Globally, almost half of all remaining tropical forest is allocated for timber production, illustrating the enormous economic asset that these forests represent to many nations. Additionally, these forests provide important societal and ecosystem services, from being sources of food through to climate change mitigation and generating income from carbon offset schemes. Compared to undisturbed forests, much less is known about previously logged and degraded forests that are regenerating. Critically, with increasingly smaller areas of undisturbed forest remaining, the economic and societal importance of disturbed forests has become greater in recent years. However, the resilience of these forests i.e. their capacity to respond to short-term perturbations (e.g. ENSO-induced drought) by resisting damage and recovering quickly, is poorly understood. If we are to manage tropical forests, both in terms of their initial exploitation and subsequent rehabilitation, we need to better understand how these systems respond to periodic drought at local to regional scales. Only then can we develop policies and practice that explicitly take into account the impacts of drought and protect the economic and societal benefits derived from these fragile ecosystems. To provide the evidence from which policy makers and practitioners can better plan forest management strategies we will examine the impact of the current ENSO drought on logged and degraded forests in Borneo, SE Asia, using a combination of ground-based and satellite remote sensing methods. In the field we will examine the response of trees to drought across a disturbance gradient, making use of a network of forest inventory plots that were established in the mid-1990s at the time of the last major ENSO drought to affect the region. In 2014 these plots were revisited as part of a long-term study into post-logging recovery of disturbed forests and as such represent a unique natural laboratory for comparing ENSO-induced changes in forest structure, composition and ecosystem functioning across a land-use gradient and addressing the interactions of logging disturbance and drought. We will revisit 25 plots at least four times during the 18 month project. At each plot we will measure a variety of leaf traits, canopy structure and tree mortality. This will be done by a joint team of UK and Malaysian research assistants who will harvest leaves and analyse leaf chemical properties in facilities at the Universiti of Malaysia. Additionally we will collect spectral reflectance measurements from the leaves. This will allow us to scale up our field observations to use multispectral satellite images to map forest response to the current drought across wider regions. In particular, we will make use of the new Sentinel-2 earth observing satellite to generate region-wide maps of current drought impact, whilst also producing a 20 year time-series drought index from NOAA AVHRR imagery. The latter will provide evidence of the temporal response of the forest to drought in comparison with non-drought conditions, whilst the former will allow us to map the spatial coherence of forest response, determining whether prior disturbance or other factors affect the resilience of forests to drought events. Finally, we will track changes in canopy structure and composition through observations from UAV-mounted sensors, from which we will examine the dynamics of liana/tree composition, which appear to change during drought conditions. With our project partners, the South East Asian Rainforest Research Programme and Permian Global, we will engage with a network of actors who are responsible for forest management across the SE Asia region. We will do this through dissemination activities including a workshop in Malaysia, where will present evidence of the impact of the current ENSO on SE Asian forests and provide a forum for discussion on how best to adapt forest management policy and practice to future ENSO events.

In the UK, as in virtually every developed country, it is widely accepted that we need more people studying and working at all levels in Science, Technology, Engineering and Mathematics (STEM). STEM industries are vital elements of the global economy with jobs in science, technology and engineering predicted to grow at double the rate relative to other careers. Yet there is a widespread consensus that there is a substantial - and growing - STEM skills gap, with insufficient numbers of suitably STEM-qualified workers to meet demand. It has also been argued that STEM skills are beneficial for a wide range of careers and can promote social mobility. Relatedly, there are serious concerns about the lack of STEM-qualified graduates entering teaching and the potential impact of this shortfall not just on schools currently, but also for the future STEM skills gap. Alongside the need to increase STEM participation, important arguments have been made for the imperative to widen participation in STEM and ensure high levels of scientific, mathematical, technical and digital literacy across the population. Specifically, there is a need to broaden the gender, ethnic and social class profile of those who study STEM post-16, particularly in the physical sciences and engineering, where women, some minority ethnic and working-class communities are starkly under-represented. Yet initiatives aimed at increasing and/or widening the profile of STEM graduates appear to have had little lasting impact on the higher education participation rates. Understanding the factors shaping STEM participation is, therefore, a key priority area for governments and a wide range of stakeholders both nationally and internationally. The proposed three year study seeks to understand the processes through which young people develop their science and career choices and trajectories from age 20-23. Specifically, the proposed study will extend the unique dataset developed by the first and second longitudinal ASPIRES and ASPIRES2 studies, which tracked the development of young people's science and career aspirations from age 10-19 (surveying over 39,000 young people at five time points between the ages of 10 and 18 and longitudinal, repeat interviews with 61 young people and their parents over the same age period). ASPIRES3 will continue tracking this cohort via a representative national survey with c.7-10,000 young people at age 20/21, sampled from those who have previously conducted ASPIRES/ASPIRES2 surveys and boosted via online (social media) recruitment, as a media that is particularly successful for recruiting this age group. The project will also conduct interviews with c.60 students who have been longitudinally tracked from age 10 and their c.60 parents. The project will also undertake secondary analysis of previous ASPIRES & ASPIRES2 survey data from over 16,000 students, matched to large national data bases to examine whether/which attitudinal and social factors at age 10/11 relate to later attainment and life outcomes at age 20/21. The knowledge generated by the research will inform inter/national STEM education policy and practice, particularly how to better increase and widen post-compulsory participation in Science, Technology, Engineering and Mathematics. The project will become the only longitudinal project to track young people's aspirations (in and out of science) from primary school, through compulsory, post-compulsory and higher education into work. In line with our commitment to achieving impactful research, the study will involve three impact collaborations with the Royal Society of Chemistry, Engineering UK and the Institute of Physics. Findings and recommendations will be disseminated via a wide range of academic journal articles and bespoke summaries, publications, events and social media for stakeholders.

There is a recognised gap in the communication of information generated by climate scientists and evidence needed by policy makers, in part because influencing policy through research is complex and requires skills that might not be valued or common in research systems. The current situation of our Earth's system, together with the social movements for climate justice, urge a step change in how policy and scientists approach Climate Change. Through this fellowship, I will develop new routes for impact in palaeoclimatology and will lead a vital step change in my field of research. Annual to decadal climate predictions may offer important information to Climate Services and Environmental Agencies, which would help guide short- and medium-term climate change strategies. For example, a better knowledge of the frequency and magnitude of floods in the UK. Decadal climate predictions are skilful for surface temperature, but confidence in projections of atmospheric pattern and the associated ecosystem response are less robust. This is, in part, because the amplitude of the decadal climate response is difficult to verify by the available instrumental data (reanalyses), which only goes back a century or two, and the impact of superimposed low-frequency variability might not be well represented. One way to provide more information on the decadal climate response is to include high-temporal resolution palaeoclimate timeseries in reanalyses. So far, the availability of proxy data suitable for this purpose is limited by the nature of the data (qualitative vs quantitative), chronological constrains (dating uncertainty and time-resolution of the proxy records) and geographical location of the proxy records (i.e limited to specific climate regions as ice-cores and corals), hence the study of decadal climate variability in the past is still in its infancy. In order to make developments in this field, I will lead an international research team that integrates palaeoclimatologists and climate modellers. We will combine emerging methodological approaches in proxy developments, chronological constraints, statistical tools and data-model comparison to provide advanced information of past decadal climate variability in the North Atlantic-European region such as shifting atmospheric circulation and occurrence of extreme weather events; and we will develop emergent constraints based on past climate scenarios to be applied to decadal prediction systems. Beyond the scientific goals, the fellowship aims at a better integration of palaeo evidence into climate policy to create a step change in how long-term climate data are viewed and used by policy and stakeholders. We will create a network of policy advisers, policy makers and other end users willing to engage. A co-development model of research will be adopted to develop shared understanding to design the research outputs, and ensure the research contributes to the specific and current needs of the decision makers across various sectors. The ultimate challenge is to create a leading centre for Palaeo Evidence for Policy at Royal Holloway University of London to: (1) build a palaeo-climate service feeding policy makers with evidence to assist decision-making; (2) support palaeoclimatologists in the UK and overseas to make impact cases studies; (3) train the next generation of early career researchers in policy skills. The fellowship will also explore art-based methods for impact. In particular, creative writing to promote climate science literacy for young children.

Our joint project is called 'Curating Crises'. It brings together teams from the Caribbean (Seismic Research Centre, University of the West Indies, and the Montserrat Volcano Observatory) and the UK (University of East Anglia, University of Oxford, the Royal Society and The National Archives), and focuses on the histories of volcanic crises in the Caribbean. In the past, environmental crises (like volcanic eruptions, or earthquakes) were seen as an opportunity by European scientists to 'drop in', make measurements, gather samples and return home to share their knowledge with other European scientists. This colonial history has left two legacies. First, while there are detailed reports of some of these past crises in European libraries, museums and archives, a lot of this information is only accessible to people who are able to visit in person. Second, the importance of local observers, and the value of their observations, has often been overlooked, or forgotten. Our project will change both of these things. We will use digital techniques to scan and share the records of these past events. We want to share these hidden histories with the communities whose history this is - the communities who lived through these events, and who may be exposed to similar events in the future. We will use these past events to think about the best way to respond to future environmental crises by working together, or by working in new ways. We will also uncover the stories of the local observers, and of the knowledge that they helped to create, and share and celebrate these pieces of environmental history with communities in the Caribbean and the UK. Our project will last 13 months. We will study events in three volcanic islands of the Caribbean: St. Vincent, Montserrat and Dominica, from 1890 - 2000. This includes several major eruptions (1902, 1979; St Vincent; 1995- Montserrat), and earthquake activity (1934-1939, Montserrat). It covers a period of time when all three islands started as British Crown Colonies; and two later became independent nations. We will also look at the way that observations and measurements have changed through time - and see how 'remote' observations, for example from satellites, or measurements by networks of automated instrument have changed the ways that data are shared, interpreted and stored. By the end of our project, our findings will help us to work out the best ways to investigate these and other examples of 'hidden histories', and will give us new ideas for ways that we can work together to understand the environment.

When Darwin was developing his theories of evolution he read avidly in popular natural history magazines and sought out information from an army of almost 2000 correspondents. Such engagement with a wide public in the construction of science became increasingly difficult with the development of professional, and highly specialised science, but the emergence of 'citizen science' projects has suggested a new way forward. With the creation of vast data sets in contemporary science, there is a need for a new army of volunteers to help classify and analyse the information. The Zooniverse platform, started in 2007 with 'Galaxy Zoo', now has over 800,000 participants who contribute to projects from astrophysics to climate science. Significant discoveries have already been made by these volunteers in the field of astronomy. Yet, the structures by which these volunteers might engage with professional science, and through which scientists themselves might draw upon their findings, are not clear, and researchers on the project have been turning to nineteenth-century models of communication to find ways of harnessing this huge popular interest in order to increase the rate of scientific progress. The information revolution in our own age has parallels in the nineteenth century which saw an explosion of print, and journal publishing; in 1800 there were only around 100 science periodicals, but by 1900 this had jumped to 10,000 worldwide. The project brings together historical and literary research in the nineteenth century with contemporary scientific practice, looking at the ways in which patterns of popular communication and engagement in nineteenth-century science can offer models for current practice. The research is timely since the digital revolution, and open-access publishing, are about to change forever the processes and forms of scientific communication and exchange. The project will be based at the Universities of Oxford and Leicester, in partnership with three of our most significant institutions: the Natural History Museum; the Hunterian Museum at the Royal College of Surgeons, and the Royal Society. Researchers will draw on their historic collections, uncovering the extraordinary range of largely forgotten science journals of the nineteenth century, from the Magazine of Natural History (one of Darwin's favourites), to Recreative Science, or Hardwicke's Science Gossip: an Illustrated Medium of Interchange and Gossip for Students and Lovers of Nature. They will also work with these institutions' science communities, addressing questions about the creation and circulation of knowledge in the digital age, and looking at innovative ways of breaking through the public/professional divide. The Zooniverse will extend the range of its work, creating four new citizen science projects which will both accelerate the rate of scientific growth in these areas, and add an estimated 100,000 to 300,000 to the dedicated ranks of citizen scientists. Drawing on the historical research, it will also develop new tools to enable better systems of exchange between professional science, and this growing army of volunteers. As part of the project there will be public symposia in the Natural History Museum, the Royal Society and the Royal College of Surgeons, as well as an exhibition in the Hunterian Museum, and a performance lecture by Professor Marcus du Sautoy. Science has suffered in the public mind from its seeming aloofness and impenetrability. This partnership between humanities and science researchers aims to break down some of those barriers, and to create a truly productive public engagement with science which will enhance the ongoing development of scientific practice.