study, together with dinoflagellates and green algae. Typical Triassic elements persist up to bed 52 (Figure 3), including Lunatisporites rhaeticus and Limbosporites spp., whereas the abundant occurrence of Retritriletes semimuris and Retritriletes austroclavatiditesin bed 53 suggests a Jurassic (Hettangian age) and the end of the ETE interval.

Figure 2. Spectral results from the clipped LITH scale of the Triassic-Jurassic boundary interval at the study site. In the evolutive FFT, white is the highest relative power and black is the lowest. In the FFT spectra, high-amplitude peaks are labeled with their periods in meters and ky (kiloyears) assuming recognition of the 405-ky cycle. In the MTM spectra, only those peaks in spectral power that have f-test significance above the 0.9 level and high amplitude are labeled, and they are labeled in meters and kiloyears based on recognition of the 405-ky level.

109

LUCCI Annual Report – work package 4 Discussion Analysis of the LITH proxy of environmental change shows that an astronomical signal in which obliquity is dominant can be extracted from lacustrine strata of the high-latitude (~60º N) Junggar Basin straddling the end-Triassic extinction and Triassic-Jurassic boundary (Figure 2). This is dramatically different from the climate precessiondominated continental tropics. In combination, the data are incompatible with published astronomical solutions for the Triassic-Jurassic in phase and amplitude, consistent with chaotic behavior of the Solar System whereas, at the same time, the Earth-Mars orbital resonance seems to have been in today’s two-to-one ratio of eccentricity to inclination, providing a constraint for the Earth-Mars orbital resonance for around 201 Ma. With the prospect of the acquisition of better temporally resolved records from deeper lake settings in the Junggar and other basins, the use of more directly climate-sensitive proxies, and additional exploration of the paleobiological context of the strata, it will be possible to test these findings, constraining the history of Solar System chaos, during this transitional time in Earth history.

Significance Geological records of paleoclimate provide the only constraints on Solar System orbital solutions extending beyond the ∼50-Ma limit imposed by chaotic diffusion. Examples of such constraints are coupled high and low latitude, Triassic– Jurassic (∼198–202 Ma) sedimentary cyclicity in coal-bearing outcrops from the ∼60° N paleolatitude Junggar Basin (Western China), and contemporaneous tropical basins. Analysis reveals climate variability dominated by obliquityscale cyclicity in the Junggar Basin and precession-scale cyclicity in the tropics. These results demonstrate the opportunity for developing a new class of solutions grounded by geological data extending hundreds of millions of years into the geologic past.

References [1] Sha, J., Olsen, P.E., Yanhong Pan, Y., Xu, D., Wang Y., Zhang, X., Vajda, V. 2015: Triassic-Jurassic climate in continental high-latitude Asia was dominated by obliquity-paced variations (Junggar Basin, Urumqi, China), 112, PNAS, 3624-3629 [2] Sha, J., Vajda, V., Pan, Y., Larsson, L., Wang, Y., Cheng, X.J, Deng, S., Yao, X., Chen, S., Zhang, X. & Peng, B., 2011: The Stratigraphy of the Triassic−Jurassic boundary successions of the southern margin of Junggar Basin, northwestern China. Acta Geolologia Sinica 85, 421–436.

110

WORK PACKAGE 5

Modeling of present and future processes and conditions with a background of Holocene and pre-Holocene scenarios and data

LUCCI Annual Report – work package 5

112

LUCCI Annual Report – work package 5

WP 5: Progress Report Chiara Molinari & Ben Smith

Researchers involved: Anders Ahlström, Jonas Ardö, Torben R. Christensen, Jennifer Grant, Anna Maria Jönsson, Wolfgang Knorr, Fredrik Lagergren, Harry Lankreijer, Dörte Lehsten, Veiko Lehsten, Mats Lindeskog, Laurent Marquer, Paul Miller, Mikhail Mishurov, Chiara Molinari, Anne Nielsen, Jörgen Olofsson, Anneli Poska, Marko Scholze, Jonathan Seaquist, Ben Smith, Vivi Vajda.

Graduated students involved: Kerstin Baumanns, Jan Blanke, Nitin Chaudhary, Andrew McRobert, Niklas Olén, Stefan Olin, Cecilia Olsson, Florian Sallaba, Minchao Wu, Zhendong Wu, Wenxin Zhang.

Work Package 5 focuses on the study of the role of climate and land use on ecosystem processes in the past, present and future at different spatial scales. Our methodology combines models development, data-model evaluations and comparisons, and the analysis of uncertainties. Since the last LUCCI Annual Report, the following WP5-associated former PhD students successfully defended their PhD: Stefan Olin, “Ecosystems in the Anthropocene: the role of cropland management for carbon and nitrogen cycle processes” - 8th of May 2015; Wenxin Zhang, “The role of biogeophysical feedbacks and their impacts in the arctic and boreal climate system” – 6th of February 2015; Cecilia Olsson, “Tree phenology modelling in the boreal and temperate climate zones: Timing of spring and autumn events” – 14th of November 2014.

Since 2015-01-14, the WP5-associated PhD student Niklas Olén started his shared leadership (with Ylva Persson from WP1) of the LUCCI R3I group. Representing some of the world´s leading scientific minds, the highly Cited Researchers list 2014 ranked Professor Emeritus Martin Sykes among the top 1% most cited researchers for their subject field. For this second time period of LUCCI (2015-2018), the common decision is to focus our attention on two main research topics:

(1) Incorporation of LPJ-GUESS ecosystem model into version 3 of the EC-Earth climate model. Main aim of researchers in WP5 is to develop EC-Earth into a state-of-the-art Earth System Model (ESM) ahead of its participation in the next Coupled Model Intercomparison Project, Phase 6 (CMIP6). CMIP6 will be a major contribution to and data source for the next IPCC report. Within EC-Earth Version 3, LPJ-GUESS will dynamically update the vegetation composition and land surface properties in response to the simulated climate. It will also send terrestrial carbon (C) fluxes to the TM5 atmospheric chemistry model for transport and mixing with C fluxes from the PISCES model of ocean biogeochemistry currently embedded in EC-Earth's ocean model, NEMO 3.6, altering and feeding back to the simulated climate through its influence on atmospheric CO2 composition. Thus, EC-Earth will give us a better understanding of the biogeophysical and biogeochemical feedbacks at play in the climate system. Summer and Autumn 2015 will be devoted to testing, tuning and evaluation of EC-Earth 3, before CMIP6 runs begin in late 2015 and early 2016. Related to this topic, LUCCI researchers Ben Smith and Paul Miller are contributing as Lund University PI’s to the Horizon 2020 project CRESCENDO, commencing in Autumn 2015, that provides a platform for ESM development and evaluation efforts among all ESM groups in Europe contributing to CMIP6.

(2) Understanding the role of fire in the terrestrial biosphere and of fire responses to biotic and abiotic processes and identification of major drivers of biomass burning during the Holocene by a comparison/combination between independent empirical observation and model projections. 113

LUCCI Annual Report – work package 5 More in detail, during the last year, this branch of WP5 lead by Chiara Molinari focused on the study of main dynamics and drivers of boreal forests fire regimes in Fennoscandia and North America during the Holocene. The decision to concentrate the attention on the boreal biome was mainly due to the fact that fire is and has been one of the most important disturbances in these regions and because the boreal forests are most likely to be significantly affected by future global warming, with a variation in fire regime. In order to reconstruct temporal and spatial patterns of boreal forests fire dynamics during the Holocene a little bit less than 140 sedimentary charcoal records were selected. For an estimation of the influence of climate change on boreal fire, fire dynamics and temperature reconstructions based on pollen data have been compared. Furthermore, the correlation between variations in biomass burning and changes in main forest composition (based on pollen data) has been statistically evaluated in order to assess the effect of vegetation on fire. The next step will be to combine palaeo-data and models for providing more a complete spatial coverage of past fire activity. Future studies should also consider the transformation of palaeo-proxies based on charcoal data (now limited to quantitative trends) and model output variables to improve the comparability between observations and simulations.

Publications 2014-2015: [1] Abdi, A.M., Seaquist, J.W., Tenenbaum, T.E., Eklundh, L. & Ardö, J., 2014. The supply and demand of net primary production in the Sahel. Environmental Research Letters 9, 094003. [2] Ahlström, A., Raupach, M.R., Schurgers, G., Smith, B., Arneth, A., Jung, M., Reichstein, M., Canadell, J.P., Friedlingstein, P., Jain, A.K., Kato, E., Poulter, B., Sitch, S., Stocker, B.D., Viovy, N., Wang, Y.-P., Wiltshire, A., Zaehle, S. & Zeng, N. 2015. The dominant role of semi-arid ecosystems in the trend and variability of the land CO2 sink. Science 348: 895-899. [3] Ahlström, A., Xia, J., Arneth, A., Luo, Y. & Smith, B. 2015. Importance of vegetation dynamics for future terrestrial carbon cycling. Environmental Research Letters 10: 054019. [4] Åkesson, C., Nielsen, A.B., Broström, A., Persson, T., Gaillard, M.-J. & Berglund, B., 2015. From landscape description to quantification: a new generation of reconstructions provides novel perspectives on Holocene regional landscapes of SE Sweden. The Holocene 25, 78-93. [5] Bentz, B.J. & Jönsson, A.M., 2015. Modeling bark beetle response to climate change. Chapter 13 In. Bark Beetles – biology and ecology of native and invasive species Eds Vega, F.E. & Hofstetter, R.W. Academic Press, Elsevier (San Diego) pp. 533- 553. [6] Blessing, S., Kaminski, T., Lunkeit, F., Matei, I., Giering, R., Köhl, A., Scholze, M., Fraedrich, K. & Stammer, D., 2014. Testing variational estimation of process parameters and initial conditions of an Earth System Model. Tellus A, 66. [7] Clear, J., Seppa, H., Kuosmanen, N., Molinari, C., Lehsten, V., Allen, K. & Bradshaw, R., 2015. Holocene fire dynamics in Fennoscandia. Geophysical Research Abstract, 17: EGU2015-13785. eISSN 1607-7962. [8] Clear, J.L., Molinari, C. & Bradshaw, R.H.W., 2014. Holocene fire review of Fennoscandia: A synthesis of published charcoal and fire scar records. International Journal of Wildland Fire 23, 781–789. [9] De Kauwe, M.G., Medlyn, B.E., Zaehle, S., Walker, A.P., Dietze, M.C., Wang, Y.-P., Luo, Y., Jain, A.K., El-Masri, B., Hickler, T., Wårlind, D., Weng, E., Parton, W.J., Thornton, P.E., Wang, S., Prentice, I.C., Asao, S., Smith, B., McCarthy, J.M., Iversen, C.M., Hanson, P.J., Warren, J.M., Oren, R. & Norby, R.J., 2014. Where does the carbon go? A model-data intercomparison of vegetation carbon allocation and turnover processes at two temperate forest freeair CO2 enrichment sites. New Phytologist 203, 883-899. [10] Edvardsson, J., Poska, A., van der Putten, N., Rundgren, M., Linderson, H. & Hammarlund, D., 2014. Late-Holocene expansion of a south Swedish peatland and its impact on marginal ecosystems: Evidence from dendrochronology, peat stratigraphy and palaeobotanical data. The Holocene 24, 466 - 476. [11] Fisher, J. B., Sikka, M., Oechel, W.C., Huntzinger, D.N., Melton, J. R., Koven, C.D., Ahlström, A., Arain, A.M., Baker, I., Chen, J. M., Ciais, P., Davidson, C., Dietze, M., El-Masri, B., Hayes, D., Huntingford, C., Jain, A., Levy, P.E., Lomas, M. R., Poulter, B., Price, D., Sahoo, A.K., Schaefer, K., Tian, H., Tomelleri, E., Verbeeck, H., Viovy, N., Wania, R., Zeng, N. & Miller, C.E, 2014. Carbon cycle uncertainty in the Alaskan Arctic. Biogeosciences 11, 4271-4288. [12] Gaillard, M.-J., Kleinen, T., Samuelsson, P., Nielsen, A.B., Bergh, J., Kaplan, J., Poska, A., Sandström, C., Strandberg, G., Trondman, A.K. & Wramneby, A., 2015. Causes of Regional Change—Land Cover. The BACC II Author Team (Toim.). Second Assessment of Climate Change for the Baltic Sea Basin (453 - 477). Springer.

114

LUCCI Annual Report – work package 5 [13] Grant, J., Wigneron, J.-P., Williams, M., Scholze, M. & Kerr, Y., 2014. Working towards a global-scale vegetation water product from SMOS optical depth, Geoscience and Remote Sensing Symposium (IGARSS) Proceedings, 2014 IEEE International, pp. 286-289, 13-18 July 2014. [14] Haverd, V., Smith, B., Nieradzik, L.P. & Briggs, P.R. 2014. A stand-alone tree demography and landscape structure module for Earth system models: integration with inventory data from temperate and boreal forests. Biogeosciences 11, 4039-4055. [15] Jamali, S., Jönsson, P., Eklundh, L., Ardö, A. & Seaquist, J.W., 2015. Detecting changes in vegetation trends using time series segmentation. Remote Sensing of Environment 156, 182-195. [16] Jamali, S., Seaquist, J.W., Eklundh, L. & Ardö, J., 2014. Automated mapping of vegetation trends with polynomials using NDVI imagery over the Sahel. Remote Sensing of Environment 141, 79-89. [17] Jönsson, A.M. & Swartling, Å., 2014. Stakeholder engagement in climate change adaptation research - Experiences from Swedish forestry. Society and Natural Resources DOI: 10.1080/08941920.2014.906013. [18] Jönsson, A.M., Anderbrant, O., Holmér, J., Johansson, J., Schurgers, G., Svensson, G. & Smith H.G., 2015. Enhanced science-stakeholder communication to improve ecosystem model performances for climate change impact assessments AMBIO 44, 249-255. [19] Jönsson, A.M., Lagergren, F. & Smith, B. 2015. Forest management facing climate change—an ecosystem model analysis of adaptation strategies. Mitigation & Adaptation Strategies for Global Change 20, 201-220. [20] Kebede, A.S., Dunford, R., Mokrech, M., Audsley, E., Harrison, P.A., Holman, I.P., Nicholls, R.J., Rickebusch, S., Rounsevell, M.D.A., Sabaté, S., Sallaba, F., Sanchez, A., Savin, C., Trnka, M. & Wimmer, F., 2015. Direct and indirect impacts of climate and socio-economic change in Europe: a sensitivity analysis for key land- and waterbased sectors. Climatic Change 128, 261-277. [21] Kemp, S., M. Scholze, T. Ziehn, T. & Kaminski, 2014. Limiting the parameter space in the Carbon Cycle Data Assimilation System (CCDAS). Geosci. Model Dev. 7, 1609-1619. [22] Kross, A.S.E., Roulet, N.T., Moore, T.R., Lafleur, P.M., Humphreys, E.R., Seaquist, J.W., Flanagan, L.B. & Aurela, M., 2014. Phenology and its role in carbon dioxide exchange processes in northern peatlands. Journal of Geophysical Research: Biogeosciences 119, 1370-1384. [23] Lindeskog, M., Arneth, A., Bondeau, A., Waha, A., Seaquist, J.W., Olin, S. & Smith, B., 2014. Implications of accounting for land use in simulations of ecosystem services and carbon cycling in Africa. Earth System Dynamics 4, 385-407. [24] Marquer, L., Gaillard, M.-J., Sugita, S., Trondman, A.-K., Mazier, F., Nielsen, A.B., Fyfe, R.M., Odgaard, B.V, Alenius, T., Birks, H.J.B, Bjune, A.E, Christiansen, J., Dodson, J., Edwards, K.J., Giesecke, T., Herzschuh, U., Kangur, M., Lorenz, S., Poska, A., Schult, M. & Seppä, H., 2014. Holocene changes in vegetation composition in northern Europe: why pollen-based quantitative reconstructions matter. Quaternary Science Reviews 90, 199-216. [25] Molinari, C., Lehsten, V., Blarquez, O., Clear, J.L., Carcaillet, C. & Bradshaw, R.H.W., 2015. Main dynamics and drivers of boreal forests fire regimes during the Holocene. Geophysical Research Abstract, 17: EGU2015-428. eISSN 1607-7962. [26] Nielsen, A.B., Poska, A., Åkesson, C., Broström, A., 2014. Modelling past land use changes and their effects on carbon pools -Using REVEALS and LPJ-Guess. 31st Nordic Geological Winter Meeting, Lund, Sweden, January 8–10 2014. Abstract volume p. 14. [27] Olin, S., Schurgers, G., Lindeskog, M., Wårlind, D., Smith, B., Bodin, P., Holmér, J. & Arneth, A., 2015. Modelling the response of yields and tissue C:N to changes in atmospheric CO2 and N management in the main wheat regions of western Europe. Biogeosciences 12, 2489-2515. [28] Olofsson, J., Akselsson, C., Zanchi, G. & Belyazid, S., 2014. Improving the understanding of nitrogen processes by combining field observations and dynamic ecosystem modelling. In: Cordovil C. M. d. S. (Ed.). Proceedings of the 18th Nitrogen Workshop – The nitrogen challenge: building a blueprint for nitrogen use efficiency and food security. 30 June - 3 July 2014, Lisboa, Portugal, pp. 163-164. [29] Olofsson, J., Broström, A., Poska, A., Trondman, A.-K., Mazier, F., Schurgers, G., Gaillard, M.-J., Hickler, T., Sykes, M.T. & LANDCLIM members, 2014. Estimating anthropogenic carbon release in north-western Europe during the Holocene – integrating pollen-based land use reconstructions with a dynamic vegetation model. Open PAGES Focus 4 Workshop: Towards a more accurate quantification of human-environment interactions in the past. 3-7 February 2014, Leuven, Belgium. [30] Olsson, C. & Jönsson, AM., 2014. Process-based models not always better than empirical models for prediciting budburst of Norway spruce and Birch in Europe. Global Change Biology 20, 3492-3507. [31] Olsson, C. & Jönsson, AM., 2015. Budburst model performance: The effect of the spatial resolution of temperature data sets, Agricultural and Forest Meteorology 200, 302-312.

115

LUCCI Annual Report – work package 5 [32] Omstedt, A., Humborg, C., Pempkowiak, J., Perttilä, M., Rutgersson, A., Schneider, B. & Smith, B. 2014. Biogeochemical control of the coupled CO2-O2 system of the Baltic Sea: A review of the results of Baltic-C. Ambio 43, 49-59. [33] Peng, S., Ciais, P., Chevallier, F., Peylin, P., Cadule, P., Sitch, S., Piao, S., Ahlström, A., Huntingford, C., Levy, P., Li, X., Liu, Y., Lomas, M., Poulter, B., Viovy, N., Wang, T., Wang, X., Zaehle, S., Zeng, N., Zhao, F. & Zhao, H., 2015. Benchmarking the seasonal cycle of CO2 fluxes simulated by terrestrial ecosystem models. Global Biogeochemical Cycles 10.1002/2014GB004931. [34] Peng, S., Ciais, P., Krinner, G., Wang, T., Gouttevin, I., McGuire, A.D., Lawrence, D., Burke, E., Chen, X., Delire, C., Koven, C., MacDougall, A., Rinke, A., Saito, K., Zhang, W., Alkama, R., Bohn, T.J., Decharme, B., Hajima, T., Ji, D., Lettenmaier, D.P., Miller., P.A., Moore, J.C., Smith, B. & Sueyoshi, T., 2015. Simulated high-latitude soil thermal dynamics during the past four decades. The Cryosphere Discussions 9, 2301-2337. [35] Piao, S., Nan, H., Huntingford, C., Ciais, P., Friedlingstein, P., Sitch, S., Peng, S., Ahlström, A., Canadell, J.G., Levis, S., Levy, P.E., Liu, L., Lomas, M.R., Mao, J., Myneni, R.B., Peylin, P., Poulter, B., Shen, M., Shi, X., Yin, G., Viovy, N., Wang, T., Wang, X., Zaehle, S., Zeng, N., Zeng, Z. & Chen, A., 2014. Evidence for a weakening relationship between variability and northern vegetation activity. Nature Communications 5, 10.1038/ncomms6018, 2014. [36] Pirzamanbein, B., Lindstrõm, J., Poska, A., Sugita, S., Trondman, A.-K., Fyfe, R., Mazier, F., Nielsen, A.B., Kaplan, J.O., Bjune, A.E., Birks, H.J.B., Giesecke, T., Kangur, M., Latałowa, M., Marquer, L., Smith, B. & Gaillard, M.-J., 2014. Creating spatially continuous maps of past land cover from point estimates: a new statistical approach applied to pollen data. Ecological Complexity 20, 127 - 141. [37] Poska, A., Trondman, A.-K., Nielsen, A.B., Pirzamanbin, B., Strandberg, G., Olofsson, J., Filipsson, H.L., Broström, A., Conley, D., Gaillard, M.-J. & Smith, B., 2014. Modelling past anthropogenic land-cover change and its effects on climate, terrestrial carbon balance and marine ecosystems in north-western Europe. Combining terrestrial and marine proxy data with dynamic ecosystem and climate modelling. 9th European palaeobotany and palynology conference (EPPC 2014). 26-31 August 2014, Padova, Italy. [38] Pulatov B., Hall, K., Linderson, M.-L. & Jönsson, A.M., 2014. Effect of climate change on the potential spread of the Colorado potato beetle in Scandinavia: an ensemble approach. Climate Research 62, 15-24. [39] Rawlins, M.A., McGuire, A.D., Kimball, J.K., Dass, P., Lawrence, D., Burke, E., Chen, X., Delire, C., Koven, C., MacDougall, A., Peng, S., Rinke, A., Saito, K., Zhang, W., Alkama, R., Bohn, T.J., Ciais, P., Decharme, B., Gouttevin, I., Hajima, T., Ji, D., Krinner, G., Lettenmaier, D.P., Miller, P., Moore, J.C., Smith, B. & Sueyoshi, T., 2015. Assessment of model estimates of land-atmosphere CO2 exchange across Northern Eurasia. Biogeosciences Discussions, 12, 2257-2305. [40] Rayner, P.J., Stavert, A., Scholze, M., Ahlström, A., Allison, C.E. & Law, R.M., 2015. Recent changes in the global and regional carbon cycle: analysis of first- order diagnostics. Biogeosciences 12, 835-844. [41] Sallaba, F., Lehsten, D., Seaquist, J. & Sykes, M.T., 2015. A rapid NPP meta-model for current and future climate and CO2 scenarios in Europe. Ecological Modelling 302, 29-41. [42] Schurgers, G., Lagergren, F., Mölder, M. & Lindroth, A., 2015. The importance of micrometeorological variations for photosynthesis and transpiration in a boreal coniferous forest. Biogeosciences, 12, 237-256. [43] Seaquist, J.W., Johansson, E.L. & Nicholas, K.A., 2014. Architecture of the global land acquisition system: applying the tools of network science to identify key vulnerabilities. Environmental Research Letters 9, 114006. [44] Seppä H., Schurgers G., Paul A. Miller, Anne E. Bjune, Thomas Giesecke, Norbert Kühl, Hans Renssen & J. Sakari Salonen, 2015. Trees tracking a warmer climate: the Holocene range shift of hazel (Corylus Avellana) in Northern Europe. The Holocene 25, 53–63. [45] Sitch, S., Friedlingstein, P., Gruber, N., Jones, S. D., Murray-Tortarolo, G., Ahlström, A., Doney, S. C., Graven, H., Heinze, C., Huntingford, C., Levis, S., Levy, P. E., Lomas, M., Poulter, B., Viovy, N., Zaehle, S., Zeng, N., Arneth, A., Bonan, G., Bopp, L., Canadell, J. G., Chevallier, F., Ciais, P., Ellis, R., Gloor, M., Peylin, P., Piao, S., Le Quéré, C., Smith, B., Zhu, Z. & Myneni, R., 2015. Trends and drivers of regional sources and sinks of carbon dioxide over the past two decades, Biogeosciences 12, 653-679. [46] Smith, B., Wårlind, D., Arneth, A., Hickler, T., Leadley, P., Siltberg, J. & Zaehle, S., 2014. Implications of incorporating N cycling and N limitations on primary production in an individual-based dynamic vegetation model. Biogeosciences 11, 2027-2054. [47] Strandberg, G., Kjellström, E., Poska, A., Zorita, E., Gaillard, M.J., Trondman, A.K., Mauri, A., Bjune, A., Davis, B.A.S., Fyfe, R., Giesecke, T., Kalnina, L., Kangur, M., Kaplan, J.O., van der Knaap, W.O., Kokfelt, U., Kuneš, P., Latałowa, M., Mazier, F., Nielsen, A.B., Smith, B., Seppä, H. & Sugita, S., 2014. Regional climate model simulations for Europe at 6k and 0.2k years BP: sensitivity to changes in anthropogenic deforestation. Climate of the Past 10, 661 - 680. [48] Tang, G., Beckage, B. & Smith, B. 2014. Potential future dynamics of carbon fluxes and pools in New England forests and their climatic sensitivities: a model-based study. Global Biogeochemical Cycles 28, 286-299.

116

LUCCI Annual Report – work package 5 [49] Tang, J., Miller, P.A., Crill, P.M., Olin, S. & Pilesjö, P., 2014. Investigating the influence of two different flow routing algorithms on soil–water–vegetation interactions using the dynamic ecosystem model LPJ-Guess. Ecohydrology, 10.1002/eco.1526 [50] Tang, J., Miller, P.A., Persson, A., Olefeldt, D., Pilesjö, P., Heliasz, M., Jackowicz-Korczynski, M., Yang, Z., Smith, B., Callaghan, T.V., & Christensen, T.R., 2015. Carbon budget estimation of a subarctic catchment using a dynamic ecosystem model at high spatial resolution, Biogeosciences, in press. [51] Trondman, A.-K., Gaillard, M.-J., Mazier, F., Sugita, S., Fyfe, R., Nielsen, A.B., Twiddle, C., Barratt, P., Birks, H.J.B., Bjune, A.E., Björkman, L., Broström, A., Caseldine, C., David, R., Dodson, J., Dörfler, W., Fischer, E., van Geel, B., Giesecke, T., Hultberg, T., Kalnina, L., Kangur, M., van der Knaap, P., Koff, T., Kuneš, P., Lagerås, P., Latałowa, M., Lechterbeck, J., Leroyer, C., Leydet, M., Lindbladh, M., Marquer, L., Mitchell, F.J.G., Odgaard, B.V., Peglar, S.M., Persson, T., Poska, A., Rösch, M., Seppä, H., Veski, S. & Wick, L., 2015. Pollen-based quantitative reconstructions of Holocene regional vegetation cover (plant functional types and land-cover types) in Europe suitable for climate modelling. Global Change Biology 21, 676 - 697. [52] Wang, W., Rinke, A., Moore, J.C., Cui, X., Ji, D., Li, Q., Zhang, N., Wang, C., Zhang, S., Lawrence, D.M., McGuire, A.D., Zhang, W., Delire, C., Koven, C., Saito, K., MacDougall, A., Burke, E. & Decharme, B., 2015. Diagnostic and model dependent uncertainty of simulated Tibetan permafrost area. The Cryosphere Discussions 9, 1769-1810. [53] Wårlind, D., Smith, B., Hickler, T. & Arneth, A. 2014. Nitrogen feedbacks increase future terrestrial ecosystem carbon uptake in an individual-based dynamic vegetation model. Biogeosciences 11, 6131-6146. [54] Weiss, M., Miller, P.A., van den Hurk, B.J.J.M., van Noije, T., Stefanescu, S., Haarsma, R., van Ulft, L.H., Hazeleger, W., Le Sager, P., Smith, B. & Schurgers, G. 2014. Contribution of dynamic vegetation phenology to decadal climate predictability. Journal of Climate 27, 8563-8577. [55] Zhang, W., Jansson, C., Miller, P.A., Smith, B. & Samuelsson, P., 2014. Biogeophysical feedbacks enhance the Arctic terrestrial carbon sink in regional Earth system dynamics. Biogeosciences 11, 5503-5519.

117

LUCCI Annual Report – work package 5

Main dynamics and drivers of boreal forests fire regimes during the Holocene Chiara Molinari (a), Veiko Lehsten (a), Olivier Blarquez (b), Jennifer Clear (c), Christopher Carcaillet (d) & Richard H.W. Bradshaw (e) (a) Department of Physical Geography and Ecosystem Science, Lund University, Sweden (b) Centre d'Étude de la Forêt, Université du Québec à Montréal, Montréal (Québec), Canada (c) Department of Forest Ecology, Czech University of Life Sciences, Prague, Czech Republic (d) École Pratique des Hautes Études, Université de Montpellier, Montpellier, France (e) School of Environmental Sciences, University of Liverpool, Liverpool, UK

Introduction Forest fire is one of the most critical ecosystem processes in the boreal mega-biome, and its frequency, size and severity have had a primary role in vegetation dynamics since the Last Ice Age [1]. Fire not only organizes the physical and biological attributes of boreal forests, but also affects biogeochemical cycling, particularly the carbon balance [2]. Due to their location at climatically sensitive northern latitudes, boreal forests are likely to be significantly affected by global warming with a consequent increase in biomass burning [3], a variation in vegetation structure and composition [4] and a rise in pyrogenic carbon dioxide emissions [5]. Even if the ecological role of wildfire in boreal forest is widely recognized, a clearer understanding of the environmental factors controlling fire dynamics and how variations in fire regimes impact forest ecosystems is essential in order to place modern fire processes in a meaningful context for projecting ecosystem behaviour in a changing environment [6]. Because fire return intervals and successional cycles in boreal forests occur over decadal to centennial timescales [7], palaeo-ecological research seems to be one of the most promising tool for elucidating ecosystem changes over a broad range of environmental conditions and temporal scales. Within this context, our aims are: 1) to reconstruct temporal patterns of boreal forests fire dynamics during the Holocene based on sedimentary charcoal records, and 2) to shed more light on different drivers of fire regime during the last 11 000 years.

Methods Palaeo-fire activity was reconstructed based on 137 sedimentary charcoal records covering part of all the last 11 000 cal yrs BP. Sites situated in North-American boreal forests were selected from the Global Charcoal Database (GCD) version3, while most of charcoal records located in Fennoscandia were collated from previous publications (Fig. 1) [8, 9].

Figure 1. Location of the selected site

118

LUCCI Annual Report – work package 5 After sites selection, data normalization and synthesis have been performed by using the paleofire R package [10]. In order to summarize the history of changing in fire regimes through time, composite charcoal curves were constructed for the North American and Fennoscandian boreal datasets (Fig. 2 a, b). Trends in biomass burning were then compared with temperature reconstructions based on fossil pollen data to better understand the importance of climate drivers on Holocene boreal fire activity (Fig. 2a, b) [11, 12]. For an estimation of the influence of vegetation (reconstructed based on pollen data) on boreal fire regime, we developed a statistical linear model to predict Z-score charcoal values in function of vegetation composition and vegetation changes during the Holocene in order to: 1. evaluate (for different selected time slice during the Holocene) the relationship between fire regime and vegetation composition; 2. explore (for each selected site) the relationship between fire dynamics and changes in vegetation composition during the Holocene. In both cases, for an evaluation of the model performance, we performed a Principal Component Analysis (PCA) between the transformed charcoal data and dimension-reduced vegetation data. Finally, we evaluated how well observed outcomes are replicated by the model by an estimation of the r-squared values.

Results The North America synthetic charcoal curve (Fig. 2a) indicates low fire activity at the beginning of the Holocene, starting to quite rapidly increase with a maximum around 10 ka BP. During the period between 9.5 and 6.5 ka BP fire activity decreases, while since 6ka BP biomass burning shows a progressive increase, reaching a pre-industrial maximum around at 2ka BP. This maximum is then followed by a downturn, with a minimum between the 16th and the 19th centuries. The last 250 years are characterised by an increase in biomass burning.

Figure 2. Reconstruction of North American (a) and Fennoscandian (b) boreal forests biomass burning over the last 11 000 years and comparison with temperatures reconstructions based on pollen data.

Also for Fennoscandia, the composite charcoal curve (Fig. 2b) shows low fire activity at the beginning of the Holocene, starting to increase quite fast and reaching a maximum around 10.5 ka BP. This maximum is then followed by a decline during the next 500 years. A rise in biomass burning occurs between 10 and 9 ka BP, while the period

119

LUCCI Annual Report – work package 5 between 9 and 6 ka BP corresponds again to a low fire activity. An increase in biomass burning characterised the following 500 years, while the period between 5.5 and 4 ka BP shows again a decrease in charcoal values. Then, biomass burning progressively increases, with a maximum around the beginning of the Industrial Era. This maximum is then followed by a downturn during the last century.

Figure 3. Examples of the PCA plots constructed for sites located in North America (a) and Fennoscandian (b) to evaluate the relationship between fire regime and vegetation composition during each selected time slice.

Figure 4. Relationship between changes in forest composition and fire dynamics at each selected sites during the Holocene.

The results of the PCA performed for different time slices during the Holocene (Fig. 3) underline a low correlation (rsquared values between 0.01 and 0.25) between forest composition and charcoal values for both North American and Fennoscandian sites.

120

LUCCI Annual Report – work package 5 On the contrary, the results of the PCA performed for the whole Holocene at each selected site, show high correlation values between changes in forest composition and fire dynamics both in North America and in Fennoscandia (Fig. 4).

Discussion The comparison between the North America charcoal composite curve and temperature reconstructions based on pollen data underlines a general agreement between the two trends at the beginning of the Holocene and between 6.5 and 3.5 ka BP. Additionally, the minimum in biomass burning between the 16th and the 19th centuries is probably due to the cold climate occurring during the Little Ice Age (LIA), while the increase in biomass burning during the last 250 years is related to the increase in temperature during this periods. In Fennoscandia boreal forests, charcoal and temperature trends show quite good similarities during the Early and the Middle Holocene, while during the last 4 thousand years the two curves are asynchronous. For what concerns the importance of vegetation composition and vegetation changes on boreal fire regime, both in North America and in Fennoscandia our results show a low correlation between forest composition and charcoal values. This means that: 1) sites with the same forest composition during a particular time slice are not characterised by a similar fire regime and 2) at a site level, fire activity is not the major determinant of vegetation composition (and vice versa). On the contrary, high correlation values exist between changes in forest composition and fire dynamics during the Holocene in both North America and Fennoscandia boreal forests. These results underline that – at most of the selected sites – changes in forest composition correspond to changes in fire regime.

References

[1] Kasischke ES et al. (2000) Fire, Climate Change, and Carbon Cycling in the Boreal Forest. Ecological Studies 138, Springer-Verlag, New York, 461 p. [2] Balshi MS et al.. (2007) The role of historical fire disturbance in the carbon dynamics of the pan-boreal region: A process-based analysis. J. Geophys. Res. 112:G2. [3] Soja AJ et al. (2007) Climate-induced boreal forest change: predictions versus current observations. Glob. Planet. Chang. 56: 274–296. [4] Johnstone JF et al. (2004) Decadal observations of tree regeneration following fire in boreal forests. Can. J. For. Res. 34: 267–273. [5] Bond-Lamberty B et al. (2007) Fire as the dominant driver of central Canadian boreal forest carbon balance. Nature 450: 89-92. [6] Kelly RF et al. (2013) Recent burning of boreal forests exceeds fire regime limits of the past 10,000 years. Proc. Natl. Acad. Sci. U.S.A. 110: 13055–13060. [7] Hu FS et al. (2006) How climate and vegetation influence the fire regime of the Alaskan boreal biome: the Holocene perspective. Mitigation Adapt. Strateg. Glob. Chang. 11: 829–846. [8] Molinari et al. (2013) Exploring potential drivers of European biomass burning over the Holocene: a datamodel analysis. Glob. Ecol. Biogeogr. 22: 1248-1260. [9] Clear et al. (2014) Holocene fire review of Fennoscandia: A synthesis of published charcoal and fire scar records. Int. J. Wildland Fire 23: 781–789. [10] Blarquez et al. (2014) Paleofire: An R package to analyse sedimentary charcoal records from the Global Charcoal Database to reconstruct past biomass burning. Comput. Geosc. 72: 255-261. Comput. Geotech. [11] Viau AE et al. (2006) Millennial-scale temperature variations in North America during the Holocene. J. Geophys. Res. 111, D09102. [12] Davis et al. (2003) The temperature of Europe during the Holocene reconstructed from pollen data. Quat. Sci. Rev. 22: 1701–1716.

121

LUCCI Annual Report – work package 5

Planned LUCCI contributions to CMIP6 with EC-Earth Paul Miller (a), Fredrik Söderberg (b) & Ben Smith (a) (a) Dept of Physical Geography and Ecosystem Sciences, Lund University (b) CEC, Lund University

Motivation and Background Whether global biophysical and biogeochemical feedbacks might act to amplify or mitigate future climate change is being addressed by LUCCI WP5. Most General Circulation Models (GCMs) still use simplified representations of the terrestrial biosphere to model the biophysical exchanges between the land surface and the atmosphere. For example, static vegetation distributions are often assumed, with leaf area index (LAI) fields that are either temporally constant or that have a prescribed seasonality. Following the successful regional climate modelling studies with RCA-GUESS (e.g. [6] & [7]), our task within LUCCI WP5 is to improve the representation of the ecosystem properties and biogeochemical and biogeophysical exchanges within the EC-Earth GCM by including LPJ-GUESS (Smith et al. 2001, 2014) as a key component of the model. The EC-Earth consortium (http://www.ec-earth.org/), are a group of 29 European national weather services, university groups and research institutes that together have developed a new GCM and model of the Earth System called ECEarth. EC-Earth simulates historical and future climate change in response to radiative forcing scenarios [1], and its scenario output has been included in the 5th phase of the Coupled Model Intercomparison Project (CMIP5) and the IPCC’s Fifth Assessment Report (AR5). Furthermore, since EC-Earth has at its core the Integrated Forecasting System (IFS), which is the current seasonal weather forecasting system used by the ECMWF, it is ideally suited to decadal climate predictions of the type that are also included in the AR5 (http://cmip-pcmdi.llnl.gov/cmip5/). A current, key focus of the EC-Earth consortium is the development of EC-Earth into a state-of-the-art Earth System Model (ESM) ahead of its participation in the upcoming Coupled Model Intercomparison Project, Phase 6 (CMIP6 http://www.wcrp-climate.org/index.php/wgcm-cmip/wgcm-cmip6). CMIP6 will be a major contribution to and data source for the next IPCC report. The EC-Earth ESM will improve existing atmosphere (IFS), ocean (NEMO 3.6), sea ice (LIM3) and land surface (H-TESSEL) components, and complement these with models of ocean biogeochemistry (NEMO PISCES), terrestrial biogeochemistry and dynamic vegetation (LPJ-GUESS [3], [4] & [5]), and atmospheric composition (chemistry and aerosols - TM5).

Progress Made within LUCCI in 2014/15 Researchers in WG5 have now completed the technical coupling of the latest version of LPJ-GUESS [5] to version 3 of EC-Earth. Within EC-Earth 3, LPJ-GUESS updates the vegetation composition and land surface properties (i.e. LAI, dominant vegetation type and cover fraction) in dynamic response to the simulated climate. It can also send terrestrial carbon (C) fluxes to the TM5 atmospheric chemistry model for transport and mixing with C fluxes from the PISCES model of ocean biogeochemistry currently embedded in EC-Earth's ocean model, NEMO 3.6, thereby altering and feeding back to the simulated climate through its influence on atmospheric CO2 composition. Thus, EC-Earth will shortly be able to provide us with a better understanding of the biogeophysical and biogeochemical feedbacks at play in the climate system.

Motivation and Planning for CMIP6 Three scientific questions motivate the CMIP6 experimental design [1], namely: 1. Climate forcing. How does the Earth System respond to forcing (e.g. CO2 emissions, land use etc.)? 2. Systematic biases in ESMs. How do these originate, and what are their consequences? 3. Future climate change. How confident can we be in predictions of climate change given the variability and uncertainty inherent in the climate system and forcing scenarios? To address these questions, the CMIP6 model runs are being coordinated by the World Climate Research Programme (WCRP) – see Figure 1. First, there is a set of standard, so-called DECK experiments that all participating models must carry out. These runs consist of an atmosphere-only simulation, a pre-industrial control simulation, and standard

122

LUCCI Annual Report – work package 5 forcing/sensitivity experiments in which CO2 is increased both by 1%/year and abruptly to 4 times the preindustrial level. These experiments will provide continuity across future phases of CMIP [1]. In addition, all models must perform a Historical simulation, i.e. a simulation with prescribed, best-guess climate forcing fields (greenhouse gases, aerosols, land use, volcanic aerosols, solar variability etc.), from 1850 to 2010. The historical simulation serves as a benchmark with which to assess model performance against observed climate variables, and as a standard against which simulations from a set of WCRP-endorsed Model Intercomparison Projects (MIP) can be assessed. The MIPs are model experiments with strict modelling protocols and standardized forcing sets carefully designed to address specifically aspects of the CMIP6 scientific questions given above (Figure 1). The experiments are open to any modelling team with the model set-up, personnel and computing resources that allow them to complete the runs in the CMIP6 time frame [1]. WP5, together with colleagues from SFO MERGE and Almut Arneth’s group at the Karlsruhe Institute of Technology (KIT), Germany, have committed to performing some of the DECK and MIP experiments for CMIP6. Specifically, we will address the role of land use in the climate system within the Land-Use MIP (LUMIP), and climate-carbon cycle interactions within the Coupled Climate Carbon Cycle MIP (C4MIP). In addition, we will contribute indirectly to the Palaeoclimate MIP (PMIP) and Aerosols and Chemistry MIP (AerChemMIP) via our LPJ-GUESS-related output, i.e. historical vegetation composition and BVOC emissions, respectively.

Next Steps The summer and autumn of 2015 will be devoted to testing, tuning and evaluation of the EC-Earth 3 model configurations, before CMIP6 runs begin in late 2015 and early 2016, a date largely determined by the eventual availability of the standardized DECK, historical and MIP forcing datasets [1]. In addition, the model output format must be standardized before the CMIP6 runs begin since all output will be uploaded to a database for use by the wider research community.

Figure 1. CMIP6 themes, and the DECK, Historical, and MIP experiments designed to address the central CMIP research questions (adopted from [2]). Adopted from The main LUCCI and SFO MERGE contributions are shown in RED.

123

LUCCI Annual Report – work package 5 References [1] Hazeleger, W., et al. 2011. EC-Earth V2.2: description and validation of a new seamless earth system prediction model. Climate Dynamics, 10.1007/s00382-011-1228-5 [2] Meehl, G. A., R. Moss, K. E. Taylor, V. Eyring, R. J. Stouffer, S. Bony, and B. Stevens, Climate Model Intercomparison: Preparing for the Next Phase, Eos, Trans. AGU, 95(9), 77, 2014. [3] Smith, B., Prentice, I.C. & Sykes, M.T. 2001. Representation of vegetation dynamics in modelling of terrestrial ecosystems: comparing two contrasting approaches within European climate space. Global Ecology and Biogeography 10: 621-637. [4] Smith, B., Wårlind, D., Arneth, A., Hickler, T., Leadley, P., Siltberg, J. & Zaehle, S. 2014. Implications of incorporating N cycling and N limitations on primary production in an individual-based dynamic vegetation model. Biogeosciences 11: 2027-2054. [5] Weiss, M., P. Miller, B. van den Hurk, T. van Noije, S. Stefenescu, R. Haarsma, L. H van Ulft, W. Hazeleger, P. Le Sager, B. Smith, G. Schurgers, 2014: Contribution of Dynamic Vegetation Phenology to Decadal Climate Predictability. J. Climate, 27, 8563–8577. doi: http://dx.doi.org/10.1175/JCLI-D-13-00684.1 [6] Wramneby, A., Smith, B. & Samuelsson, P. 2010. Hotspots of vegetation-climate feedbacks under future greenhouse forcing in Europe. Journal of Geophysical Research 115, D21119 [7] Zhang, W., C. Jansson, P. A. Miller, B. Smith, and P. Samuelsson (2014) Biogeophysical feedbacks enhance Arctic terrestrial carbon sink in regional Earth system dynamics. Biogeosciences, 11, 5503-5519, doi:10.5194/bg-11-55032014, 2014.

124