https://watch.eeg.cl.cam.ac.uk/about/instance/homehttps://watch.eeg.cl.cam.ac.uk/videos/browse?scope=localhttps://watch.eeg.cl.cam.ac.uk/homehttps://watch.eeg.cl.cam.ac.uk/w/ekJtJzPgKGfbxurRgehdxehttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e2a9b0ee-4415-4b13-859e-f04f55f5858e.jpgInvestigating forest dynamics through individual-based modeling (E-Ping Rau)E-Ping Rau: Investigating forest dynamics and functioning in response to disturbances through individual-based modeling Due to the difficulty in conducting real-life experiments on forests, models are often used by forest ecologists to formalize the complex processes and interactions in forest ecosystems, to perform virtual experiments at scale, and to test for ecological hypotheses by validating model predictions against field or remote sensing data. In recent decades, increase in computer power has reduced constraints on model complexity and facilitated development of models with more detailed process representation. In particular, individual-based models take a bottom-up approach and construct a forest through individual-level processes, and generate stand-level patterns as emergent properties of the individual-level interactions. In this talk, I will give a brief introduction to how individual-based forest models could help answer questions about forest dynamics and functioning in response to disturbances, and what work should be prioritized to improve their representation of the disturbance processes. E-Ping Rau is a first-year postdoctoral research associate at the University of Cambridge. He is co-supervised by Srinivasan Keshav (computer science) and David Coomes (plant sciences). His research focuses on combining modeling and remote sensing approaches to evaluate long-term tropical forest resilience and permanence faced with natural or human-caused disturbances. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/6c081cc0-b999-4b00-9ad5-40ad53f4c833/8c037669-7caf-4779-a02a-61a5fd5f9138-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ekJtJzPgKGfbxurRgehdxe3449002024-03-01T09:23:10.525ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/fimoCwGs5bfE3omWHECfxUhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b9e56ec6-5399-4002-bca3-c720e8c3dbb3.jpgData-driven Building Energy Systems: Applications, Platforms, and BenchmarkingThe traditional way to control energy systems is by strategies based on principles of physics. Now there is a transformation where the decision-making is driven by information and data technologies. In this talk, we will briefly review some data-driven applications. With an increasing number of applications, a new challenge is to support such applications at scale. The main problem is how to handle various levels of heterogeneities in different building energy systems so that data and machine learning models can be automatically handled by a platform with minimal human involvement. We present our recent works to automatically translate data into standards and extract data under various local data conventions; as well as schemes to evaluate appropriate machine learning models at scale. We are working with the Electrical and Mechanical Services Department (EMSD) of the Hong Kong government to benchmark data and machine learning models, in the hope to accelerate AI deployment in the campaign towards a smart city and a green city. Dan Wang is a professor in the Department of Computing, The Hong Kong Polytechnic University. His research interests lie in smart energy systems. He published in ACM eEnergy and ACM Buildsys, and he won the best papers in both conferences. He is currently the steering committee chair of ACM eEnergy. He is an advisor of EMSD, the Hong Kong SAR government. He has extensive experience in applied research and his research results have been adopted by industry, including Huawei, IBM, Henderson, etc.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/73cc26b6-3aca-442d-a9ab-658c49598952/9458e59d-45d0-4089-98f2-fd4141f50576-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/fimoCwGs5bfE3omWHECfxU3015012024-03-01T09:23:24.936ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/nXWHbhDZt7TbtbYxgFAgYghttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a7f78e5a-b963-4269-b27d-6da2371080b9.jpgDynamic Distribution Network Reconfiguration with Generation and Load UncertaintyGiven the uncertainty in load demand and renewable energy sources, the distribution network reconfiguration (DNR) problem is a stochastic mixed-integer nonlinear optimization program with a running time that scales exponentially with the number of sectional and tie line switches. Stochastic optimization techniques require knowledge of the stochastic processes of the uncertain parameters, which may not be available in practice. In this seminar, we introduce a deep reinforcement learning algorithm to solve the DNR problem by determining the optimal network configuration using a deep neural network architecture. Vincent Wong is a Professor in the Department of Electrical and Computer Engineering at the University of British Columbia, Vancouver, Canada. His research areas include protocol design, optimization, and resource management of communication networks, with applications to the Internet, wireless networks, smart grid, mobile edge computing, and Internet of Things. Dr. Wong is the Editor-in-Chief of the IEEE Transactions on Wireless Communications.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b1df8911-1a66-4a26-a346-c058f15c02bb/0ed23164-a183-4fe1-ad1e-b8d39b2e4a14-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/nXWHbhDZt7TbtbYxgFAgYg3458012024-03-01T09:23:42.402ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/8BMkpoxwcu5tGezWPVuFRFhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a0a60c32-5bb1-449c-b9e6-b43b439cb9b0.jpgTaking the Long ViewTitle: Taking the Long View: Enhancing Learning On Multi-Temporal, High-Resolution, and Disparate Remote Sensing Data for Measuring the Spread of Buildings Abstract: The progress made in computer vision and satellite technology has opened up new possibilities for observing societies and infrastructure. Powered by a lengthening historical record of high-resolution satellite imagery and an ever-growing set of labels for training machine learning models, it is becoming feasible to investigate a decade of changes in built infrastructure and landscapes. These analyses can assist decision-makers with valuable insights into population shifts, economic trends, and infrastructure performance. Nevertheless, challenges inherent with this kind of imagery — such as varying image quality, imbalances in data collection between urban and rural areas, high costs, and the absence of image metadata — can impede the use and efficacy of these methods. Bio: Jay Taneja is an Assistant Professor in the Manning College of Information and Computer Sciences at the University of Massachusetts, Amherst. As the lead of the STIMA Lab, he and his team develop and study applications of sensing and communications technology on the measurement and planning of societal-scale infrastructure systems in developing regions. He also leads the e-GUIDE Initiative, a multi-university consortium that collaborates with partners across sub-Saharan Africa with an aim to transform the approaches used for planning and operations of electricity infrastructure in the region. Prior to joining UMass, he was a Research Scientist leading the Energy team at the IBM Research - Africa lab in Nairobi, Kenya, from 2013 to 2016. His website is www.jaytaneja.com.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3daef0e3-8d0c-401c-aeb3-0c626e7ee9cd/981e5bb0-db0c-46e5-8394-662defdf28ec-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/8BMkpoxwcu5tGezWPVuFRF3341002024-03-01T09:23:54.686ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/57ydQPqPFEJ4Wdv2NT4dh5https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/19430b50-3494-4a1f-9c9d-01caac8ca1d3.jpgDesigning PV-EV integrated Residential MicrogridsSolar photovoltaic (PV) systems, electric vehicles (EVs), and electrical storage devices are all now available off-the-shelf. Nevertheless, choosing how many solar panels to install, how to place them, and how much storage to buy are costly--yet poorly understood—decisions. Moreover, there is little understanding of the impact of EVs on sizing. This is due to the inherent variability of all three processes: solar generation, residential load, and EV arrivals and departures. In this talk, I will outline the problem, why it is challenging, and recent breakthroughs in my research group in placement, sizing, and operation algorithms that are theoretically well-grounded yet applicable in practice. Srinivasan Keshav is the Robert Sansom Professor of Computer Science in the Department of Computer Science and Technology at the University of Cambridge. His interests lie broadly at the intersection of computer science and sustainability. He received a Ph.D. in Computer Science from the University of California, Berkeley in 1991 and was subsequently employed at AT&T Bell Labs and Cornell University. Most recently, he was a Professor at the University of Waterloo in Canada. He is a Fellow of the Royal Society of Canada, the Association for Computing Machinery, and the Institute of Electrical and Electronic Engineers, and a Distinguished Alumnus of the Indian Institute of Technology, Delhi.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/214edfd9-3a85-4c88-9a20-76090aaa677c/239cd802-b823-40ec-b833-6023d8b034e2-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/57ydQPqPFEJ4Wdv2NT4dh52997002024-03-01T09:24:08.728ZenergysmartgridurbanbatteriesYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/d9VmYKbT8xJvkKMHU3qdv4https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/105f642c-cb02-45db-9fae-49b1879bde48.jpgTokenized Carbon CreditsCarbon credits are one potential tool for climate change mitigation. For various reasons, proof of stake (i.e. energy efficient) blockchains are a natural technological choice for trading carbon credits and offsetting emissions. In this talk, we will discuss tokenized carbon credits, or carbon credits which are distributed, traded, and retired on the blockchain. We will discuss challenges to their efficacy as a tool for mitigating climate change. Derek is a member of EEG and has recently submitted his PhD. His work focuses on smart contracts or blockchain-based programs. He studies their difficult-to-specify properties with the aim of developing high-assurance financial smart contracts.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/626c7fd8-3dce-45ad-aa38-c6e5ef1fbc65/bbe3a767-3a1e-4b69-b246-b9d937be4bfd-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/d9VmYKbT8xJvkKMHU3qdv43393002024-03-01T09:24:19.385Z4ccarbonsensingblockchainforestsYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/uoH2Gie4WiiAocQJYLi9imhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/cdc66707-ea10-4631-b91d-4e22be1065f4.jpgForest biomass mapping and monitoring with NASA LidarsNASA currently has two active satellite lidar missions in orbit that are collecting 3D Earth surface structure data, GEDI and ICESat-2. The GEDI mission was launched in 2018 and has been collecting near-global forest height and structure measurements with a primary science goal of constraining the global carbon cycle. GEDI’s biomass products use simple empirical models to translate lidar waveforms into estimates of forest aboveground biomass (carbon). GEDI’s products cover the full tropical and temperate forests, and ICESat-2 data has been recently used to extend this to fill the northern boreal data gap for circa 2020 global coverage. Similar statistical techniques have been used for ICESat-2 boreal biomass mapping. This talk will present an overview of current products, their limitations, and next steps toward product improvement and change monitoring with a time series of satellite lidar datasets. Dr. Duncanson is an assistant professor at the University of Maryland, college park. She uses lidar data to understand and monitor forest carbon dynamics at local to global scales. She co-leads the development of empirical biomass models for NASA missions (GEDI and ICESat-2), and is involved in several international efforts aimed at bringing consistency and transparency to the field of remote sensing of forest biomass.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/e5eb87d3-c4c8-49d5-b074-94c6a38ba8f6/663dbb1b-5e91-4d2b-bfbc-9c1085f7c8f8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/uoH2Gie4WiiAocQJYLi9im3540022024-03-01T09:24:39.855ZceoforestscarbonclimatelidarYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/uARXZnUm3ys71HkZBDssYZhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/77c0af94-93ce-4a86-a2d4-85e686df78ee.jpgTelling tree stories through laser scans and AIRecent technological advancements in laser scanning have enabled the capture of high-resolution three-dimensional (3D) forest scenes, marking a significant shift in forest management towards sustainability. This presentation highlights the intersection of these cutting-edge laser scanning techniques with the evolving field of deep learning, emphasizing their combined potential in revolutionizing forest mapping and analysis. The focus is on recent developments in deep learning methodologies applied to 3D forest data and the exploration of machine learning-ready datasets that facilitate detailed, tree-level analysis. Specifically, we will delve into the application of deep learning algorithms for critical 3D point cloud tasks, such as tree instance segmentation, forest scene panoptic segmentation, and tree species classification. These advancements not only enable a more granular understanding of forest ecosystems at an individual tree level but also open avenues for optimizing biomass production and stock management while preserving vital ecosystem functions. This confluence of high-resolution 3D data acquisition and sophisticated deep learning approaches represents a transformative approach to sustainable forest management, promising significant contributions to both ecological conservation and resource optimization. Stefano is a research scientist at the Norwegian Institute of Bioeconomy research doing research on the use of proximal laser scanning and deep learning for capturing individual tree properties and applying this information to forest management.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/e79df67f-f14a-47c2-bed9-16efcfca3f91/8a8335d6-7420-4ba3-b243-f3826f9328ec-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/uARXZnUm3ys71HkZBDssYZ3457002024-03-01T09:25:04.587ZforestsceoaisensingYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/gAbSVSej7JgVdEYpknzdmjhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c997e728-761e-4cc4-b3c1-f8e3abde3452.jpgBatteries beyond BatteriesThis talk will present our recent research on Cyber-Physical Systems (CPS) and the Internet of Things (IoT), covering on two closely related research concepts originating from batteries. The first concept involves the traditional use of batteries as power supplies. For this, this talk will introduce a set of relaxation-assisted battery management solutions to enhance the performance of mobile device batteries by allowing them to rest when possible. The second research theme explores a novel concept of utilizing batteries as sensors and actuators to improve system monitoring and control. We will showcase this through a battery-enabled cybersecurity management system for vehicles and discuss its broader applications across various platforms, including satellites and industrial control systems. The talk will conclude by highlighting several future research directions. Dr. Liang He is an Assistant Professor at the University of Colorado Denver (CU-Denver), CO, USA. Prior to joining CU-Denver, he worked as a research fellow at the University of Michigan (2015-2017) and Singapore University of Technology and Design (2012-2014). Dr. He's research focuses on cyber-physical systems and internet-of-things, with applications to batteries, vehicles, and manufacturing systems. He has authored over 100 research papers published in premier academic conferences such as ACM MobiCom, MobiSys, SenSys, UbiComp, and journals like Proceedings of IEEE and TMC. Dr. He is the inventor of 15 US and international patents. He has contributed as a Technical Program Committee (TPC) member for various academic conferences, including ICCPS, IoTDI, and e-Energy. Additionally, he serves as an Associate Editor for IEEE Transactions on Vehicular Systems.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/7e3f126a-612a-4fc6-af31-746116461462/cba68849-2c40-4668-8c8c-3b8e69020d32-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/gAbSVSej7JgVdEYpknzdmj3503002024-03-01T09:25:26.619ZgridenergybatteriesYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/33ThbDGL3TcJMPnfaAd3BJhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/9dd77901-aea0-43a1-8366-fe23b84922bc.jpgThe Real Information in Climate SimulationsProgress towards more reliable weather and climate forecasts is limited by the resolution of numerical models and the complexity of simulated processes. Performance is therefore a major bottleneck and most current models are barely computationally efficient. High-precision calculations are unnecessary, despite being the standard, given the uncertainties in the climate system and the errors from discretisation, data assimilation and unresolved climate processes. I will outline several aspects of low-precision climate computing to preserve information despite fewer bits: (1) The real bitwise information content used to compress the very large volume of climate data produced by numerical models, while minimising information loss. (2) Understanding rounding errors in simple dynamical systems, that arise from the standard floating-point numbers and other number formats. (3) Advances towards 16-bit climate models, which would be a major step towards computationally efficient digital twins of the Earth's climate system. (4) How the Julia programming language enables number format-flexible models without sacrificing performance while accelerating productivity with interactivity and extensibility. Milan is a postdoctoral associate in climate modelling at the Massachusetts Institute of Technology. He received his PhD from Oxford working on low-precision climate computing and data compression. During his PhD, Milan established the concept of the bitwise real information content for data compression. He worked with posit numbers and stochastic rounding and invented a logarithmic fixed-point number format. He ran the first 16-bit weather and climate simulation on Fujitsu's A64FX, the CPU that powers Fugaku. He writes and maintains many Julia packages. Most recently, he wrote SpeedyWeather.jl, an atmospheric general circulation model with a focus on interactivity and extensibility to further accelerate research into computationally efficient weather and climate models.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/10995dde-fd27-4ca2-871e-5e78b777e000/5791e774-1a58-4f22-896e-33d5d883b4ce-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/33ThbDGL3TcJMPnfaAd3BJ3722002024-03-01T09:25:39.150ZclimatemodellingceoYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/whvGL8xNsiq29kwddRUDCDhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/2cf6430a-8f49-4a1e-b8ad-5a74773be4bc.jpgIf a Tree Falls in the Forest, Does It Cause a Fire?Overstory uses satellite imagery to understand individual trees - at scale. I'll be talking about how we do this, and the practical impact Overstory has on wildfire and power outage prevention. Dr Laura James is an engineering leader who builds practical and innovative internet tech systems and organisations in diverse contexts. She is Engineering Director at Overstory, and a co-founding trustee at CoFarm.co and board chair at Now Play This. Find out more at LBJ.org.ukhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f5405d77-78d2-4929-930c-42471e559f31/75ebc29f-1bcf-4511-b14b-4915fec97bb1-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/whvGL8xNsiq29kwddRUDCD3476012024-03-01T09:25:56.754ZforestsceosensingwildfiresYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/8HKQYCt7LA8o4hvCumEBmNhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/f1bab962-93f3-4a49-8cd7-b9f66e3cebe9.jpgProgramming for the PlanetTitle: Programming for the Planet: the Challenges of Repeatable and Reproducible Computing for Environmental Sciences Abstract: As we race to deal with the climate crisis facing the planet, ecologists are increasingly turning to computational analysis, both in the domains of academic research, and in other domains like assessing carbon impact of field projects for carbon offsets. But as if this wasn’t enough, both the expression of computation and the management of the vast amounts of data involved prevent significant hurdles in ensuring that we can generate results that are repeatable, transparent and timely. In the EEG group we are looking at how computer science needs to step in to support ecologists achieve this goal, and this talk will outline the specific challenges we’ve observed working closely with members of the Cambridge Conservation Initiative and avenues we’re taking to solve these problems. Bio: Michael is a research associate in the Computer Lab, part of the EEG, working at the intersection of computer science and ecology: how can computing best support those trying to save the planet. Previously he’s worked in many different domains, all linked by an aim to use technology to solve problems people face: at Bromium he lead the Mac product team working on virtualisation based security software, he worked at Ndiyo, an attempt to make thin client computing a solution to equitable internet access in the developing world, and he’s worked in gaming and education in the earlier days of geolocation in mobile. Outside of computing, he builds guitars that mix traditional woodwork with the cutting edge of 3D printing and generative design.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3e847c3a-e6af-434f-9900-2c74f4cd54d2/f01cc874-563a-4e09-a2aa-bb782bcded76-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/8HKQYCt7LA8o4hvCumEBmN3673002024-03-01T09:26:24.307Zceo4cforestscarbonspaceYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/sNjKKf9wYUfv472trpMXmjhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e5f06b2c-36ce-45ce-8b79-258c6add073e.jpgUncertainty at Scale: How CS Hinders Climate ResearchAbstract: Computer science is a powerful tool for enabling data-driven advances in global ecology and conservation. However, the amplification cuts two ways, as mechanisation can also compound problems inherent with just how uncertain anything to do with natural ecosystems is! Species habitat datasets are uncertain, local observations are uncertain, the resulting inferences about species distributions are uncertain, and side effects from interventions are uncertain; conservation action has evolved to take this into account. Computer science when applied without consideration of these factors can amplify the uncertainty by running ever-larger datasets through increasingly complex data pipelines and algorithms, all built upon wobbly foundations. What exact version of the dataset is being used? What exact version of the dataset did you use? What assumptions went into generating that dataset? What libraries, system dependencies and environment variables were used to calculate the results? In this talk, we first segment sources of uncertainty across ecological data sources and computation over them, and then reflect on how these uncertainties impact ecological research and how we might cleanly bound the uncertainty for future conservation research. Bio: Patrick Ferris is a research assistant in the Energy and Environment Group at the Department of Computer Science, University of Cambridge. He works alongside colleagues in Plant Sciences, Ecology and Zoology to better understand climate change, forests and biodiversity. He is particularly interested in: - How technology can help tackle climate change (e.g. geocaml) and potential issues it creates. - The changing landscape of human rights as a consequence of technology and climate change. - Communicating science effectively to a wide audience (e.g. NI Forests) and improving diversity and inclusion in tech (see Outreachy).https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/d9056054-5bab-432b-a6c9-2e6436e804ae/4f18e13b-bccf-44fa-9e05-a2ce724a3f25-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/sNjKKf9wYUfv472trpMXmj3592042024-03-01T09:26:34.625ZpolicyclimateceoYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/wd9tbpCsgcBR3Rf4MF3Y6Chttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/cf3a0368-cd03-4d59-9415-f242ce7414f8.jpgCarbon Clarity in the Global Petrochemical Supply ChainAbstract: Our world could not function without plastics, synthetic fertilisers, and chemical derivatives. Yet making these products exacts a significant toll on the environment. The petrochemical sector is accountable for a third of industrial final energy use, a tenth of global oil and natural gas demand, and almost a fifth of global industrial carbon emissions. And demand for chemicals is expected to double by 2050. A future sustainable petrochemical sector will need to innovate quickly to deliver products with net zero GHG emissions and a reduced burden on the environment and natural resources. However, the decarbonisation of the petrochemical sector is challenging given the complexity of supply chains and thermodynamic constraints of the chemical reactions. Emissions are released throughout their life cycles, with varying stages dominating for different products. Mitigation options are therefore less straightforward than in other industries, requiring a system-wide approach. The nature of the sector also makes data collection and analysis challenging, meaning there is currently no reliable, comprehensive picture of GHG emissions or energy, mass, and trade flows of the petrochemical sector. In this talk, I will present finding from C-THRU a 3-year international research project with the aim of delivering carbon clarity in the global petrochemical sector. The project aims to build the world’s most comprehensive, reliable, and transparent account of current and future emissions from the petrochemical sector and explore how future interventions and innovations could minimise greenhouse gas emissions. The research supports strategic policy and business decision-making to promote the sustainability of the petrochemical sector, making it compatible with climate change mitigation goals. Bio: Jonathan Cullen is the Professor of Sustainable Engineering at the University of Cambridge. Jonathan's research interests span energy and material systems, efficien...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f4a44dd3-60f4-4768-9094-3a22612396c6/011810be-21c0-41f4-b417-9e52747ddea6-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/wd9tbpCsgcBR3Rf4MF3Y6C2782022024-03-01T09:26:43.735ZcarbonpolicyclimatesupplychainsYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/t8yxUErtY3tRn3r8rDAACbhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c5ef6c24-63e2-4016-ae42-9874efb4a199.jpgAI-Refined Radiative Transfer Modelling to Retrieve Biophysical Variables in ForestsRecent advancements in machine learning, combined with the availability of vast remote sensing data, have led to significant progress in ecology and climate science. However, the lack of interpretability in learned representations limits their application to crucial environmental challenges. Understanding the biophysical properties of forests, for instance, is essential in comprehending their role in mitigating climate change. In the field of remote sensing, scientists have attempted to retrieve the biophysical variables by inverting the radiative transfer models (RTMs). However, classical approaches overlook the presence of systematic bias in RTMs, which is particularly problematic when extracting variables from complex forest structures. Motivated by physics-informed machine learning and disentangled representation learning, we propose an innovative approach that integrates the RTM with an auto-encoder-based architecture. Our approach integrates an RTM into a contemporary machine-learning framework and effectively corrects its bias, resulting in improved variable extraction. The developed pipeline holds broad applicability in other machine-learning problems involving physical models. Our research advances the integration of RTMs and machine learning, enabling more accurate analysis of remote sensing data and facilitating a better understanding of forest biophysical properties. Yihang She is a first-year PhD student in Computer science at the University of Cambridge. His PhD focuses on the development of 3D vision algorithms to enable real-time and low-cost forest carbon estimation.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/dbb5028b-bc9c-4e37-ab0d-6a30d185327a/50e6ea46-05d5-436e-b8f2-acf5af18fd73-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/t8yxUErtY3tRn3r8rDAACb3357002024-03-01T09:26:53.835ZceoaisensingforestsclimateYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/boqxJ3DXmtokY4ykdyhS3Dhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/eafc17b9-1dde-444b-bd76-20ac45b04c00.jpgChallenges in Computing Climate Sensitivity and Climate SolutionsFull Title: Challenges in Computing Climate Sensitivity and Climate Solutions: Examples from Glaciology and CO2 Storage Challenges in Computing Climate Sensitivity and Climate Solutions: Examples from Glaciology and CO2 Storage "Understanding a predicting the observed changes in key climatic systems, or predicting and de-risking climate solutions pose challenges, particularly in assessing poorly constrained properties of the system whose behaviour at small scales often determines the response. In this talk I’ll give two physical examples, and an approach of reduced modelling which focuses on the key uncertainties. The two large global ice sheets are loosing significant mass, but Greenland predominantly looses mass by surface melting, while Antarctica looses mass by melting of ice shelves and in the zone where ice becomes grounded. In both cases, the response is dictated by the temporal and spatial evolution of a complex subglacial hydrological system, whose properties are difficult to observe remotely. We’ll use reduced models of the subglacial system to understand the key role of subglacial systems and pose the question of how best to constrain this subglacial hydrological system. Geological storage of CO2 is one method of reducing anthropogenic CO2 emissions, and involves the injection of large volumes of CO2 into the subsurface. In order to remain trapped in the subsurface, buoyant CO2 is typically injected beneath a relatively impermeable cap rock. Observations of CO2 stored in various storage sites demonstrates that the subsequent flow of CO2 is dominated by variability in caprice topography and variations in permeability which are not readily observed remotely. We’ll again use reduced models of CO2 spreading to understand the sensitivity to geological heterogeneity, and pose the question of how best to estimate the uncertainties associated with the storage of large volumes of CO2 in the subsurface. Lerome Neufeld is Professor of Earth an...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/541ce08d-3665-45de-9a83-f55404cb1f61/5588d8ec-1ed1-47af-86f4-e8b45a3e1822-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/boqxJ3DXmtokY4ykdyhS3D3856002024-03-01T15:59:19.544ZclimateglaciologyceoYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/mnPcScreqb3j4pMSAxtu3ohttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/22a429d9-963f-4174-a59a-9b99b1b18938.jpgAnalyzing the Impacts of Extreme Heat on BuildingsAbstract: As our planet warms, the world faces increasingly severe and frequent heatwaves, challenging the comfort and safety of residential environments. While existing studies often focus on historical conditions and general future weather predictions, the impacts of extreme climate events on buildings remain underexplored. This leaves critical questions unanswered regarding the vulnerability of buildings and the well-being of occupants in the context of climate change. To address these challenges, our research analyzes both historical data and future weather scenarios to assess the resilience of the UK's residential housing against extreme heat. In my talk, I will present the insights gained from our analysis, highlighting key factors that contribute to the vulnerability of homes to heatwaves and the expected severity of these conditions. The focus hereby lies on indoor thermal comfort and the direct consequences for occupants. We further introduce a novel metric for quantifying liveability in buildings based on ranges of possible activities. The results are made available through the Heatscore application, allowing users to assess the liveability within homes across the UK under different climate conditions. Finally, I will present the Heatalyzer, a novel Building Energy Modeling tool developed to make our methodology accessible to a wide audience. This tool allows users to evaluate and understand the impacts of extreme heat on indoor environments for various geographic locations. Bio: Livia Capol is a Master’s student in Computer Science at ETH Zurich. She has been working on her thesis at the University of Cambridge since last September under the joint supervision of Keshav and Zoltan Nagy (The University of Austin at Texas).https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/a5027bb4-9161-4b65-882a-bd0c636f74e2/d132ffd2-1a74-48e6-80f7-564af929a2dd-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/mnPcScreqb3j4pMSAxtu3o3173002024-03-11T10:37:22.973ZenergyurbanclimategeospatialYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/g7kZvxQfBRpTqpmLzfA1QDhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/97bad843-0324-4577-adf8-a6407d1d42b9.jpgRenewables as Reserve Providers; Turning a Challenge into an OpportunityAbstract: The expansion of renewable energy generation increases the need for short-term reserve facilities to compensate for their short-term variations. This makes reserve markets increasingly profitable and attractive for renewable energy producers (REPs), who are facing diminishing subsidies and returns. Policymakers would also value REPs as reserve providers because conventional reserves are generally carbon-intensive. The major hurdle is, however, that reserve markets require reliability, while renewables are intermittent. This brings financial risks to REPs and reliability risks to the system operators. Two remedies to alleviate these risks are intraday trading and storage. The open question is whether, these hedging instruments, individually or combined, can resolve the financial and reliability risks and facilitate REP'S participation in reserve markets. This is currently unknown and, among others, depends on the micro-structure of the intraday markets, mainly distinguished as discrete (D-ID) or continuous (C-ID) markets. We study this problem by formulating the operation of a profit-maximizing REP, with and without storage, providing reserve services as a multi-stage stochastic integer program, separately with the support of D-ID or C-ID markets. We combine the Benders decomposition and stochastic dual dynamic programming algorithm (SDDP) to solve the problem efficiently. Our analysis of real data from the German market provides interesting insights into REPs' participation in short-term reserve markets. Importantly, we find that C-ID trading is the best enabler among all, facilitating the profitable and reliable participation of REPs in the FCR market. In this case, batteries not only do not help FCR participation but also worsen the reliability. Conversely, D-ID markets do not help FCR participation and REPs need batteries for a reliable and profitable FCR participation. Thus, system operators should discourage the use of batteries (for REPs) in ca...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/7a5bdcdc-e5a8-4266-bd29-053e59a624d5/481f5d14-f64e-4321-8a57-30864c3adba8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/g7kZvxQfBRpTqpmLzfA1QD3472002024-03-15T16:23:48.844ZenergygridbatteriespolicyYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/vvMsqPd7bJps5Xa1BsquD3https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/f1e13d72-7e4a-43af-8cf2-d75ae1bc1522.jpgThe Future of 24/7 Clean Energy driven by AIDeep decarbonization and rapid electrification of energy will require greater penetration of renewables of energy supply, and electrification of energy demand. As renewables penetration crosses 10-20% of the grid electricity demand (and other supply sources correspondingly adjust), the intermittency and volatility of renewable supply & new electrified demand will increasingly dominate the market. Renewable supply and grid electricity demand needs to be matched through a combination of multiple markets, energy storage and an orchestrated portfolio of flexible resources. The future of renewables will fundamentally be driven by software and AI on the cloud to manage this transition. This talk will unwrap the various challenges and opportunities around this transition. We will also cover examples of the application of multi-modal Generative AI for advanced remote operations, health & safety and multi-lingual knowledge access in the context of problem-solving in field operations. Shiv's Bio: Shiv is CTO, Energy Industry, Asia at Microsoft. Previously he was Executive General Manager of Growth Offerings at GE Power Conversion responsible for a new Line of Business development in e-Mobility, Commercial & Industrial Solar and digital/AI innovations. Earlier he was at IBM Research - India, and the Chief Scientist of IBM Research - Australia. Before IBM, he was a tenured Full Professor at Rensselaer Polytechnic Institute in Troy, NY, USA. Shiv has degrees from Indian Institute of Technology, Madras (B.Tech, CS), Ohio State University (MS, PhD) and RPI (Executive MBA). Shiv is a Distinguished Alumnus Awardee of IIT Madras (2021, recognizing 0.3% of IITM's alumni over the years) & Ohio State University (2021), Fellow of the IEEE (2010), Fellow of Indian National Academy of Engineering (2015), ACM Distinguished Scientist (2010), MIT Technology Review TR100 young innovator (1999). Srinivasan's Bio: Srinivasan Iyengar is a Senior Program Manager at Microsoft's Energy...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ef019a7f-acba-4050-be5e-9b38b4c1f214/df12819a-3bc3-4cb2-877d-9afd8f984396-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/vvMsqPd7bJps5Xa1BsquD33712012024-03-22T17:30:33.252ZgridenergyaieconomicsYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/ajc61T5nszFP1NcnwXbnVrhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/658c1a8c-1391-456e-97bd-3fdf438505e1.jpgTowards Global, General-Purpose Geographic Location EncodersGeospatial data is common across a wide range of disciplines and modeling tasks, e.g. in ecology or urban analytics. Location features are often not readily available and need to be obtained via individual data collection and fusion. This opens the opportunity for a new class of "foundation models": global, general-purpose geographic location encoders, which provide vector embeddings summarizing the characteristics of a location for convenient usage in downstream tasks. I will outline the intuition and technical challenges for building these models, and contextualize them with respect to other geospatial foundation models in the vision, language and geophysical domain. Bio: I am a postdoctoral researcher at Microsoft Research New England and part of the Machine Learning and Statistics group. My research focuses on the representation of geographic phenomena in machine learning methods, particularly in neural networks. My recent work includes the integration of notions of spatial dependency into neural networks and the unsupervised training of location encoders that learn characteristics of a given location and can be deployed in different downstream tasks. My work is motivated by real-world challenges such as climate change and increasing urbanisation, combining technical and methodological research with application and deployment studies. I have a PhD in Computer Science and Urban Science from the University of Warwick and spent time as a visiting student at NYU, as an Enrichment student at the Alan Turing Institute and as a Beyond Fellow at TUM and DLR. Website: https://konstantinklemmer.github.io/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4b6c77d4-060b-4365-bff0-6f108baf858f/0849bda3-a41b-45a4-b1ba-a97645127103-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ajc61T5nszFP1NcnwXbnVr3523052024-03-28T21:21:09.708ZcoesensingaispatialYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/c89onE3seuEkT2j7RqyNAyhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/89034b84-e89f-4873-a32f-f39708ff2dfd.jpgTracking Compound Heat and Humidity ImpactsAbstract: Compounds of high humidity and heat are deadly. With our changing climate an increasing proportion of the world’s land surface is being exposed to higher temperatures and humidity, with this rise being faster in populated areas. One metric to measure heat stress, a rise in body heat as the result of external environment and exertion is Wet Bulb Globe Temperature (WBGT) the International Organization’s occupational health metric. However, Wet Bulb Globe Temperature is often not observed at a meteorological station because of a key instrument – a globe thermometer being expensive to purchase and in local cases often being stolen from sites. This led to several approximations being developed, with differing levels of accuracy. In this talk, I will discuss a new method developed for numerical weather prediction models to produce accurate WBGT measurements. I will then discuss gaps within literature for WBGT and biases that exist within the literature. Before, discussing how we are applying the new WBGT data to research such as heat related child mortality and our developments in comparing WBGT heatwaves to temperature heatwaves through a heatwave algorithm. Finally, I will discuss how this WBGT data could be applied to energy systems and chances for future avenues of research. Bio: Dr Chloe Brimicombe is a post-doctoral researcher in the Social Complexity and System Transformation Research Group at the Wegener Center, University of Graz, Austria and visiting fellow at the London School of Economic and Public Policy. She is a Climate Scientist and Extreme Heat Researcher focused on heat impacts such as those on pregnant women and children's health. She previously helped to develop thermal comfort indices for the European Center for Medium Range Weather Forecasts Numerical Weather Prediction Products.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/5a13e648-f353-4130-8ef0-fc204e7273ec/f93326d9-397f-4fed-ae90-48b3adaa83e9-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/c89onE3seuEkT2j7RqyNAy2582032024-04-05T16:06:48.924ZclimateconservationcoesensingYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/bjwjsPSyfuzFpTYnsM42LZhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/1f8fb483-2068-4ddb-b2ad-c27ece48cf6f.jpgWhat is 4C? About the Cambridge Centre for Carbon Credits(Created Nov 2021) We are building a trusted, decentralised marketplace where purchasers of carbon credits can confidently and directly fund trusted nature-based projects. The Cambridge Centre for Carbon Credits (4C) has two primary goals: 1. To support students and faculty members conducting foundational research in the relevant areas of computer science, environmental science, and economics 2. To create a trusted decentralized marketplace where purchasers of carbon credits can confidently and directly fund trusted nature-based projects that ties together corporate funders to conservationists via automated and transparent global oracles. This marketplace will be compatible with existing schemes such as IPCC Article 6 of the Paris Agreement. 4C seeks to partner with the private sector, NGOs, and governments to iteratively refine these ideas. Our goal is to exponentially increase the number of real nature-based conservation and restoration projects by channeling billions of dollars of worldwide funding towards them via market-based instruments. All of the instruments will be bootstrapped by existing government initiatives to handle Know Your Customer (KYC) and land ownership requirements, depending on the geography involved.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/53917573-1f77-4c61-97e5-aeb5bc71001d/4074fe93-2f5d-4a71-887c-8b13657c9540-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/bjwjsPSyfuzFpTYnsM42LZ1371012024-04-05T18:05:27.730ZYESCambridge Centre for Carbon Credits (4C)NOhttps://watch.eeg.cl.cam.ac.uk/w/kJQ3vG6FYBfyrxqbMH4kPrhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/271155a6-aa29-4cf0-ab48-e3bfcdd0b9bf.jpgOur approach: 4C explainerHalting global deforestation via carbon credits: an overview of our approach in 4C ------ Prof Srinivasan Keshav (@ProfKeshav) is using computer science to move towards a sustainable future through clean energy and environmental conservation. He is a co-director of 4C - the Cambridge Centre for Carbon Credits - and the Robert Sansom Professor of Computer Science in the Department of Computer Science and Technology at the University of Cambridge. ------ The world is facing two major crises: a climate crisis and a biodiversity crisis. While the former is now very familiar with the public, fewer people know about the latter. The biodiversity crisis refers to the dramatic loss of species globally due to human activities – the two most important of which are habitat loss and climate change. Preserving forests can help to tackle both these crises in conjunction – it prevents carbon dioxide from leaving the trees and retains biodiversity. Nowhere is this more important than in tropical rainforests, where a large fraction of the world’s biodiversity exists and which are still relatively intact habitats compared to other parts of the planet. How do we prevent deforestation? We believe we need to pay for it. If someone who lives near the forest decides it is economically beneficial to cut the trees and sell it for timber, and then raise cattle on the plot, then they will most likely do so – especially if that person is one of the millions around the world in poverty. However, if we can pay them not to convert their land, then that will balance the scales and incentivise protection of the forest. Thanks for this summary to James Miller, student environmentalist and film-maker.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/9fd87b09-f8f2-41ed-a862-912ef5ba1b13/957d374d-f262-479b-a404-1b0f063b878c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/kJQ3vG6FYBfyrxqbMH4kPr129502024-04-05T18:05:32.843ZYESCambridge Centre for Carbon Credits (4C)NOhttps://watch.eeg.cl.cam.ac.uk/w/vcz5HNHhmH6LVWKumuC22phttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c1851945-3591-467a-a575-64b9386114c7.jpgAdditionality: 4C explainerCalculating the "additionality" of deforestation avoidance interventions ------ Prof Srinivasan Keshav ( @ProfKeshav ) is using computer science to move towards a sustainable future through clean energy and environmental conservation. He is a co-director of 4C - the Cambridge Centre for Carbon Credits - and the Robert Sansom Professor of Computer Science in the Department of Computer Science and Technology at the University of Cambridge. ------ When evaluating forest protection projects, the concept of additionality is crucial. A landowner might claim to have saved a patch of forest from destruction, and ask to be paid on that basis. However, it could be that the forest was not saved as a result of their actions – it might be that the forest was never going to be destroyed in the first place, whether that landowner managed it or not. We want to pay only for services rendered – for "additional' value conferred. How can we determine whether or not that was the case? It is impossible to prove what would have happened in the absence of an intervention, but we can gather evidence to make a case using counterfactual scenarios. For example, if there are two identical patches of forest adjacent to a road, one of which was managed by the landowner, one of which was not, and after twenty years only the forest managed by the landowner remained unfelled, then that would make it more likely that the landowner had made a positive difference to the fate of the forest. This concept can be generalised – we can take samples from within a project area, called ‘pixels’ and pair them up with other pixels taken from outside the project area, ‘counterfactuals’. These pairs of pixels are matched in all major characteristics – forest cover, distance from roads, elevation, soil type, rainfall etc. If we do this well, we expect the paired pixels to experience similar rates of deforestation. These pixel pairings can then be monitored throughout time using satellite images taken over m...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ec7696f9-5b0e-40a0-b638-cd97712eebd5/bf09d00e-681f-4d64-b4ad-87806ab7d90a-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/vcz5HNHhmH6LVWKumuC22p680002024-04-05T18:05:36.618ZYESCambridge Centre for Carbon Credits (4C)NOhttps://watch.eeg.cl.cam.ac.uk/w/sNFpsJKXxZAimA4YNpkdkphttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c5ccb4f7-e4cc-4b1b-bdcf-6dc899f369ed.jpgPermanence: 4C explainerAdjusting impermanent nature interventions to be equivalently permanent to geological sequestration ------ Prof Srinivasan Keshav (@ProfKeshav) is using computer science to move towards a sustainable future through clean energy and environmental conservation. He is a co-director of 4C - the Cambridge Centre for Carbon Credits - and the Robert Sansom Professor of Computer Science in the Department of Computer Science and Technology at the University of Cambridge. ------ When it comes to carbon sequestration, there are two traditional categories. Permanent geological sequestration entails CO2 being pumped down into cavities in rock to form a layer of calcium carbonate. This method is quite effective and very long-term but it is expensive. The second category is nature-based solutions, like planting a tree. When a tree is planted, it locks down carbon in its tissue – both in the above ground biomass that we can see, and in the below-ground biomass, like roots. This type of sequestration is far more cost-effective and comes with many cobenefits for humans and wildlife, but it has a key disadvantage – it is impermanent. Eventually the tree will die, and much of that captured CO2 will be released back into the atmosphere. Does that impermanent sequestration of carbon dioxide still have value in the fight against climate change? How can we begin to measure what that value is? And how do we know how long that carbon dioxide will be sequestered for? The framework we use for working this out rests on a concept known as the ‘social cost of carbon’. CO2 in the atmosphere lasts for 300-500 years, where it has a number of detrimental effects. The damage of those effects to society can be estimated in dollars, and that total can be divided to attribute a cost to each single tonne of carbon released in a particular year. The damage inflicted by a single tonne throughout its entire lifespan is the social cost of carbon. It is worth noting that this estimate encompasses a dis...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/d9121a3b-2e0b-4669-aac6-1cb44e450dbd/07078950-0e7e-4906-bdf4-440969876917-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/sNFpsJKXxZAimA4YNpkdkp1016002024-04-05T18:05:43.809ZYESCambridge Centre for Carbon Credits (4C)NOhttps://watch.eeg.cl.cam.ac.uk/w/6rodKhXimH1G5AdT3W7hnHhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e1d52a82-fbb2-4ee0-a94e-04bbefbe2971.jpgLeakage: 4C explainerExplaining the leakage effect resulting from negative externalities in deforestation avoidance ------ Prof Srinivasan Keshav (@ProfKeshav) is using computer science to move towards a sustainable future through clean energy and environmental conservation. He is a co-director of 4C - the Cambridge Centre for Carbon Credits - and the Robert Sansom Professor of Computer Science in the Department of Computer Science and Technology at the University of Cambridge. ------ Another integral and historically neglected concept in nature protection schemes is leakage. It might be that when you protect an area from deforestation, you simply displace any harmful activity from inside the protected area to neighbouring unprotected areas (local leakage). Or, for some internationally traded products, that activity could be shifted to an area on the other side of the world (global leakage). For example, imagine someone is paid to not cut down trees in a project area. They might then might instead simply begin cutting down trees in the areas adjacent to the project instead, rendering no net benefit to the forest. That would be local leakage, and it’s a commonly documented phenomenon. For an example of how global leakage could occur, imagine what would happen if the landowner stopped cutting trees entirely, and that cut supply of timber to the international market that they originally supplied. Purchasers might instead decided to buy timber from another part of the world, increasing the deforestation rate in that area to meet the increased demand. Again, money has been spent, but there is a minimal contribution to carbon sequestration and biodiversity loss – perhaps none at all. How can we detect leakage? We can create a ‘buffer zone’ around the project area, normally around 5km radius, in which we also monitor deforestation using satellites. If we detect increasing deforestation in this buffer zone since the project began, then that implies that leakage is likely to have occurre...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/2c08fb20-aa41-454a-b1ce-decbbe8bf77b/4bd1426d-8e43-4e7c-ab9d-8bfe49f95c9f-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/6rodKhXimH1G5AdT3W7hnH222002024-04-05T18:05:45.683ZYESCambridge Centre for Carbon Credits (4C)NOhttps://watch.eeg.cl.cam.ac.uk/w/xaxnfeUZC12Az2E2NHjUJQhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/8d023c3e-f554-481d-84d1-2678e4691999.jpgICFP23 Keynote: Functional Programming for the PlanetAs simultaneous crises in emissions and biodiversity sweep the planet, computer systems that analyse the complex interplay of our globe’s health are ever more crucial to guiding policy decisions about how to get out of the mess we’re in. In this talk, I examine how functional programming can contribute to building systems that are more resilient, predictable and reproducible in the face of huge amounts of input data (such as from satellites and ground sensing) that demands precise access control (or else poachers and malicious actors go straight to the source) and requires interactive exploration from non-CS-experts at different levels of the software stack (to do climate science). I will also highlight how our ongoing cross-disciplinary research is having real impact on conservation projects that are sorely underserved by current systems/PL infrastructure, and also how we went about forging these links. I hope to encourage some of you into forming your own local collaborations with your colleagues working on the climate crisis! See https://icfp23.sigplan.org/details/icfp-2023-icfp-keynotes/50/Programming-for-the-planethttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/fc605293-4c1c-41a5-9b04-3b03c65a2664/944bca34-f0be-453a-bb62-63006f13abe0-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/xaxnfeUZC12Az2E2NHjUJQ3688022024-04-05T18:35:55.284ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/qEsMt2Ayk37SaKgxrfwoBthttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/2624fe8f-9073-43af-b483-aa4e145f40e4.jpgMechanisms Against Climate ChangeAbstract: The problem of climate change is essentially about cooperation over a common resource. It is challenging because it is global, and countries are free to cooperate or decline. The good thing is that we know a lot about what properties of agreements make cooperation work. The bad thing is that our current approaches, such as the Paris Agreement, get almost all of these properties exactly wrong. In the talk, I will explain why this means that the Paris Agreement is exceedingly unlikely to succeed. I’ll also attempt to be constructive, and look at what agreements which might actually work in practice would have to look like. Bio: To see Carl Edward Rasmussen's bio, please click the link: mlg.eng.cam.ac.uk/carl.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c7ba18e6-2fca-4ecc-9e25-e07b5b101991/de1211de-ba0b-4db5-8a2f-c7f1b9f3945c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/qEsMt2Ayk37SaKgxrfwoBt41025252024-04-12T15:33:05.834ZclimatepolicyYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/kgP3qnPWMv9nNvSk9fzWXMhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c7f2c5e4-87f7-41cd-94ec-b1401a6d6edb.jpgFinancing Forests: A Credible Approach towards Halting Tropical DeforestationThe rate of tropical deforestation is continuing to increase, and with it comes an enormous loss of biodiversity and natural resources. Ideally we need a massive number of new conservation projects that will provide alternatives to deforestation in the regions worst affected in the tropical equatorial belt. How can we figure out where to focus our efforts, and to verify their progress? I will discuss how we are using satellite remote sensing to construct a global view of tropical rainforests, and to measure the effectiveness of conservation interventions from a viewpoint of the additionality gained vs natural forest dynamics, the tracking of leakage due to negative externalities, and the permanence of resulting changes. We use these measurements to build up a quantitative model of the conservation efforts that share common baselines globally, and then combine with qualitative metrics such as biodiversity, local livelihood and justice to build up a complete project assessment that can be used as the basis for trustworthy, verifiable carbon credits. I will also discuss the computer science challenges required to build this infrastructure: the need for planetary scale computations, the need for digital permanence so projects can be tracked over decades, and the need for decentralised trust to operate globally. Departmental Seminar at Cambridge: https://talks.cam.ac.uk/talk/index/185282https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/9c12c815-f9f7-4c1d-b005-bbe870a02603/54a19556-1d30-4820-90f9-ce2ce45d30a7-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/kgP3qnPWMv9nNvSk9fzWXM3719002024-04-13T12:00:00.824ZYESCambridge Centre for Carbon Credits (4C)NOhttps://watch.eeg.cl.cam.ac.uk/w/tcDcPLhSGbiXNVpXy6AHvThttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a49545dd-360d-466e-894f-8849724ea4e6.jpgNet Zero Speaks with Srinivasan KeshavWelcome to Net Zero Speaks With Srinivasan Keshav, hosted by Climate Activist Prachi Shevgaonkar. Prachi Shevgaonkar (Founder of Cool the Globe) engages with renowned computer scientist and environmental advocate, Professor Srinivasan Keshav from the University of Cambridge. Discover the groundbreaking "Permanent Additional Carbon Tonne" (PACT) method, revolutionizing carbon credit valuation for forest conservation as Net Zero continues monitoring sustainability and climate action. CAST & CREDITS PRODUCER Vanessa Vela ART DIRECTOR Will Wells ASSOCIATE PRODUCER Philo Magdalene _________________________________ Curated by Protect Our Planet & Planet Classroom Uploaded by Planet Classroom 🌎 Follow us on: ❤️ Instagram: https://www.instagram.com/planetclassroom ❤️ Facebook: https://www.facebook.com/planetclassroom ❤️ Twitter: https://www.twitter.com/planetclassroom ❤️ LinkedIn: https://www.linkedin.com/company/planet-classroom 🌎 View more content at https://www.planetclassroom.world The Planet Classroom Network, brings together musicians, dancers, video game creators, filmmakers, activists, learning innovators, and emerging technologists from around the world to entertain, educate and engage youth, and to provide a rich cultural experience for all. Content showcased for youth and by youth on the Planet Classroom Network is provided by over 30 cultural organizations from around the world. Young people globally play a significant role in conceptualizing, creating, and producing the network's vision and programming. #RacetoNetZero, #UniversityofCambridge, #ComputerScience, #VisionPro, #TimCook, #Metaverse, #SavethePlanet, #ClimateCrisis, #planetclassroom Mirrored from: https://www.youtube.com/watch?v=Dzl_sx2rPgAhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/dc46da5c-d764-4110-a00c-0966a3283e69/f1dfe3b8-7f7d-42be-a44a-9d76f4161ccd-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/tcDcPLhSGbiXNVpXy6AHvT509012024-04-13T12:12:51.236ZclimatepolicyYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/m89L5hCqgf4Jcp1qD3MHEmhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/fd4948c6-b014-4399-86e5-a0c99a0cc28f.jpgUncertainty at Scale - how Computer Science hinders Climate Science"Computer science is a powerful tool for enabling data-driven advances in global ecology and conservation," says PhD student Patrick Ferris. "But the amplification cuts two ways, as mechanisation can also compound problems inherent with just how uncertain anything to do with natural ecosystems are! "In this short talk, I lookd at the different ways computer science amplified uncertainties in our work to analyse avoided deforestation projects in tropical moist forests."https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/a2f660c8-fd90-4e6e-a154-254b846b21bc/f948e3a3-34de-499b-b079-4741e8c81a54-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/m89L5hCqgf4Jcp1qD3MHEm537012024-04-13T17:36:53.620ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/wrjcJLXYCRjxA1sENch22thttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ab3d9fa6-e186-48b1-8cd6-cdae175f5e7c.jpg[PROPL'24] Discussion and brain storming: How can the CS/PL community help address the cur...[PROPL'24] Discussion and brain storming: How can the CS/PL community help address the current planetary crises? Dominic Orchard, Anil Madhavapeddy No description availablehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f67af8be-a8b7-45e3-b42d-6ee3df23b0e9/2bd23d51-2cea-4f40-9db9-0912219bab97-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/wrjcJLXYCRjxA1sENch22t1028042024-04-13T18:33:54.927ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/owMMPGXfJAZKWTcY84WCGchttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/74120273-3084-443f-a386-cbd4ba4164db.jpg[PROPL'24] Can computer science help climate policy making?[PROPL'24] Can computer science help climate policy making? Nicola Botta, Patrik Jansson The rational of the workshop is that, to tackle the crises induced by CO$_2$ emissions, we need to translate “a wealth of new data about our natural environment” into trustable insights that allow “to make decisions that affect the lives of billions of people worldwide” but that the computer systems currently available for this translation are not adequate. Hence the need to “close the gap” between state-of-the-art programming methods being developed in academia and climate science. We argue that this analysis is correct but too narrow and that computer science can contribute more than by “just” helping programming the computer systems that are needed to process planetary data. We propose a computational, Leibnizian approach in dealing with the climate crisis and outline work done over more than a decade in this direction. (A sequence of papers related to this are available online here: https://github.com/DSLsofMath/FPClimate )https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b675952e-3ad7-47f0-921d-09837d0a79ab/264ce44c-53ce-4c73-a7a2-80cbddaddc4c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/owMMPGXfJAZKWTcY84WCGc1602012024-04-13T18:34:05.713ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/gyxUz5FaRaQuSYAQEVaP6hhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/094f8424-df70-482b-9fc5-e4a9239dba2c.jpg[PROPL'24] Scalable agent-based models for optimized policy design: applications to the ec...[PROPL'24] Scalable agent-based models for optimized policy design: applications to the economics of biodiversity and carbon Sharan Agrawal No description availablehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/7e048b25-fe8c-4343-b687-377bd9243d46/a8291bbf-ba37-4612-aa05-843ed2c1e7f7-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/gyxUz5FaRaQuSYAQEVaP6h12380102024-04-13T18:34:14.816ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/kfmWHK5rXVrx6T3fwogx47https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b80be2b2-0f2f-4428-bc31-4840666cbd06.jpg[PROPL'24] Discussion on multidisciplinary PROPL-work[PROPL'24] Discussion on multidisciplinary PROPL-work Patrick Ferris, Michael Dales No description availablehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/9bdef444-d3a5-438d-b865-9b383a5bfe08/74a48ede-9a11-421f-87c4-cafc7eb67528-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/kfmWHK5rXVrx6T3fwogx471919012024-04-13T22:24:21.761ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/3nGExywoVm6XFRBA2zYxSLhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/aafe2bbb-a9b0-4bd7-8f0c-cdbc8bfb57f8.jpg[PROPL'24] Toward a Live, Rich, Composable, and Collaborative Planetary Compute Engine[PROPL'24] Toward a Live, Rich, Composable, and Collaborative Planetary Compute Engine Alexander Bandukwala, Andrew Blinn, Cyrus Omar Addressing the climate crisis poses many computing challenges for a variety of stakeholders, many of whom are not CS experts but rather scientists, policymakers, journalists, and members of the public. In order to solve these challenges there needs to be a large-scale collaborative compute engine that is live, rich, composable, and collaborative. Specifically, we present Planet Hazel, a vision of the Hazel programming environment geared toward planetary computing.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1339f4f8-2f31-44a7-a770-6f468cfb8f4c/1d1e74a2-b72f-4f7e-9d31-7d56e4c3b985-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/3nGExywoVm6XFRBA2zYxSL1302032024-04-13T22:24:30.581ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/arxCNHWWFZME4M86WXvBmvhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4ba0ebe7-5293-43e6-9cfa-d23ec9e8adb8.jpg[PROPL'24] Fluid: towards transparent, self-explanatory research outputs[PROPL'24] Fluid: towards transparent, self-explanatory research outputs Joe Bond, Cristina David, Minh Nguyen, Roly Perera This talk will introduce a “transparent” programming language called Fluid which incorporates a bidirectional dynamic dependency analysis into its runtime. Fluid keeps track of dependencies as outputs, such as charts or tables, are computed from data, and automatically enriches rendered outputs with interactions allowing a reader to interactively explore their relationship to inputs.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4c735672-f83f-405d-bd21-cc7ffff59909/739af3cb-be47-4110-b94d-4c50b3f31f81-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/arxCNHWWFZME4M86WXvBmv1145062024-04-13T22:24:38.780ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/w1gDefyifHmrZv86DTX3phhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c372d2e3-e405-4868-8636-3f325a34ce15.jpg[PROPL'24] Assessing the availability, reproducibility and reuseability of research softwa...[PROPL'24] Assessing the availability, reproducibility and reuseability of research software Vashti Galpin Different communities use different processes to assess the availability, reproducibility and reusability of research software and other research artifacts such as datasets. The Programming Languages (PL) community (together with other Computer Science communities) uses the artifact evalution process. Outside of the computer science, the FAIR (Findable Accessible Interoperable Reusable) principles for research data have been developed in the open science community and are now applied to research software. Within climate science, there has been a particular focus on climate modelling. The goal of this presentation is to describe some of these different assessment processes with the aim of promoting discussion and potential collaboration.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f2fbcca8-671a-4be5-806c-0c11c12c9c36/abee9f8e-2e2a-4d3a-b855-aa333eac8724-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/w1gDefyifHmrZv86DTX3ph1289012024-04-13T22:24:47.364ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/fqSh6LUVBHWK3r9rL6Kz7Vhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/3d017a8c-a7fe-4b1e-ac2c-f4ace1e08e06.jpg[PROPL'24] Kepler Watt Store: Kepler Software Watt Watcher StoreTitle Kepler Watt Store: Kepler Software Watt Watcher Store Parul Singh, Huamin Chen, Christophe Laprun Efficient power utilization has become pivotal in modern software development and deployment strategies. This abstract presents the comprehensive framework of Kepler Watt Store, a set of Cloud Native power-aware tools that incorporate Kepler and Peaks at various stages to monitor and optimize power efficiency throughout the software lifecycle, from development to deployment. Two tools proposed in this abstract, Runtime Monitoring and Power-Efficient Scheduling, are based on existing open-source projects—Kepler and Peaks—and are driven by community collaborations. Continuous Integration and Development Phase are research ideas that we are exploring for their viability in integrating with current software development and CI infrastructure. Some tools in the Kepler Watt Store focus on software development, while others target infrastructure. On the infrastructure side, we propose an operator-based approach to monitor infrastructure power consumption, ensuring backward compatibility for existing infrastructure and future-proofing for new software. Kepler serves as the cornerstone of this framework, functioning as a sophisticated power observability framework. It is a Kubernetes-based Efficient Power Level Exporter that estimates power consumption at the process, container, and Kubernetes pod levels. The architecture is designed to be extensible, allowing industrial and research projects to contribute novel power models for diverse system architectures. In more detail, Kepler utilizes a BPF program integrated into the kernel’s pathway to extract process-related resource utilization metrics. Kepler also collects real-time power consumption metrics from the node components using various APIs, such as Intel Running Average Power Limit (RAPL) for CPU and DRAM power, NVIDIA Management Library (NVML) for GPU power, Advanced Configuration and Power Interface (ACPI) fo...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/74d8c6c0-bea9-45a1-b45b-110528fcb9fd/70b2cd02-0ad1-4493-8ae6-0bcc6e5476eb-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/fqSh6LUVBHWK3r9rL6Kz7V1603022024-04-13T22:24:56.097ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/buhZDvtPuip8ceXSy99HBvhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/6e2d60e3-2383-416f-a0f2-1f946280db2f.jpg[PROPL'24] Formal Methods to Save the Earth[PROPL'24] Formal Methods to Save the Earth Hongyi Huang, Jialin Li, Umang Mathur With the rise of power demands for energy-greedy applications, such as large-scale data centers, machine learning models, and block-chain computation, higher proportion of energy consumption and carbon emissions can be attributed to computing machines [Manyika et al . 2011] [Liu 2013]. It is crucial to make computing more sustainable and environment-friendly. In a race with time, enabling regulatory bodies to enforce strict energy budgets on computing applications may fast track the progress towards the goal of green computing. Towards this, we revisit some approaches to analyzing the energy consumption of software and hardware applications, and analyze their strengths and weaknesses. We then turn to deductive methods and propose the operational semantics behind a program logic, which we call Power Logic, that can be used to formally specify and verify energy consumption. This set of operational semantics is abstracted from real-world computer systems and aims to reflect the ground truths about energy consumption in a system during small steps of computation. On top of these small-step rules, we hope to prove or specify the energy consumption of a certain program. Finally, we discuss potential directions in which Power Logic can further evolve.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/54eeaf5e-00c4-4293-8a0f-c682c6c9798f/dabffa1e-9415-48e5-859c-fa8d294d26e6-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/buhZDvtPuip8ceXSy99HBv1364052024-04-13T22:25:04.530ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/2kEyG3VQgVNRP3Yyd8fMi6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/04f07f7d-97ca-4aba-87ba-4cfd35613732.jpg[PROPL'24] Categorical Composition of Discrete Exterior Calculus Climate Models[PROPL'24] Categorical Composition of Discrete Exterior Calculus Climate Models Luke Morris, George Rauta, James Fairbanks The Decapodes.jl framework offers a domain-specific language for specifying systems of multiphysics equations, and formally composes such complex systems via undirected wiring diagrams. Since equations are encoded in the Discrete Exterior Calculus (DEC), these composed Decapodes can be compiled, automatically generating performant simulation code. Composition of Decapode diagrams along physical quantities manages the complexity that arises when bringing together multiphysics systems in a climate modeling context. In this talk, we briefly cover the Decapodes.jl framework and three example problems: the Halfar glacial model, the Halfar model’s composition with the Budyko-Sellers energy balance model, and the Klausmeier vegetation model.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/0ad80935-691b-4098-bced-635d2ca5c053/15b6ed75-8a1a-4f98-b97c-0ad68513bb88-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/2kEyG3VQgVNRP3Yyd8fMi61306002024-04-13T22:25:11.823ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/4sHR1xAj1NPdrkYxpYsHcWhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4d7d339e-9852-40be-a3d9-ed0e48e7f320.jpg[PROPL'24] The programming challenges of climate data analysis[PROPL'24] The programming challenges of climate data analysis Ezequiel Cimadevilla The ongoing evolution of the climate system and assessment of climate change requires sophisticated tools and methodologies for the analysis of vast and complex datasets generated by climate models, emerging records of satellite data and observational datasets. The pipelines and workflows involved in this complex task are performed in several phases that include multi-year and international planning, acquisition of data from heterogeneous data sources, complex infrastructures supporting the distribution of information and global collaborations for evaluation and archival. This work provides an update on concrete state-of-the-art methods currently used in data analysis workflows of climate data generated by Global Climate Models, focusing on the activities of the Climate Model Intercomparison Project (CMIP) and its underlying data infrastructure, the Earth System Grid Federation (ESGF). The presentation will dive into different topics of interest for practitioners and software developers of climate data analysis tools, including an overview of the software libraries from the Python ecosystem used for multidimensional climate data analysis and storage, a formal definition of climate data in the scope of the relational data model and an overview and description of requirements of different storage systems (HPC and cloud) used by the climate community.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1c068ab9-5e6b-40a4-9829-6646fab26528/e7b61403-f0a2-48da-8978-fa6a6c5b5036-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4sHR1xAj1NPdrkYxpYsHcW748002024-04-13T22:25:18.783ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/omPJqWDXch9YNq2QEC7vkYhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b443743a-561a-4c88-b579-b0ae862ee585.jpg[PROPL'24] Building Open Source Software for Climate Change Research — Lessons Learned fro...[PROPL'24] Building Open Source Software for Climate Change Research — Lessons Learned from Mimi.jl Lisa Rennels Mimi.jl is an open-source computational platform for integrated assessment modeling, a key tool in assessing coupled human-natural systems to evaluate the economic impacts of future climate change used widely in policy analysis, including recent work for the US government. This is an important step in enabling research and collaboration between academic research groups and with policy-making communities. Mimi also serves as a case study for developing tools and languages to support work by domain-specific communities that do not necessarily possess formal computer science training. In this talk I will first present the Mimi framework and the research-policy nexus at which it sits. Second I will present our recent study which employed observations and semi-structured interviews to (1) direct observe domain experts’ user experience with a climate change economics embedded domain-specific language (eDSL) (2) examine the impact of eDSL design and implementation decision son user experience and (3) suggest design considerations for future eDSLs.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b511576e-c731-4d66-8027-1bf86250c50a/4543a115-840e-4bc5-9eb1-f549f512b9be-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/omPJqWDXch9YNq2QEC7vkY2528032024-04-13T22:25:31.358ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/iGBKuvuc9fNNUCW5sBpQn4https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/340d08f8-cbe2-492a-9ab6-0ae36acac4ec.jpg[PROPL'24] Setting the stage for AI for biodiversity[PROPL'24] Setting the stage for AI for biodiversity Drew Purves We are suddenly experiencing an unprecedented demand for actionable information on biodiversity from governments, NGOs, and corporations worldwide. These stakeholders require biodiversity information that is finer grained (in space and time), more accurate, and more comprehensive than anything available today. Obviously AI (by which we mean scaled up deep learning) has great potential to both provide and interpret this next generation of information – from gathering more ecological data; to integrating that data into more useful information; to helping groups of stakeholders to make informed decisions against that information. However, this kind of transformative progress in AI for biodiversity will require a new level of integration of data, organizations, people, and technology. After an expanded version of the above intro, touching on some specific opportunities and challenges, I’ll raise the (open!) question of how, and how deeply, we should try to build what we might call nature’s schemas (i.e. phylogeny, traits, interactions, geographic and environmental space, and ecological time) into the next generation of planetary computing platforms.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8f5723c0-1b05-46fc-b3f0-de6dc47e548d/b5bedf45-91a0-4ed4-9038-98e400fbbb94-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/iGBKuvuc9fNNUCW5sBpQn428555162024-04-13T22:25:45.333ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/r5Se12vQoxGmjLmNezAVixhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b020cd55-cc40-48fd-870a-04aa78fd9ebb.jpgMaths, Models and Melting Ice - Ep78: Emily ShuckburghProfessor Emily Shuckburgh OBE is Director of Cambridge Zero, the University of Cambridge's major climate change initiative. Emily is a climate scientist and mathematician, a Fellow of the Cambridge Institute for Sustainability Leadership, an Associate Fellow of the Centre for Science and Policy and a Fellow of the British Antarctic Survey. She leads the UKRI Centre for Doctoral Training on the Application of AI to the study of Environmental Risks. A polar expert, she previously led a UK national research programme on the Southern Ocean and its role in climate. In 2016 she was awarded an OBE for services to science and the public communication of science. She is co-author with HRH The Prince of Wales and Tony Juniper of the Ladybird Book on Climate Change. Further reading: Official bio https://www.cst.cam.ac.uk/people/efs20 Cambridge Zero https://www.zero.cam.ac.uk/ Cambridge Zero Policy Forum https://www.csap.cam.ac.uk/Research-Policy-Engagement/cambridge-zero/ AI4ER https://ai4er-cdt.esc.cam.ac.uk/ Centre for Landscape Regeneration https://www.clr.conservation.cam.ac.uk/ Originally from https://www.youtube.com/watch?v=4PNzUbQJS7Uhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/cb2264af-ea36-467c-99aa-a4203ac3af6d/b4f84294-f554-437d-8ab4-f778e5d4515c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/r5Se12vQoxGmjLmNezAVix3908012024-04-13T22:26:11.775ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/ntRDot5MpBLtxhmJgtsatRhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e58a0b0a-5287-48d1-b98e-260c2028de21.jpgBringing Scale and Trust to Carbon Credits Through Computer SciencePresented by S. Keshav, University of Cambridge at the ICCS Summer School 2022 Held at the Centre for Mathematical Sciences, University of Cambridge, 20-22 Sept 2022. Carbon credits–especially those derived from nature-based solutions such as reforestation or averted deforestation–are deservedly viewed as being untrustworthy and their use by airlines and oil companies a barely-concealed form of greenwashing. In this talk, I will present a so- lution to these issues that leverages advances in earth observation, AI, cloud storage, and blockchain. This solution is being prototyped by the Cambridge Center for Carbon Credits (https://4c.cst.cam.ac.uk) and I will discuss the current status and our vision for the future. Originally from https://www.youtube.com/watch?v=cSK537TT-C4https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/adf39696-143e-401d-bafe-81426150b433/5d7f3d3b-1b4d-47af-a256-a24e737501ce-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ntRDot5MpBLtxhmJgtsatR3526022024-04-14T00:02:41.591ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/dD1H4v6mJZPjfDUksgMyhAhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5bd2744c-6905-4a77-b4c5-f3a94615dc77.jpgAndy Hopper: Computing for the Future of the PlanetDigital technology is becoming an indispensable and crucial component of our lives, society, and the physical environment. A framework for the role of computing in dealing with sustainability of the planet will be presented. The framework has a number of goals: an optimal digital infrastructure, sensing and optimising the physical world with a global model, guaranteeing the performance of indispensable systems, and digital alternatives to physical activities. Practical industrial examples will be given as well as longer-term research goals. About the Speaker Andy Hopper is Professor of Computer Technology at the University of Cambridge and Head of Department of the Computer Laboratory. His research interests include computer networking, pervasive and sensor-driven computing, and using computers to ensure the sustainability of the planet. Andy Hopper has pursued academic and industrial careers simultaneously. In the academic career he has worked at the Computer Laboratory and the Department of Engineering at Cambridge. In the industrial context he has co-founded over a dozen spin-outs and start-ups, three of which floated on stock markets, as well as working for multinational companies. He is Chairman of RealVNC Group. The companies he co-founded have received numerous Queen’s Awards for Enterprise and the MacRobert Award. Professor Hopper received the BSc degree from the University College of Swansea (1974) and the PhD degree from the University of Cambridge (1978). He is a Fellow of the Royal Academy of Engineering (1996) and of the Royal Society (2006). He was made a CBE for services to the computer industry (2007). He is a past President of the Institution of Engineering and Technology (2012). In 2014 the Science Council named him as one of UK’s “100 Leading Practising Scientists”, and the Sunday Times listed him as one of the “100 most influential Britons of the modern age”.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/66589ea4-1f87-4e33-8169-98e4c94dd13a/07a7068b-23f5-43d5-89f6-df7435e66898-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/dD1H4v6mJZPjfDUksgMyhA3981022024-04-14T00:03:04.794ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/qaofRR1NkFQ1oD4su25M8Khttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a8736f23-de39-4b52-84e6-3b1780fe7bb7.jpgUpcoming BIOMASS Earth Explorer 7 mission: concept and validation challengesJerome Chave: Upcoming BIOMASS Earth Explorer 7 mission: concept and validation challenges The ESA BIOMASS mission aims to provide wall-to-wall aboveground biomass map products globally every year with an accuracy of 20% at the 4-ha scale. This will reduce the major uncertainties in carbon fluxes linked to land use change, and forest degradation, will provide support for nationally determined contributions, will become an essential input to the climate modelling community, and will provide key information on forest resources and ecosystem services. This is a technological challenge, as this satellite will be the first ever to embark a radar sensor at P band. Validation is a key component, and relies on the long-term onsite commitment of forestry and ecological science. Such an ambitious science and technology project should benefit to the Global South. Jerome Chave is senior scientist with CNRS based in Toulouse. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c3aabcd5-daf8-4fc6-9f44-837a0adeb745/2f74e6e8-33d2-4b44-8767-1fe884c2985f-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/qaofRR1NkFQ1oD4su25M8K3273002024-04-14T01:26:01.666ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/9YFmPtXS5Q2tCiA2NV5N8Ahttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/2181d5f3-7217-4c2d-bcb7-37d1ba755ce3.jpgFresh Thinking at Fitz: What next for UK trees?Conversation with Professor David Coomes, recorded on Thursday 28th January.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/48b30426-5609-4956-a483-aa40940e5620/40402480-4e4d-4f41-bd79-9da8cf1a355b-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/9YFmPtXS5Q2tCiA2NV5N8A3374012024-04-14T01:26:21.457ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/ixwketKS6Y52Bqci4EN63xhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/65bad799-ef88-4a1e-ab38-b6b3ab59775e.jpgPhone based measurement algorithm to assess forest healthAssessing the health of forests using rapid phone-based measurement algorithm is the next step forward for forest remote sensing research. Measuring tree diameter is critical to monitoring forest health and carbon sequestration levels. Streamlining this process allows scientists to determine the health of a forest and understand the effects on the wider forest ecosystem. Researchers from the University of Cambridge have developed an algorithm that provides an accurate measurement of tree diameter using low-cost, low-resolution LiDAR sensors incorporated into mobile devices. This novel approach provides accurate and timely results compared to previous manual measurement techniques. In this study, the researchers developed the algorithm to accurately measure trees almost five times faster than traditional methods. The research, published in Remote Sensing, discusses how the highly valuable automated measurement technique advances scientists’ ability to calculate a wide range of tree variability efficiently. Previous manual methods to discover how much carbon a forest is sequestering are hugely valuable, but also time-consuming, explains first author Amelia Holcomb from Cambridge’s Department of Computer Science and Technology. “We wanted to develop an algorithm that could be used in more natural forests, and that could deal with things like low-hanging branches, or trees with natural irregularities,” said Holcomb. The researchers collected their own dataset and trained the algorithm to differentiate trunks from larger branches and determine tree angle of lean using image processing and computer vision techniques. The algorithm app was tested in three different study sites – one each in the UK, US and Canada – in varying seasons. Since the measurement tool requires no specialised training and uses sensors that are already incorporated into an increasing number of phones, it could be an accurate, low-cost tool for forest measurement, even in com...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8e121b4a-035a-428d-bad5-4cefeb410e77/e1ca8aa0-9cd3-47c4-849d-6d8e368cb4b7-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ixwketKS6Y52Bqci4EN63x32012024-04-14T02:21:08.997ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/bRX4oratYEq75PM8p9U5r9https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b8874945-3c53-4be4-a2ae-e45fcdae35f5.jpgFusing GEDI and LandSat data to estimate tropical forest recovery rates (Amelia Holcomb)Amelia Holcomb: Fusing GEDI and LandSat data to estimate tropical forest recovery rates across the Amazon Tropical secondary forests are ecosystems of critical importance for protecting biodiversity, buffering primary forest loss, and sequestering atmospheric carbon. Monitoring the growth and sequestration patterns of secondary forests has historically been difficult at scale, but the recent launch of the Global Ecosystem Dynamics Investigation (GEDI), a space-borne LiDAR sampler, provides accurate aboveground biomass density (AGBD) estimates across the tropics. However, fusing GEDI data (25 m circular samples with geolocation uncertainty) with historical forest change maps derived from LandSat (30 m x 30 m square wall-to-wall pixels) remains a challenge. In this work, we propose a generalizable Monte Carlo-based method for fusing GEDI and LandSat-based maps while robustly propagating uncertainty. The method also allows flexible filtering for high-confidence data points, and we provide open-source code for distributing the computation. Using this novel approach, we estimate the carbon sequestration rate of regrowth forest across the Amazon. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/57f4c324-227b-4b5c-8517-aabd8309f782/32d9f545-7f23-413c-8496-a1b025e37e68-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/bRX4oratYEq75PM8p9U5r93444002024-04-14T16:58:30.633ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/85L3LuC1KJJJPHNcFRtCa5https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ebec9602-c6f4-4a6f-8753-78ed67bbd1b9.jpgOptimization-in-the-loop AI for energy and climate (Priya Donti)Priya Donti: Optimization-in-the-loop AI for energy and climate Addressing climate change will require concerted action across society, including the development of innovative technologies. While methods from artificial intelligence (AI) and machine learning (ML) have the potential to play an important role, these methods often struggle to contend with the physics, hard constraints, and complex decision-making processes that are inherent to many climate and energy problems. To address these limitations, I present the framework of “optimization-in-the-loop AI,” and show how it can enable the design of AI models that explicitly capture relevant constraints and decision-making processes. For instance, this framework can be used to design learning-based controllers that provably enforce the stability criteria or operational constraints associated with the systems in which they operate. It can also enable the design of task-based learning procedures that are cognizant of the downstream decision-making processes for which a model’s outputs will be used. By significantly improving performance and preventing critical failures, such techniques can unlock the potential of AI and ML for operating low-carbon power grids, improving energy efficiency in buildings, and addressing other high-impact problems of relevance to climate action. Priya Donti is the Co-founder and Executive Director of Climate Change AI, a non-profit initiative to catalyze impactful work at the intersection of climate change and machine learning, which she is currently running through the Cornell Tech Runway Startup Postdoc Program. She will also join MIT EECS as an Assistant Professor in Fall 2023. Her research focuses on machine learning for forecasting, optimization, and control in high-renewables power grids. Specifically, her work explores methods to incorporate the physics and hard constraints associated with electric power systems into deep learning models. Priya received her Ph.D. in Computer S...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/395a1750-0bd7-4e05-bc51-0d844416f786/cd60582c-19ae-4c4d-be76-7404d083b4e6-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/85L3LuC1KJJJPHNcFRtCa53611002024-04-14T16:58:47.612ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/8ywdeCj8aJBA2t7dMJ3gd8https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/456b9cba-431d-496b-84d8-4fbfef0a7477.jpgAssessing indirect business risk from climate change (Daoping Wang)Daoping Wang: Assessing indirect business risk from climate change Increasingly complex linkages of the global economy can be a channel for the propagation of negative climate-related shocks with the potential to deteriorate resilience and sustainability in societies. Given projected increases in disaster risks due to climate change, developing models that more accurately simulate the propagation of shocks through economic networks becomes crucial for designing effective prevention and recovery strategies. Except for some studies that use macroeconomic models to assess indirect risk at the sectoral level, indirect risk at the firm level is rarely well quantified. In this work, I propose an indirect risk assessment approach that bridges macroeconomics and microenterprises. This approach divides the indirect risk of enterprises into two parts: the average risk of the industry determined by the characteristics of the industry to which the enterprise belongs, and the risk coefficient determined by the characteristics of the enterprise itself. We use an agent-based model equipped with general equilibrium theory to evaluate the former, and machine learning techniques to evaluate the latter. On top of a more comprehensive risk assessment, the proposed approach provides a modeling framework for analyzing the indirect risk face by specific firms. Daoping Wang is a Hoffmann Research Fellow at the University of Cambridge. He worked with academics from the Department of Computer Science and Technology and the Judge Business School at the University of Cambridge, to explore the application of AI technologies in climate risk assessment and governance, especially the indirect risk propagated on global supply-chain networks. He is also in collaboration with the Centre for Climate Engagement in Hughes Hall and the World Economic Forum to bring knowledge on global climate governance to policymakers, stakeholders, and the general public. This talk is part of a seminar series fro...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3d3a63b0-df07-4fca-b9db-0c8267e90ceb/f1ef747f-2917-4175-b4e0-df4cb525bddc-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/8ywdeCj8aJBA2t7dMJ3gd82641002024-04-14T16:59:20.255ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/7neGeAbNLUUyPDG53dS7q6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/2afaa690-1ad5-42d2-939f-aea7dcab43ed.jpgMeasuring emissions by satellite to support the Paris AgreementStephen Briggs: Measuring emissions by satellite to support the Paris Agreement The Paris Agreement and its associated Global Stocktake are designed to assess progress towards emissions reduction. Individual countries report on the evolution of their emissions. Satellite observations can provide independent estimates of emissions but their relation to geographical location is extemely complex both in terms of the suite of observations required and the inverse modelling process. Stephen Briggs is the former Head of EO Science, Applications and Future technologies at the European Space Agency and was Chairman of the Global Climate Observing System during 2013-2019. He is the assessor for the EU Copernicus CO2 emissions programme and is a visiting Professor at Reading University Meteorology Dept and a visitor to the Cambridge University Chemistry Department. He is on the Board of the Cambridge Centre for AI and Environmental Risk. He has worked on the use of satellite observations for, inter alia, climate studies for more years than he would care to remember. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/338de3e2-213b-440b-8aaa-4ad64436801d/604736dd-d5d4-491d-b033-7fd2ca4414bb-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/7neGeAbNLUUyPDG53dS7q63631002024-04-14T16:59:09.028ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/sJpAvgUSp54ectBXpnL5ERhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/9bca8268-14e5-4af4-8b59-116a2862038b.jpgA framework for measuring biodiversity impacts of land use changeAlison Eyres: A framework for measuring biodiversity impacts of land use change If applied appropriately Nature-based carbon projects such as conservation, restoration and expansion of habitat offer an exciting opportunity to tackle both the climate and biodiversity crises. However, co-benefits are not assured and there is a growing concern that funding through carbon credits encourages creation of fast-growing monocultures that offer short term benefits of rapid sequestration and miss opportunities to support biodiversity. Although recognised as important, a lack of robust, consistent and scalable metrics for considering biodiversity impacts of projects has largely prevented meaningful comparisons. I plan to present the framework that we have developed within 4C to quantify the impact on biodiversity of land use change and two key applications. Firstly, to assess the biodiversity impact of carbon projects and secondly to examine the value of restoration and conservation to biodiversity globally. I would like to discuss ways in which to develop this framework further e.g. by incorporating additional threats, habitat quality and integrating on the ground observations. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/d8796266-ab6e-4853-a07d-60a608a2ec2d/8b420d6a-9fed-4aca-a7f1-ba370f6f5b98-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/sJpAvgUSp54ectBXpnL5ER2592002024-04-14T16:59:35.590ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/4sSMsuMjdDGFXrEkxfzW1ahttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/f6b6efec-8fea-4cea-9ff4-10d59a2b6d84.jpgChallenges in Designing Control Strategies for BuildingsOmid Ardakanian: Challenges in Designing Control Strategies for Buildings Building operation is responsible for a large share of global energy consumption and carbon emissions. Despite the growing adoption of sensors in buildings, controls are still simple, reactive, and must be customized for every building based on its size, floor plan, location, and occupancy schedule. In this talk I will give an overview of key challenges in designing safe and optimal control strategies for buildings, and present some of the recent work we did to address these challenges. Bio: Omid Ardakanian is an Assistant Professor in the Department of Computing Science at the University of Alberta. He received the BSc in Computer Engineering at Sharif University of Technology, and the MMath and PhD in Computer Science at the University of Waterloo. Before joining UofA, he was an NSERC postdoctoral fellow at the University of California, Berkeley, and the University of British Columbia. Dr. Ardakanian’s research focuses on the design and implementation of intelligent networked systems, such as smart buildings and grids. He received the best paper award at ACM e-Energy (2013), ACM BuildSys (2016, 2022), and IEEE PES General Meeting (2017). He was a guest editor of a special issue of IEEE Transactions on Smart Grid, and TPC co-chair of ACM e-Energy 2021. He is currently serving on the executive committee of ACM SIGENERGY and is an editor of SIGENERGY Energy Informatics Review (EIR). This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1c0c0cf3-848d-47a0-87c0-d608106a1ac9/5191e4ad-92d3-4fa5-b16f-874518ce3050-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4sSMsuMjdDGFXrEkxfzW1a3627002024-04-14T16:59:52.337ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/4zefDpfhoBLfT7RgxM1VVPhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/8b08a868-fbef-496d-a939-e347125ebef7.jpgQuantifying changes in Above Ground Biomass in degraded and restored forests (Charlotte Wheeler)Charlotte Wheeler: Quantifying changes in Above Ground Biomass in degraded and restored forests: Challenges & opportunities Forestry based solutions, including forest restoration and avoided deforestation and degradation are vital for climate change mitigation and adaption. Thus methods to quantify changes in above-ground biomass (AGB), in the form of emissions from forest degradation and removals from forest restoration are vital. Radar data has been widely used to map and quantify changes in AGB , however mapping these subtle changes in AGB over time is challenging. Here I discuss the use of L-band Radar data in mapping AGB change, highlighting the advantages and disadvantages. I will also discuss new opportunities for AGB mapping from new spaceborne sensors, detailing some of the methods we will be testing in the 4C project. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1cef20a5-e8f1-472b-b877-1ce7d45a6b15/e734dd05-3c55-427c-859e-4b85fb25adf6-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4zefDpfhoBLfT7RgxM1VVP3235002024-04-14T17:00:05.868ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/p3roLCwZbkPCCzh1haDWLjhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/eef04fde-9fba-410c-87cd-a9b61dcf7bca.jpgMachine Learning in Climate ActionDavid Rolnick: Machine Learning in Climate Action Machine learning (ML) can be a useful tool in helping society reduce greenhouse gas emissions and adapt to a changing climate. In this talk, we will explore opportunities and challenges in ML for climate action, from optimizing electrical grids to monitoring crop yield, with an emphasis on how to incorporate domain-specific knowledge into machine learning algorithms. We will also consider ways that ML is used in ways that contribute to climate change, and how to better align the use of ML overall with climate goals. David Rolnick is an Assistant Professor and Canada CIFAR AI Chair in the School of Computer Science at McGill University and at Mila Quebec AI Institute. He is a Co-founder and Chair of Climate Change AI and serves as Scientific Co-director of Sustainability in the Digital Age. Dr. Rolnick received his Ph.D. in Applied Mathematics from MIT . He is a former NSF Mathematical Sciences Postdoctoral Research Fellow, NSF Graduate Research Fellow, and Fulbright Scholar, and was named to the MIT Technology Review’s 2021 list of “35 Innovators Under 35.” This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ba995350-ca12-4b9c-ad3f-7bba9bd60ffa/a2b08eda-873e-45fb-9ecb-85863001f792-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/p3roLCwZbkPCCzh1haDWLj3532002024-04-14T17:00:22.694ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/qg3KyXuFQSUfUQCGEMZHTChttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/d7322ea7-1bc3-49e6-b827-4ee998a856a3.jpgEcovisor: A Virtual Energy System for Carbon-Efficient ApplicationsDavid Irwin: Ecovisor: A Virtual Energy System for Carbon-Efficient Applications The growth of cloud platforms is raising significant concerns about their impact on the environment. To reduce their carbon footprint, future cloud platforms will need to broadly adopt low- or zero-carbon energy sources, such as solar or wind. A distinguishing characteristic of clean energy is its unreliability. Unfortunately, today’s energy systems mask this unreliability in hardware, which prevents applications from optimizing their carbon-efficiency, i.e., work done per kilogram of carbon emitted. To address the problem, we design an ``ecovisor,’’ which virtualizes the energy system and exposes software-defined control of it to applications. Our ecovisor enables each application to handle clean energy’s unreliability within the software stack based on its specific requirements. We implement a small-scale ecovisor prototype that virtualizes a physical energy system capable of regulating power flow between the grid, a solar array, batteries, and a cluster of microservers. We evaluate our ecovisor’s flexibility by showing how a range of applications can exercise their virtual energy system in different ways to optimize carbon-efficiency. For example, we show how using a 1kWh battery to reduce renewable energy volatility decreases the running time and energy usage of an elastic Spark job by 4.4x by eliminating recomputation overhead due to power shortages. David Irwin is an Associate Professor in the Department of Electrical and Computer Engineering and an Adjunct Associate Professor in the College of Information and Computer Sciences at the University of Massachusetts Amherst. He leads the Sustainable Computing Lab, which focuses on designing distributed software systems with an emphasis on improving efficiency and sustainability. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c4752e0b-7514-429b-9a8d-ab7f91a81eee/c7e55069-8ac3-439c-8621-ff83b0d4bd9b-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/qg3KyXuFQSUfUQCGEMZHTC3217002024-04-14T17:00:39.642ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/vFu5CsuQoekMwLS7sTpGDNhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/dfb40bdb-7ce1-4364-ab25-dea9032a961a.jpgNew framework for carbon offsets in global drylandsAdam Pellegrini: New framework for carbon offsets in global drylands Drylands, which span an area five times the size of the Amazon, have massive potential to serve as either a carbon sink or source depending on shifting resource availability and disturbance events. In my talk, I will present the current theory and understanding of dryland carbon cycling, how ecosystem models capture these trends, and what challenges lie ahead. I will then present a new European Research Council Starter Grant project that aims to deliver models capable of forecasting nature-based climate solutions in drylands and collaborate with industry players. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f05c553d-10fa-4291-9d36-ace6b7dc02b8/5b5e2eef-daae-4c6c-a7aa-bfceaf4c332f-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/vFu5CsuQoekMwLS7sTpGDN3088002024-04-14T17:00:55.200ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/t3SU1tRd5Hi7j4EpND4SXEhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/651b91cd-caeb-4ee2-8827-744f48237d9a.jpgCompetitive Online Peak-Demand Minimization using Energy StorageMinghua Chen: Competitive Online Peak-Demand Minimization using Energy Storage We consider an increasingly popular demand-response scenario where large-load customers, e.g., datacenters, utilize energy storage to reduce the peak procurement from the grid, which accounts for up to 80% of their electric bills. We focus on minimizing the peak-demand charge using energy storage under the online setting, where the loads and renewable generations are revealed sequentially in time but we have to make irrevocable decisions at current epoch with little or no future information. Such an online problem is uniquely challenging due to (i) the coupling of irrevocable decisions across time imposed by the inventory constraints and (ii) the noncumulative nature of the peak procurement. We tackle this issue by developing an optimal online algorithm for the problem that attains the best possible competitive ratio (CR) among all deterministic and randomized algorithms. We show that the optimal CR can be computed in polynomial time, by solving a linear number of linear-fractional problems. More importantly, we generalize our approach to develop an anytime-optimal online algorithm that achieves the best possible CR at any epoch, given the inputs and online decisions so far. The algorithm retains the optimal worst-case performance and achieves adaptive average-case performance. Simulation results based on real-world traces show that, under typical settings, our algorithms improve peak reduction by over 19% as compared to baseline alternatives. This is a joint work with Yanfang Mo, Qiulin Lin, and Joe Qin, all from City University of Hong Kong. This talk is part of a seminar series from the Energy and Environment Group, Department of Computer Science and Technology, University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/db0d98f9-d4e5-43ce-9539-e03b648f77ec/9211284e-9758-4308-9375-b103febdd61a-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/t3SU1tRd5Hi7j4EpND4SXE3290002024-04-14T17:01:14.371ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/aFUFSzJ4n2X6nQFf4EVdu3https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/d7151c9f-f85c-42e4-8900-0237aabf0025.jpgRemote sensing enables monitoring life above and under waterRemote sensing enables monitoring life above and under waterhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4e74b98a-2bc0-4b2b-a524-6898f21fbe92/6e907b6d-f27f-4cbf-9094-b114f7f4a9a7-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/aFUFSzJ4n2X6nQFf4EVdu33295002024-04-14T17:01:27.457ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/qxYRpu8gNG2RzTxDiGsmJQhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/16d259e6-0391-45b8-9336-583d5e7aba0d.jpgAssessing the Impact of Drought on Tropical Forests using GEDIAssessing the Impact of Drought on Tropical Forests using GEDIhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c6d26bbe-e8ac-463e-a763-0bc832661874/da3cfbf4-a0a4-479a-b95b-6139c49f0cb5-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/qxYRpu8gNG2RzTxDiGsmJQ3545002024-04-14T17:01:48.655ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/u3jmHnFXbp7siwD7MTv1H6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/23542596-6757-4ac4-b365-e24bfb4fa38e.jpgGenerative model in remote sensing the role and applications of radiative transfer models in the optGenerative model in remote sensing the role and applications of radiative transfer models in the opthttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/e312aed9-2633-4ac9-90c8-e237cb065127/c5b9bdc3-4857-462e-8c30-4d099d6936f0-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/u3jmHnFXbp7siwD7MTv1H63310032024-04-14T17:02:02.100ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/o6rcihw7SCb7VHSNB7biddhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/671f5b6e-1c89-487e-9c4e-60e8b384f97e.jpgDesigning PV-EV integrated Residential Microgrids in the post-COVID WorldDesigning PV-EV integrated Residential Microgrids in the post-COVID Worldhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b2eb4bd4-9acd-4bbd-926b-86d5e9bfaa78/916fbb49-26f5-46b2-a3d6-13b80b755a38-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/o6rcihw7SCb7VHSNB7bidd2990002024-04-14T17:02:15.165ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/7z1QcpDF69BPdYFu7Z3cQzhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/206bb972-c18a-47c9-a951-412f3d5e9e45.jpgTransforming Your Understanding of Large Language Models: A PrimerThis talk gives an introduction to Large Language Models for non-specialists, especially those who might be considering applying them to nature or climate projects. We will cover the recent history of language models and then deep dive on a modern transformer-based language model. We will look at tokenisation, training methods, preference alignment and the future trends in model architectures. Bio: https://toao.com/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3532e373-49c0-4682-bd05-ebaf02c6e6ad/149b5582-d7fb-4efa-9075-bfc2488a7f29-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/7z1QcpDF69BPdYFu7Z3cQz3320072024-04-19T15:57:54.661ZllmsaiYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/aaY3rcVQmibWxQhx9KYaGrhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ec79ffae-2195-4244-bf56-4ce82967a186.jpgUsing AI to Accelerate Evidence Synthesis and Decision Support to Save BiodiversityConsider a planner laying a new train line through ecologically sensitive areas. They need to balance economic benefits with ecological damage and think through mitigations such as fencing, animal overpasses or habitat compensation areas, each of which have costs. To maximise benefits and reduce costs, such decisions must be informed by prior research knowledge. However, scientific evidence is difficult to find, synthesise and assess. The Conservation Evidence (www.conservationevidence.com) has manually screened over 1.5M papers in 17 languages, so far distilled to 8,636 relevant studies that test 3,690 conservation actions for 24 species groups, habitats and other conservation issues. However, the laborious process of evidence gathering and synthesis means we can only add a few hundred studies per year, whilst the rate of evidence generation is increasing each year. In this talk, I'll explain how we are planning to accelerate the evidence synthesis pipeline and improve decision support using Large Language Models to ensure future conservation action is informed by scientific research. Bio: Alec is a Henslow Research Fellow at Downing College Cambridge and works with the Conservation Evidence group in the Department of Zoology and Cambridge Conservation Initiative (CCI). He did his PhD in Zoology in the same group in 2017 and before that he studied for a BSc in Marine Biology at the University of St Andrews. His work has focused on understanding the gaps and biases in the evidence for biodiversity conservation and improving decision support to enable more evidence-based conservation.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4a467a9d-69ea-44de-a191-6e56898e6e9d/9590da42-59ba-4ccd-bbcf-2247ed72b3ad-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/aaY3rcVQmibWxQhx9KYaGr3568002024-04-26T14:45:58.305ZllmsaiconservationbiodiversitycoeYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/ot9M3LTp18sB8zsmaLfk4yhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e2f726bb-9064-408d-b49a-00ca81537c5e.jpgUsing AI to Accelerate Evidence Synthesis and Decision Support to Save BiodiversityConsider a planner laying a new train line through ecologically sensitive areas. They need to balance economic benefits with ecological damage and think through mitigations such as fencing, animal overpasses or habitat compensation areas, each of which have costs. To maximise benefits and reduce costs, such decisions must be informed by prior research knowledge. However, scientific evidence is difficult to find, synthesise and assess. tHE Conservation Evidence (www.conservationevidence.com) has manually screened over 1.5M papers in 17 languages, so far distilled to 8,636 relevant studies that test 3,690 conservation actions for 24 species groups, habitats and other conservation issues. However, the laborious process of evidence gathering and synthesis means we can only add a few hundred studies per year, whilst the rate of evidence generation is increasing each year. In this talk, I'll explain how we are planning to accelerate the evidence synthesis pipeline and improve decision support using Large Language Models to ensure future conservation action is informed by scientific research. Bio: Alec is a Henslow Research Fellow at Downing College Cambridge and works with the Conservation Evidence group in the Department of Zoology and Cambridge Conservation Initiative (CCI). He did his PhD in Zoology in the same group in 2017 and before that he studied for a BSc in Marine Biology at the University of St Andrews. His work has focused on understanding the gaps and biases in the evidence for biodiversity conservation and improving decision support to enable more evidence-based conservation.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b5f38a87-d43b-4883-8564-557aca39db0a/8f3c841e-a735-40a9-b470-7ee37ae4778c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ot9M3LTp18sB8zsmaLfk4y3568022024-04-29T20:41:18.381ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/qUZF6cyxeoDs5XFsYZqvkihttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/6fa2e76d-d27a-4e90-854b-3a3bdc6a5861.jpgDigital Participatory Tools for Rural Communities in India to Adapt to Climate ChangeWith weather patterns becoming erratic, rural communities in India dependent upon agriculture, livestock, and forests for their sustenance face an intersecting crisis of environment, livelihood, and social justice. Navigating this crisis requires a multi-dimensional approach of sustainable natural resource management, done in an equitable manner to benefit the most marginalized populations, and with collectivization efforts to improve consensus building and cooperation in communities. Can data and digital technologies play a role here? I will describe the complexity of socio-ecological problems in the context of rural central India and opportunities for ICT-based interventions that can enable communities to build a shared understanding of changes taking place in their landscape, use it to plan and demand natural resource management works, and bring changes in their day-to-day resource utilization and regeneration practices. Our work leverages geospatial algorithms, machine learning on satellite data, and novel data oralization and visualization ideas, that sit in a technology stack of building blocks on which further new innovations can be created. We are also attempting a co-creation methodology to build this stack through collaboration across disciplines and borders, to solve for complexities that are beyond a single research group to manage. Bio: Aaditeshwar Seth is a Professor in the Department of Computer Science and Engineering at the Indian Institute of Technology Delhi, and co-founder of the social technology enterprise Gram Vaani. He is passionate about building appropriate technologies and participatory tools that can empower marginalized and oppressed communities to collectivize and voice themselves. Several million people, and over 150 organizations worldwide, have directly touched technology platforms built by Aaditeshwar’s team at Gram Vaani and his students at the ACT4D (Appropriate Computing Technologies for Development) research group at IIT...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c9c18bf5-d875-4b25-ab12-5c64ed4243c3/fb2d8ca7-d7f3-4073-a3bf-815ca1913686-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/qUZF6cyxeoDs5XFsYZqvki3865522024-05-03T16:43:17.906ZcoesensingclimateconservationbiodiversityYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/hz7GSpqZycMvZWzbiXGrfehttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/35b48716-211a-424e-8ed3-2f691a01071e.jpgAutomated Fact-Checking of Climate Change Claims with Large Language ModelsThis talk introduces Climinator, a novel AI-based tool designed to automate the fact-checking of climate change claims. Utilizing an array of Large Language Models (LLMs) informed by authoritative sources like the IPCC reports and peer-reviewed scientific literature, Climinator employs an innovative Mediator-Advocate framework. This design allows Climinator to effectively synthesize varying scientific perspectives, leading to robust, evidence-based evaluations. Our model demonstrates remarkable accuracy when testing claims collected from Climate Feedback and Skeptical Science. Notably, when integrating an advocate with a climate science denial perspective in our framework, Climinator's iterative debate process reliably converges towards scientific consensus, underscoring its adeptness at reconciling diverse viewpoints into science-based, factual conclusions. While our research is subject to certain limitations and necessitates careful interpretation, our approach holds significant potential. We hope to stimulate further research and encourage exploring its applicability in other contexts, including political fact-checking and legal domains. Bio: Dominik Stammbach just recently finished his PhD in Natural Language Processing at ETH Zurich and is an incoming postdoc at Princeton University in Fall 2024. Dominik's research interests include developing NLP methods which can be applied in the context of misinformation, online safety and developing methods at the intersection NLP and climate change. Among others, he wants to detect company greenwashing, the practice of companies making generic, misleading or false claims to boost their environmental credentials.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8631e514-e5b6-477c-b94c-10b61421450d/7e8a694b-1290-41c0-b4fc-73ac45927bfc-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/hz7GSpqZycMvZWzbiXGrfe3198022024-05-10T17:42:53.140ZllmsaipolicyYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/5Z7QzVfQEqpjo2KRVgPLKChttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/d96e8df4-d8e0-4112-941e-725c81d69bae.jpgAn App for Tree Trunk Diameter Estimation from Coarse Optical Depth MapsAbstract: Trunk diameter is related to the overall health and level of carbon sequestration in a tree. Trunk diameter measurement, therefore, is a key task in both forest plot and urban settings. Unlike the traditional approach of manual measurement with a measuring tape or calipers, several recent approaches rely on sophisticated technologies such as LiDAR and time-of-flight cameras that provide fine-grain depth maps. These technologies are supported only on specialized devices or high-end smartphones. We present a mobile application called GreenLens that only uses coarse-grain depth maps derived from an optical sensor, and so can be run on most common Android devices. Moreover, we use a state-of-the-art deep neural network to estimate trunk diameter from an image and its corresponding coarse depth map (RGB-D). We tested our app under challenging conditions including occlusion, leaning trees, and irregular shapes and found that our algorithm is comparable to accuracy from fine-grain depth maps. Currently, we are developing GreenLens2. Unlike our previous work, we are using a game engine (Unreal Engine) to create a highly photo-realistic virtual forest, making it easy to collect unlimited and diverse data for training neural networks. At the same time, we have proposed a multi-task neural network that performs trunk segmentation and end-to-end trunk diameter prediction simultaneously. We have also refined the app's user journey to make it more interactive, straightforward, and user-friendly. Bio: Frank Feng is currently an undergraduate researcher and will join the Department of Computer Science and Technology at the University of Cambridge as a PhD student in October 2024.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/285e29d4-21dd-45e2-abe3-0f026654ea3a/3867756d-df67-4837-acee-464299acb31c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/5Z7QzVfQEqpjo2KRVgPLKC2426022024-06-07T21:19:47.749ZcoeforestssensingmobileYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/nfXCsAiyawBPM2qzdpRCfBhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/29855b7a-fe9c-4344-9661-a2137554ff3b.jpgTackling the Hidden Costs of Computational ScienceTitle: Tackling the Hidden Costs of Computational Science: GREENER Principles for Environmentally Sustainable Research Abstract: From genetic studies and astrophysics simulations to AI, scientific computing has enabled amazing discoveries and there is no doubt it will continue to do so. However, the corresponding energy usage and environmental impacts are a growing concern in light of the urgency of the climate crisis, so what can we all do about it? Tackling this issue and making it easier for scientists to engage with sustainable computing is what motivated the Green Algorithms project. Through the prism of the GREENER principles for environmentally sustainable science, we will discuss what we learned along the way, how to estimate the impact of our work and what levers scientists and institutions have to make their research more sustainable. We will also debate what hurdles exist and what is still needed moving forward. Bio: Dr Loïc Lannelongue is a Research Associate in Biomedical Data Science in the Heart and Lung Research Institute at the University of Cambridge, UK, and the Cambridge-Baker Systems Genomics Initiative. He leads the Green Algorithms project, an initiative promoting more environmentally sustainable computational science. His research interests also include radiogenomics, i.e. combining medical imaging and genetic information with machine learning to better understand and treat cardiovascular diseases. He is a Software Sustainability Institute Fellow, a Post-doctoral Associate at Jesus College, Cambridge, and an Associate Fellow of the Higher Education Academy.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ac269b62-fbca-402a-a180-3facc44d5f37/2823b501-2763-4a2e-8b3d-9570f6c53407-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/nfXCsAiyawBPM2qzdpRCfB3477002024-06-21T15:03:59.936ZcloudenergyYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/pUULKS4bi9hG9uUy9prhF2https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a94c262b-56e5-4d03-aae7-777436e92114.jpgHyWay: Enabling Mingling in the Hybrid WorldWe present HyWay, short for “Hybrid Hallway”, to enable mingling and informal interaction among in-person and remote users in semi-structured and unstructured hybrid settings. Key to the design of HyWay is bridging the awareness gap between physical and virtual users and providing the virtual users the same agency as physical users. We explain how HyWay achieves these goals with an agile, light touch approach that centers on commodity hardware, frictionless bootstrapping, and a deploy-learn-refine flywheel to improve the system continually. We present our learnings from multiple deployments and discuss ongoing research aimed at weaving together more tightly the virtual and the physical realms. Bio: Venkat Padmanabhan is the Managing Director of Microsoft Research India in Bengaluru. He was previously with Microsoft Research Redmond, USA for nearly 9 years. Venkat’s research interests are broadly in networked and mobile computing systems, and his work over the years has led to highly cited papers and paper awards, technology transfers within Microsoft, and industry impact. He has received several awards and recognitions, including the Shanti Swarup Bhatnagar Prize in 2016, four test-of-time paper awards from ACM SIGMOBILE, ACM SIGMM, and ACM SenSys, and several best paper awards. He was also among those recognized with the ACM SIGCOMM Networking Systems Award 2020, for contributions to the ns family of network simulators. Venkat holds a B.Tech. from IIT Delhi (from where he received the Distinguished Alumnus award in 2018) and an M.S. and a Ph.D. from UC Berkeley, all in Computer Science, and has been elected a Fellow of the INAE, the IEEE, and the ACM. He is an adjunct professor at the Indian Institute of Science and was previously an affiliate faculty member at the University of Washington.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c1a5630c-9132-40cb-9e31-f840a1ab986f/7f07ce41-28d6-4d06-8ad7-efb90295f286-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pUULKS4bi9hG9uUy9prhF23220002024-06-28T17:13:11.687ZnetworkscommunicationsYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/2kgatqvAtb6XUAgF2UoKZzhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a9c0ba7e-dfe4-4944-af60-d4410450ef58.jpgAnnual Monitoring of Forest AGBTitle: Annual Monitoring of Forest AGB over a Period of 10 years Using SSL-derived Representations from Optical Time Series Abstract: I recap the functioning of our fully self-supervised learning pipeline based on the spectral-temporal Barlow Twins. The SSL approach generates highly informative representations at 10m spatial resolution from cloud-corrupted optical time series. The resulting representations are well correlated with GEDI-derived relative height measurements so that an AGB model for vegetation/forest of up to 300-500 t/ha can be derived. I show that the model transfers well between years making it possible to train the model on (for example) one year of Sentinel-2 data together with the corresponding GEDI measurements, and applying the frozen model to Landsat data acquired in previous years. Bio: 2010-2023: Full Professor for Remote Sensing/Geomatics - since 2016 Lead of Mantle's research team.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/0ac99d71-fc87-4d6f-a289-a74fb0539da7/66f4b45a-f1fe-43fd-b1ac-e3f0c26dc707-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/2kgatqvAtb6XUAgF2UoKZz3181032024-06-28T17:13:31.303ZforestssensingcoeYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/iSPamqxUdmP2CwNNdGyQSNhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b2187fda-c0c8-497f-b53e-83b56fe26187.jpgGrid-friendly Energy Community Coordination for Reducing Grid Reinforcement NeedsAbstract: This talk discusses how market mechanisms and automated trading strategies can be used to control the flexible consumption and generation units of the community members in such a way that they make the best possible use of existing distribution networks and support the network operator in avoiding and eliminating congestion situations. This ultimately helps avoiding grid reinforcements or allows to provide a better service with the existing grid, keeping in mind that it takes much longer to reinforce the grid than to build and connect many new (fluctuating) decentralized renewable generators and new loads such as heat pumps and electric vehicles. Bio: Since 2017: Professor of Control and Integration of Grids at INATECH; before: Professor for Energy Systems Technology and Energy Economics, in particular intelligent decentralized structures for sustainable power supply (Smart Grids) at Offenburg University of Applied Sciences, Fellow and head of the research project “Smart Grids” at the foundation neue verantwortung, Berlin, Senior Researcher and Project Manager in the research area “Future Energy Systems”, SAP AG, Research Assistant at the University of Mannheim, Research Fellow at the Iowa State University, Research Assistant at the University of Karlsruhe (TH)https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/90c39c20-5cbc-4bf9-b046-a3b72e6c19c2/ac586ae5-35bc-43f8-bb19-a7489f4dc55c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/iSPamqxUdmP2CwNNdGyQSN3450022024-07-05T17:14:38.368ZenergyeconomicsYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/tyPqbNvp3isgTDZVVoLFD1https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/bfc3817e-4fd7-43d0-a617-204427a3ac1b.jpgReal-time Wildlife MonitoringAbstract: This research introduces an AI-based alert system to reduce human-wildlife conflicts in the Romanian Carpathian Mountains. Globally, conflicts between people and wildlife are rising due to population growth, shifting land use patterns and climate change. In Romania, mountain communities are impacted by bears and wild boars, which damage livestock, crops and property. These conflicts can undermine conservation efforts and may result in the killing of problematic animals. In collaboration with Fundația Conservation Carpathia, this research supports Rapid Intervention Teams who respond to wildlife activity in mountain villages. Six years of camera trap data are used to train and test AI models to detect and classify European mammals. These models are integrated into an alert system and deployed in three locations. The new pipeline improves on the state-of-the-art for detecting and classifying bears and wild boars. Preliminary results from the field deployment show a positive impact on conservation efforts. This is the first known study to use remote processing of 4G-enabled camera trap images to operate a human-wildlife conflict alert system, with potential wider applications as cellular connectivity expands to more remote locations. Bio: Tom is an MRes student on the AI for Environmental Risk Centre for Doctoral Training at the University for Cambridge. He previously spent 10 years working for the UK's Foreign, Commonwealth and Development Office, where he designed and managed sustainable development projects while on postings in DRC, Sierra Leone and Tanzania.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/df3bc427-6fd0-468b-a20b-22039e89b1fe/87053087-bb50-453f-89dd-a8b5a8f8a5bf-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/tyPqbNvp3isgTDZVVoLFD12152032024-07-12T18:52:48.788ZbiodiversitysensingforestswildlifeconservationYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/pQBnfPWJi9kxLdeHY9YAA7https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b7db3ae6-e38b-4f16-8bbd-04e1ee8ff965.jpgPartner-driven Environmental Sensing:Full Title: Partner-driven Environmental Sensing: Co-design with Indigenous Ojibwe Scientists and Malagasy Conservationists Abstract: Evolving environmental sensing technologies present a myriad of opportunities for gathering data to understand and promote environmental justice, biodiversity, and climate change mitigation. However, technical development from academic and commercial settings often struggle to translate to accessible solutions for marginalized communities. In this talk, I will explore the opportunities of partner-driven co-design, share the findings from a qualitative study of field scientists’ use of technology, and present two case studies: (1) designing environmental sensors with Indigenous Ojibwe scientists for manoomin (wild rice) conservation and (2) partnering with Malagasy conservation organizations to understand the role that technology can play in reforestation and biodiversity monitoring. Bio: Eric Greenlee (he/him) is a PhD student in the College of Computing at Georgia Tech, co-advised by Ellen Zegura and Josiah Hester. Conducting research at the intersection of the Computing and Society Lab and the Ka Moamoa Lab, Eric explores partner-driven processes with communities often cut out of technology development to co-create emergent environmental sensors to address challenges in environmental justice, biodiversity loss, and climate change mitigation. By leveraging qualitative methods, he aims to strengthen connections across traditional silos to design and deploy user-friendly, networked, and low-power embedded systems. Prior to pursuing his PhD, Eric worked as a Radio Frequency engineer for the U.S. Federal Government and studied electrical engineering at Dartmouth College.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c10baf7d-454c-4a2e-b50f-fd91d1986382/7cb30f49-30b9-4066-8b1c-552daf227964-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pQBnfPWJi9kxLdeHY9YAA73115042024-10-04T18:15:38.900ZconservationsensingbiodiversitycoeYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/9CqWsuQQykVtbuDPwuLwZshttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/59843611-82b6-4d0d-901c-44becad91ca0.jpgVisual Digital Twins of ForestsAbstract: Methods developed by the computer graphics community allow for the photorealistic rendering of complex geometry. In this talk we explore how such mathematical procedures can be leveraged to describe the growth, biomechanics, and combustion of trees at a detailed spatial level. These models facilitate a realistic 3D visualization of these processes at forest scale which allows exploring illustratively a variety of hypothetical environmental scenarios. Potential applications of such methods include the educational dissemination of environmental concepts, the generation of synthetic image data for training vision-based AI models, and the evaluation of ecological hypotheses expressed at plant organ scale. Bio: Wojtek Palubicki is a Professor at Adam Mickiewicz University where he leads the Natural Phenomena Modelling Group. The research group uses methods from computer graphics and AI to describe and investigate natural pattern genesis. Before that he held a post-doctoral research scientist post at the SLCU developing mathematical models of plant developmental biology.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/45df410f-f99d-47f3-b930-957b167f9adc/e2ebc718-28e8-44c1-9840-12a5910b8af4-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/9CqWsuQQykVtbuDPwuLwZs3220012024-10-11T19:45:49.498ZforestssimulationcoeYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/7XijwZ8ZtKtnLgRXKgt9G1https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/26335bf6-1d55-4365-9ba7-f343555f18de.jpgChallenges in Cyber-Physical Energy Systems - of Agents and DataAbstract: In this talk, first, some special challenges in cyber-physical energy systems will be reflected. Then, examples from research projects and field tests will be discussed to show how multi-agent systems can be used to tackle these challenges. Finally, the topic of research data management and its role in open research will be discussed. Bio: Prof. Dr.-Ing. Astrid Nieße has been Professor for Digitalized Energy Systems at the University of Oldenburg since 2020 and a member of the Energy Division Board of the OFFIS - Institute of Computer Science. From 2018 to 2020 she was Professor for Energy Informatics at Leibniz University Hannover. Astrid Nieße received her doctorate from the University of Oldenburg in 2015; her doctoral thesis dealt with the application of distributed algorithms in the field of decentralized energy systems . Astrid Nieße studied computer science and biology at the University of Bremen and at the University of Oldenburg.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/384f69b1-8aa5-4a49-8ed4-1ab7df483778/cbf3fef2-e18a-4217-aea8-adf211f5f01d-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/7XijwZ8ZtKtnLgRXKgt9G13265002024-10-18T15:03:59.257ZenergyabmYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/pxkLZ4jgVJMqjwZuhWicrKhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/8695c05d-5776-485e-9111-7752d65c4a2e.jpgA CarbonFirst Approach for Decarbonizing Cloud ComputingAbstract: The exponential growth of cloud computing has been a defining trend of our time, fueled by rapidly growing demands from data-intensive and machine learning workloads. Despite the end of Dennard scaling, the cloud's energy demand grew more slowly than expected over the past decade due to the aggressive implementation of energy-efficiency optimizations. Unfortunately, there are few significant remaining optimization opportunities using traditional methods, and moving forward, the cloud's continued exponential growth will translate into rising energy demand, which, if left unchecked, will translate to increasing carbon emissions. In this talk, I will argue for a CarbonFirst approach to designing cloud computing systems by making carbon efficiency a first-class design metric, similar to traditional metrics of performance and reliability. I will explain how today's systems can be made first carbon-aware by exposing energy and carbon usage information to software platforms and then made carbon-efficient by providing control over the system's carbon usage. I will present an initial design of a system to enable such carbon awareness and management and present several application case studies on how modern cloud applications can employ these mechanisms to reduce their carbon footprint. I will end with open research challenges in the emerging field of computational decarbonization. Bio: Prashant Shenoy is currently a Distinguished Professor and Associate Dean in the College of Information and Computer Sciences at the University of Massachusetts Amherst. He received the B.Tech degree in Computer Science and Engineering from the Indian Institute of Technology, Bombay and the M.S and Ph.D degrees in Computer Science from the University of Texas, Austin. His research interests lie in distributed systems and networking, with a recent emphasis on cloud and sustainable computing. He has been the recipient of several best paper awards at leading conferences,...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/bea26ee5-370d-402e-820a-c05d2fa48de9/91eb9249-5cb3-44dc-88ef-618b94ed7a95-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pxkLZ4jgVJMqjwZuhWicrK3371012024-11-01T15:46:48.984ZenergycloudYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/pMzCFQKTrRtQ6jotF1z12Vhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/55059937-1455-43bb-84b6-e33fd9cca89c.jpgLong-term Biodiversity Monitoring at ScaleAbstract: Comprehensive data on global biodiversity patterns is only obtainable through in-situ distributed sensor networks. However, these multi-device networks are constrained by battery lifetimes, must gather rich data from power-hungry sensors, and yet must be deployed in remote environments for long periods. We look at the feasibility of a prototype multi-sensor device using on-device reinforcement learning for power management. Bio: Josh Millar is a PhD based at the NetSys Lab at Imperial-X. Their current research interests include: - energy-aware ML - IoT and on-device ML - applied ML for sustainabilityhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c09f62b7-a2f6-4875-8c45-a98d0f64b6b7/0cc13b9b-4044-4ade-aa6c-dd0760ffe9a8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pMzCFQKTrRtQ6jotF1z12V1355002024-11-15T16:16:36.544ZbiodiversitysensingconservationurbanYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/ijC1E36q7fn2qwxs7opSJqhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/3702883a-bfe5-477d-b4dc-63529d47dfda.jpgConservation EvidenceGrey literature’s inherent nature means that it is a difficult form of media to discover, typically being hidden deep within websites, analyse, following no standard file formats or structures, and process, due to the sheer volume of existing and actively produced literature, this forms a massive cost and time problem for organisations that require such literature in their function. We devise and implement a pipeline that uses Common Crawl internet archives to locate & scrape potential grey literature; then process it for use in a multistage machine learning pipeline to classify and output relevant media. Bios: Shrey Biswas is a second-year Computer Science Student at Pembroke College. Radhika Iyer is a second-year Computer Science Student at Murray Edwards College. Kacper Michalik is a Second-year Computer Science Student at Pembroke Collegehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8c44f016-07c0-4484-833f-b554679f175c/3dc4cf76-1b7b-4d68-8f08-7085cc8f077a-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ijC1E36q7fn2qwxs7opSJq3231042024-11-29T17:39:54.533ZconservationbiodiversityevidenceYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/uFyApvuvALLv66D7x36FErhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b63a5a7b-4279-438a-81ec-efc99732f940.jpgSmart Grid Trust AssessmentAbstract: Energy systems are highly complex. State determination and detection of anomalies, faults or even attacks are only possible to a limited extent with traditional approaches. This talk will investigate how such systems can be planned and operated in the future in the area of conflict between high automation and trust by human operators. Bio: Sebastian Lehnhoff is a Full Professor of Energy Informatics at the University of Oldenburg. He received his doctorate at the TU Dortmund University in 2009. Prof. Lehnhoff is chairman of the board of the OFFIS Institute for Information Technology and speaker of its Energy R&D division. He is a board member of the section „Energy Informatics“ within the German Informatics Society (GI) as well as an active member of numerous committees and working groups focusing on ICT in future Smart Grids. In 2022 he was appointed to the Board of Trustees of the Volkswagen Foundation (VolkswagenStiftung). He is the CTO of openKONSEQUENZ e.G. – a registered cooperative industry association for the development of modular Open-Source SCADA/EMS. He serves as Chairman of the Executive Board of the Energy Research Centre of Lower Saxony (EFZN) as well as an Executive Committee Member of the ACM Special Interest Group on Energy Systems and Informatics (SIGEnergy). Prof. Lehnhoff is a member of the German Academy of Science and Engineering (acatech) as well as a member of the Berlin-Brandenburg Academy of Sciences and Humanities (BBAW).https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/e845f9cc-b5cb-4e91-b97e-ac0705a78979/8e0163d3-2a22-4661-8f95-3c35234bfb54-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/uFyApvuvALLv66D7x36FEr3387002024-12-06T16:40:43.895ZenergysmartgridYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/feDup1JutmgQkC6ipGF9r5https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/03d54d07-6179-4a5b-ba2d-9c69e29ffd7b.jpgOptimising Sustainable Energy with Functional ProgrammingAbstract: This talk describes some results from a collaboration between Computer Science, Physics, and Climate Impact Research on theories and tools for performance optimisation of strongly coupled physical systems with a large parameter space. The first part of the talk discusses computing optimal policies; we have used these techniques for climate decisions and for fusion energy designs. The second part of the talk will focus on one particularly important concept: the Pareto-front, which mathematically captures the trade-offs between two (or more) conflicting objectives. The core object of study is an expensive black-box function computing multiple objectives, for which we approximate the Pareto front using adaptive mesh refinement. Bio: Patrik Jansson is a professor in the Computer Science and Engineering Department, joint between Chalmers University of Technology and the University of Gothenburg, Sweden. His main research areas are Programming Languages, Functional Programming, Domain-Specific Languages, and their application to climate, physics, etc. His research focus is on systems for constructing correct and reusable software. The goal is to develop the programming languages of the future and theories, tests and proofs of the correctness of high-level models of complex systems. Important techniques include functional programming, domain-specific languages and type theory. Examples of applications are climate impact research, physics, and language technology but many results are also curiosity-driven basic research with generic applicability in most areas. Patrik has been on sabbatical in Oxford, as a Visiting Fellow of Kellogg College for Michaelmas term 2024, visiting Prof Jeremy Gibbons.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/7347b6cb-172c-4988-a5ef-6c59398e2f86/96573d23-b78d-4a3a-8f7f-15499df2c0a5-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/feDup1JutmgQkC6ipGF9r53825022024-12-13T18:23:12.951ZfpenergyfusionYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/8PhivRm85jZuFg8v55yo7Fhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4c397ffc-9800-4077-bda2-e92dd042a793.jpgUsing Low-cost, Research-led, Decentralised NetworksFull Title: Using Low-cost, Research-led, Decentralised Networks to Increase Access to High Quality Microspatial Data on Building Stocks, and the Built and Natural Infrastructure Abstract: The Colouring Cities Research Programme (CCRP) is overseen by an informal, international academic consortium that uses its decentralized research-led network, to co-create and manage, permanent open data/visualisation platforms across countries. These provide standardised, open microspatial data on the characteristics, performance, and short/long-term dynamics of building stocks, and built and natural infrastructure. They also test feedback loops between live streaming, computational inference, and crowdsourcing approaches to improve coverage and reliability of data, and to support cross sector/multidisciplinary engagement. Research Institutions from 30 countries are currently involved. The programme has been set up to accelerate progress towards UN SDGs; to provide access to big data required to exploit the potential of AI & ML and gain insights at scale; to reduce research costs and overlaps and speed up testing of research applications by pooling expertise, funding and ideas, and; to ensure that areas such as data standardisation, uncertainty in data, and citizen privacy and security are prioritised, as commercial demand for microspatial data grows. Bio: Polly Hudson is a Senior Research Fellow at The Alan Turing Institute and PI for the Colouring Cities Research Programme. She was previously a Senior Research Fellow at the Centre for Advanced Spatial Analysis University College and held a Visiting Fellowship at the Kellogg Centre for the Historic Environment, University of Oxford. Relevant appointments include advisory/board positions for the Department of Culture, Media and Sport, English Heritage, The Royal Institute of British Architects, and the National Lottery (charitable arm). Polly trained as an architectural historian, and initially worked in furniture making, ...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3f49f969-1e5b-480a-9525-45fcb1e12e1b/6410c0a2-1770-4cc4-95f7-3b41a12c1f2e-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/8PhivRm85jZuFg8v55yo7F3144032025-02-14T16:25:58.351ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/9hADtA5Fov2vdDt9iNVjJQhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a9fe2f3c-5817-495e-97aa-292e1ab1b348.jpgRobotics and Sensing for Sustainable Crop ProductionAbstract: Crop farming is essential in our society, providing food, feed, fiber, and fuel. We heavily rely on crop production, but at the same time, we need to reduce the production footprint. We aim to address this key challenge by investigating new solutions to produce crops more sustainably. We study novel technology-driven approaches to move toward sustainable crop production. Agricultural robots offer promising directions to address management challenges in agricultural fields or support plant breeding efforts through large-scale trait acquisition. For that, field robots need the ability to perceive and model their environment, predict possible future developments, and make appropriate decisions in complex and changing situations. This talk will showcase our recent developments in robotics for crop production, incorporating machine learning to support farmers in operating more sustainably and reducing some negative impacts on the ecosystem. Bio: Cyrill Stachniss is a full professor at the University of Bonn and heads the Photogrammetry and Robotics Lab. He is also a Visiting Professor in Engineering at the University of Oxford and is with the Lamarr Institute for Machine Learning and Artificial Intelligence. Before his appointment in Bonn, he was with the University of Freiburg and ETH Zurich. Since 2010, he has been a Microsoft Research Faculty Fellow and received the IEEE RAS Early Career Award in 2013. From 2015 to 2019, he was senior editor for the IEEE Robotics and Automation Letters. He is the spokesperson of the DFG Cluster of Excellence "PhenoRob" at the University of Bonn, together with his colleague Heiner Kuhlmann. His research focuses on probabilistic techniques as well as learning approaches for mobile robotics, perception, and navigation. The main application areas of his research are autonomous service robots, agricultural robotics, and self-driving cars. He has co-authored over 300 publications and has coordinated multiple large-scale resear...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/431a5c6c-f0d5-4fed-a846-b0c2aef19f84/45aafb41-d764-485f-b7be-8e2d9ad57765-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/9hADtA5Fov2vdDt9iNVjJQ3033002025-02-21T16:26:06.356ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/dFShkouits1FFyUctiSSH5https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/d18c98fb-8346-443f-b33d-8ea967483bf5.jpgAn Introductory Guide on How to Create Beautiful SlidesFrank Feng is a first-year Ph.D. student in the Department of Computer Science and Technology at the University of Cambridge. His research interests lie at the intersection of machine learning and earth sciences, with a particular focus on the application of self-supervised learning in remote sensing.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/66bea879-6796-4b4b-bff9-446823a854d6/027fc202-0f56-4c5f-82ca-b5b6155cd068-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/dFShkouits1FFyUctiSSH53147032025-02-28T17:59:20.609ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/oW6eqJBH1Hkwu6wE7XzQT3https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c1cbe483-ee25-42f9-9bea-6f492d2244c3.jpgIdentifying Key Countries in the Illegal Elephant Ivory Trade NetworAbstract: Illegal wildlife trade is a key driver of biodiversity loss, but targeting policy to maximise disruption to trade remains a key challenge. A network approach was applied to seizure data to prioritise national action disrupting the illegal trade of elephant ivory. By simulating the removal of countries from trade, targeting groups of countries was found to be most effective due to network redundancy. Despite temporal variability, trade was highly concentrated and cessation in less than 10 countries would have disrupted 75% of trade in 2018-2020. These findings support evidence-based legislation and efficient allocation of conservation resources for tackling illegal wildlife trade. Bio: Jakob is a PhD student in the Conservation and Development Lab (Department of Geography). His research focusses on evaluating policy for sustainable land systems, supervised by Prof. Rachael Garrett and Prof. Srinivasan Keshav. This work is supported by the Centre for Doctoral Training on Artificial Intelligence applied to the study of Environmental Risk (AI4ER CDT). Before starting his PhD, Jakob completed an MRes with AI4ER in Environmental Data Science, where he collaborated with TRAFFIC to develop data-driven tools to inform international illegal wildlife trade policy. Previously, Jakob completed an undergraduate degree in Natural Sciences at the University of Cambridge, specialising in Plant Sciences, and contributed to research on metrics for biodiversity offsetting, novel approaches to wildlife monitoring and forest ecology.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b9b67028-a982-4794-8a9b-fe026db5d728/44081e58-77e8-43b2-8336-9412c979e638-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/oW6eqJBH1Hkwu6wE7XzQT31553522025-03-14T17:34:06.978ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/3exAV8tLbnPSGqoKv2mZtshttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5a1d7cf6-171a-401f-a2b1-2bcccde81915.jpgGlobal Maps of Human Threats to Biodiversity and Species ExtinctionsFull title: Towards Global Maps of Anthropogenic Threats to Biodiversity and Their Contributions to Species Extinctions Abstract: Species extinctions are primarily driven by loss of habitat, which is relatively easy to monitor by satellite remote sensing; other anthropogenic threats to biodiversity, like hunting, are much more difficult to observe directly. My PhD project draws on local studies which capture the population effect of some anthropogenic threat, scaling these results using machine learning and remote sensing. In this talk, I will discuss my first attempt at this through quantifying species-specific responses to hunting pressure. I find that machine learning methods can offer marked improvements over (linear) statistical models, which are commonly used in ecology, but model validation must be done carefully to properly contextualise predictive performance. I will preview my plans for integrating these hunting pressure models with the LIFE biodiversity metric framework to express pressure in terms of extinction risk. If there is time, I will also discuss future plans for my PhD. Bio: Emilio is a PhD student in the Department of Zoology at the University of Cambridge in the Conservation Science Group and the Energy and Environment Group. He is supervised by Andrew Balmford, with co-supervision from Anil Madhavapeddy and Tom Swinfield. He is also part of the AI for Environmental Risks Centre for Doctoral Training, a researcher at the Cambridge Centre for Carbon Credits, and a member of Churchill College. His research focuses on the uses of predictive modeling for biodiversity conservation, with an emphasis on quantifying species-specific responses to human disturbance.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/12166bf8-1c6a-4233-94d9-e5dc367fea4c/d7f39bab-db28-4b35-b6ea-bae3e1eca714-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/3exAV8tLbnPSGqoKv2mZts2568012025-03-21T16:11:39.039ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/f1Uxw34FRLEfVNBBpzbsgDhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c58ec009-4e3c-4f50-86d0-7bfcd6578aac.jpgBlockchain-Based Carbon Trading for Stakeholders in BrazilFull Title: Democratizing Carbon Markets: A Blockchain-Based Emission Trading System for Small and Large-Scale Stakeholders in Brazil Abstract: The integration of blockchain technology into carbon markets offers a unique opportunity to create more transparent, inclusive, and efficient trading mechanisms. This presentation introduces a novel Blockchain Emission Trading System (BETS) model designed to align with Brazil’s new carbon market legislation (Law 15042/2024), ensuring that both large landholders and small rural producers can participate fairly. Our approach leverages official land registries, such as SICAR, to create spatially and temporally verifiable carbon credits, preventing fraud and double counting while enabling greater accessibility for smaller stakeholders who often struggle to enter regulated markets. By decentralizing the issuance and trading of carbon credits, our model aims to reduce intermediaries, lower costs, and promote broader participation, ultimately fostering a more equitable environmental and economic transition. Through a systematic mapping study, we identify key challenges and research directions for blockchain-based carbon markets and propose a framework that ensures compliance with national and international standards while prioritizing social and economic inclusivity. Bio: Jean is a professor at the Federal University of Santa Catarina (UFSC) in Brazil, specializing in information security, blockchain technology, and electronic documents. He holds a PhD in Computer Science from the University of Cambridge, where his research focused on cryptographic protocols and secure execution of code. Over the years, he has worked extensively on the development of blockchain-based solutions, particularly in the areas of digital identity, electronic signatures, and regulatory compliance. His recent work explores the use of blockchain to improve transparency, security, and inclusivity in digital ecosystems, including its application in carbon...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/71805280-20ba-443a-8b81-215fffbe8173/93683ea6-f068-4a10-9e10-833997842938-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/f1Uxw34FRLEfVNBBpzbsgD3461032025-03-28T17:09:11.873ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/7aqBd2Nn9E6QpMvnoBPxuQhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ae6196a1-0295-477f-8f97-7e773b882435.jpgTowards Global-scale Species Distribution ModellingAbstract: Estimating the geographical range of a species from sparse observations is a challenging and important geospatial prediction problem. Given a set of locations where a species has been observed, the goal is to build a model to predict whether the species is present or absent at any location. This problem has a long history in ecology, but traditional methods struggle to take advantage of emerging large-scale crowdsourced datasets which can include tens of millions of observations of hundreds of thousands of species in addition to the availability of multi-modal data sources such as paired images and natural language descriptions. In this talk, I will present recent work from my group where we have developed deep learning-based solutions for estimating species' ranges from sparse presence-only data. I will also discuss some of the open challenges that exist in this space.   Bio: Oisin Mac Aodha is a Reader in Machine Learning in the School of Informatics at the University of Edinburgh. He is also an ELLIS Scholar and former Turing Fellow. He obtained his PhD from University College London and was a postdoc at Caltech prior to his current role. His current research interests are in the areas of self-supervised learning, 3D vision, fine-grained learning, and human-in-the-loop learning. In addition, he works on questions related to AI for conservation and biodiversity monitoring. More information can be found on his website: https://homepages.inf.ed.ac.uk/omacaodhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/31e7b0c5-1cdc-4de9-b965-a3662d0caeec/e93c2225-26eb-43bd-98dc-eeb2cd4110bb-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/7aqBd2Nn9E6QpMvnoBPxuQ3606042025-04-11T17:19:28.339ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/pBZXQsbjuXLtLfyTWx4LWnhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/faa6f06b-f5e3-4546-970b-c0e561b3349c.jpgLOCO-2024 Lightning TalksLightning talks, 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. List of talks: + Assessing the ecological impact of AI: From Ethics of Technology to Engineering—and back. Sylvia Wenmackers. KU Leuven, Belgium + Running a DIY assessment against the Web Sustainability Guidelines. James Smith. + RSS Podcast Feed Inefficiency: trimming climate impact. Damon Hart-Davis. University of Surrey, UK. + Juice Sucking Servers. Axel Roest Phluxus. the Netherlands. + Bridging Models and Reality: Real-World-Oriented Methodologies for Estimating ICT Carbon Footprint at Imperial College London. Yurong Yu, Niel Hanham, Harriet Wallace, Jasmin Cooper, and Mark Sinclair. Imperial College London, UK. + Advancing data center sustainability: carbon-aware computing utilizing AI automation. Imran Latif, Mohtadi Mahim, and Marwan Ruby Brookhaven. National Laboratory + Stony Brook University + SUNY Farmingdale State College, USA. + Energy-Efficient Computing Using Alternative Architectures (ARM and RISC-V). Emanuele Simili. University of Glasgow, UK. + Simulating Carbon Opportunity Costs at DESY. Dwayne Spiteri. DESY, Germany. + Grid site operational techniques to reduce carbon emissions. Gordon Stewart, David Britton, Emanuele Simili, Sam Skipsey, and Bruno Borbely. University of Glasgow, UK. + Using ILNP to Do More for Less. Gregor Haywood. University of St Andrews, UK. + Smart Procurements. Dan Protopopescu, University of Glasgow, UK. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/bf48f120-56ef-457a-b906-8acc0aacd269/23273532-c858-4a7d-8806-77522ab61d66-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pBZXQsbjuXLtLfyTWx4LWn3632012025-04-16T23:16:24.530ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/3JbRfyyN2wH4811yaNMAamhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/04149e24-4b5c-4909-ade3-2dd77646b817.jpgLOCO-2024. Teaching Sustainable Creative Technologies. Chelsea Thompto. Virginia Tech, USATeaching Sustainable Creative Technologies. Chelsea Thompto. Virginia Tech, USA. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/161633a1-f561-4ab3-adcc-6423a0a1276e/4c4a5f11-7587-474e-b66e-1330df7b3542-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/3JbRfyyN2wH4811yaNMAam656002025-04-16T23:16:34.731ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/cXKBBwBaML8vpjD2pcNhJchttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4d13005b-b4d4-4613-9497-90fdd87158b7.jpgLOCO-2024. The Dark Side of Dark Mode. Zachary Datson, BBC, UKThe Dark Side of Dark Mode. Zachary Datson. BBC, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/60dd4f89-ea5f-4e80-973b-8447d32b6ecf/69430b94-c3c1-4d73-ab56-698bca9cd22c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/cXKBBwBaML8vpjD2pcNhJc474032025-04-16T23:16:37.958ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/tuKQAjhTTmLi2irBMKpCAwhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/7b45761d-22b2-421e-b651-df6c7d305f36.jpgLOCO-2024. On Comb. Two Server Control Policies for Energy Eff. J Dai et al., McMaster Uni., CanadaOn Combining Two Server Control Policies for Energy Efficiency. Jingze Dai and Douglas Down. McMaster University, Canada. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/deaa95db-3093-47da-83f6-e6552a4a898a/da3f8b6a-fe6b-4aba-970c-28aade4fa8d4-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/tuKQAjhTTmLi2irBMKpCAw760002025-04-16T23:16:40.023ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/sSdg6PKCYV4ZcpKuU8ZKrChttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5968f105-65ea-4979-b1fa-94a223c2a3f0.jpgLOCO-2024. Choosing the Right Battery Model for Data Center Sims. Paul Kilian et al., KTH, SwedenChoosing the Right Battery Model for Data Center Simulations. Paul Kilian, Philipp Wiesner, and Odej Kao. KTH Royal Institute of Technology, Sweden + Technische Universität Berlin, Germany. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/d9905970-4ef7-49bb-89d6-db71c7d1e132/e026f03c-8125-42f0-bbdf-6c102f70dd55-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/sSdg6PKCYV4ZcpKuU8ZKrC601022025-04-16T23:16:45.496ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/7TjHCjY8ijMpjRwVdqKJ22https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/1abf5563-e282-4707-8622-95d9a8a92c57.jpgLOCO-2024. A Digital Twinning Approach to Decarbonisation... B. Archibald et al., Univ. of GlasgowA Digital Twinning Approach to Decarbonisation: Research Challenges. Blair Archibald, Paul Harvey, and Michele Sevegnani. University of Glasgow, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/37c14ca6-c1df-4d91-90af-6d8da8f6791b/2e2cc962-8485-4de5-a2e8-c3dfc2b6a8fd-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/7TjHCjY8ijMpjRwVdqKJ22655002025-04-16T23:16:48.810ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/pwD8wfj9tjETp83LkU9WuFhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/90e016f1-8860-42bb-8358-99bd244324b5.jpgLOCO-2024. Coop. Sensor Networks for Long-Term Biodiv. Monitoring. J Millar et al., Imperial CollegeCooperative Sensor Networks for Long-Term Biodiversity Monitoring. Josh Millar, Sarab Sethi, Hamed Haddadi, Michael Dales, and Anil Madhavapeddy. Imperial College London + University of Cambridge, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/be89625e-c671-4e2c-8261-a98b1361a077/eb9e3b7c-909b-492c-a0c1-3f47210db06d-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pwD8wfj9tjETp83LkU9WuF603032025-04-16T23:16:53.312ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/9hTnxoqSoq5YreoFfhMB8dhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/0ae3c328-a144-446b-b93d-c3dcde67db51.jpgLOCO-2024. Emission Impossible: privacy-preserving carbon emiss. claims. J. Man et al. Cambridge, UK(No sound in the first two minutes) Emission Impossible: privacy-preserving carbon emissions claims. Jessica Man, Sadiq Jaffer, Patrick Ferris, Martin Kleppmann, and Anil Madhavapeddy. University of Cambridge, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4324ab18-f3b2-4fdd-883f-a4188dee5816/ed3edafa-45d6-40d1-b502-b1f8d1951e2a-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/9hTnxoqSoq5YreoFfhMB8d919092025-04-16T23:16:58.077ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/jxT3JgDufjqH7hbFGDVfUXhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/652134e1-2f97-41e4-8ae7-386260a729ab.jpgLOCO-2024. Incr. Awareness for Energy Cons. in Jupyter Notebooks. M. Garus et al., Potsdam, GermanyIncreasing Awareness for Energy Consumption in Jupyter Notebooks. Marcel Garus, Sven Köhler, and Andreas Polze. Hasso Plattner Institute, University of Potsdam, Germany. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/963802f8-ef36-4e89-b994-6639d848bf4f/0fb8599c-fb37-4c3e-a094-d9a93071eb9a-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/jxT3JgDufjqH7hbFGDVfUX652002025-04-16T23:17:01.083ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/6eS8k9SmVyXrb3bkwK15CYhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/dc9c169b-3590-4bd6-bd3f-d35c48c3795b.jpgLOCO-2024. Green Metrics Tool: Measuring for fun & profit. G.-D. Hoffmann et al., GCS, Germany.Green Metrics Tool: Measuring for fun and profit. Geerd-Dietger Hoffmann and Verena Majuntke. Green Coding Solutions + HTW Berlin, Germany. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/2a6d40f8-74f9-4fd6-b433-4b70e56fd920/3f0c9863-3c4a-4269-aae2-929b1a4723f2-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/6eS8k9SmVyXrb3bkwK15CY614012025-04-16T23:17:04.894ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/oofSsxe38AiRbx2Hu6T6qwhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a009e492-583c-4602-a685-5d90e3f005e6.jpgLOCO-2024. Ichnos: A Carbon Footprint Est. for Sci. Workflows. K. West et al., Univ. of Glasgow, UKIchnos: A Carbon Footprint Estimator for Scientific Workflows. Kathleen West, Yehia Elkhatib, and Lauritz Thamsen. University of Glasgow, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b54493d0-48b6-4bed-86d4-c4eb68d3f5ca/c0f9e5c3-5651-495f-9502-1e1b0f16fbb1-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/oofSsxe38AiRbx2Hu6T6qw1040002025-04-16T23:17:09.307ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/2JauY9jo1rwmWH3L5wpRGkhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/21c965a3-fcd4-4d5d-a118-fe34e811fb88.jpgLOCO-2024. Putting green software principals into practice. James Uther, Oliver Wyman, UKPutting green software principals into practice. James Uther, Oliver Wyman, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/0dfc3b32-bbac-451a-9765-3b5b4a6821ff/c6850ac6-d223-4224-bbc1-d6abfcb6e7ee-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/2JauY9jo1rwmWH3L5wpRGk820012025-04-16T23:17:12.208ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/943rW9UpVVYRPYHfG48CCAhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5ee12c89-ff84-45d4-b5a0-4a84dfb15ac3.jpgLOCO-2024. The belief in Moore’s Law is undermining ICT climate action. A. Friday et al., Lanc. Uni.The belief in Moore’s Law is undermining ICT climate action. Adrian Friday, Christina Bremer, Oliver Bates, Christian Remy, Srinjoy Mitra, and Jan Tobias Mühlberg. Lancaster University + University of Edinburgh, UK + Université Libre de Bruxelles, Belgium. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4135d943-ac52-45bd-80ae-ef65658d78a2/d1786f93-1115-449e-9fa8-a89419e16521-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/943rW9UpVVYRPYHfG48CCA752002025-04-16T23:17:16.228ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/6NuDERX3vTaqfDdyMZZ9i6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/995496b0-390b-4710-b57f-700b84efaf15.jpgLOCO-2024. Exploring Privacy & Security as Drivers for Env. Sust... J. Kayembe et al. ULB, BelgiumExploring Privacy and Security as Drivers for Environmental Sustainability in Cloud-Based Office Solutions. Jason Kayembe, Iness Ben Guirat, and Jan Tobias Mühlberg. Université Libre de Bruxelles, Belgium. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/2efb9005-fa97-4da6-a24a-67dba46be497/0e559353-9758-432a-b5cc-1d904394279e-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/6NuDERX3vTaqfDdyMZZ9i6851022025-04-16T23:17:18.740ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/r5VcsyFUaZRp5awyuWjBdchttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/09462f7d-0a53-430d-9536-7745100bde6e.jpgLOCO-2024. Lineage first comp.: towards a frugal userspace for Linux. M Dales et al., Cambridge, UKLineage first computing: towards a frugal userspace for Linux. Michael Dales, Patrick Ferris, and Anil Madhavapeddy. University of Cambridge, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/cb2439c9-d160-4daa-8103-b952c5aa2c5f/671dad57-da5d-4482-b766-31dd20531f37-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/r5VcsyFUaZRp5awyuWjBdc8585332025-04-16T23:17:24.567ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/f43a7YVK4w1ZkwXmRWThBXhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/50ceb95e-7fd1-4bfd-a7d0-f38c684d8581.jpgLOCO-2024. Carbon-Aware Microservice Deployment for Opt. UX on a Budget. K. Kreutz et al., TU BerlinCarbon-Aware Microservice Deployment for Optimal User Experience on a Budget. Kevin Kreutz, Philipp Wiesner, and Monica Vitali. Technical University of Berlin, Germany + Politecnico di Milano, Italy. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/71cc8083-2394-4230-8a32-163b0d8383dd/49c20c9f-e002-4da0-a8cb-e09e80ee70c4-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/f43a7YVK4w1ZkwXmRWThBX550002025-04-16T23:17:26.786ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/5yC2UmcNZa4hjU58akF7z3https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a9f6b275-6718-4975-b921-712d68b6738d.jpgLOCO-2024. Environmentally-Conscious Cloud orchestration... G. Attenni et al., Sapienza, ItalyEnvironmentally-Conscious Cloud orchestration considering Geo-Distributed Data Centers. Giulio Attenni and Novella Bartolini. Università di Roma “La Sapienza”, Italy. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/24f28fb0-4883-46fa-b4ad-6ac29fec12fc/fe981ab6-a6a3-4da4-b54a-f636736c6260-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/5yC2UmcNZa4hjU58akF7z3538002025-04-16T23:17:30.297ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/voNk95u6aQVRsByHbzbkNMhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/748250a2-73b4-415c-8f72-354ca97a7c49.jpgLOCO-2024. MAIZX: A Carbon-Aware Framework for Optimal Cloud Comp. Emissions. F. Ruilova et al., KTHMAIZX: A Carbon-Aware Framework for Optimizing Cloud Computing Emissions. Federico Ruilova, Ernst Grunnar Gran, and Sven-Arne Reinemo. KTH Royal Institute of Technology, Sweden + Norwegian University of Science and Technology, Norway. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ee07ef9a-57b2-4fac-9c83-0f4bb81b9465/51fb98c5-b256-4a20-b91c-2b5797b9a293-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/voNk95u6aQVRsByHbzbkNM863012025-04-16T23:17:34.848ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/aukfwbzvrxoUacYdmcAft3https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/885c58da-6e95-4e8b-9280-5bbe291571bc.jpgLOCO-2024. Carbon-Aware Name Resolution. Ryan Gibb et al. University of Cambridge, UKCarbon-Aware Name Resolution. Ryan Gibb, Patrick Ferris, and Anil Madhavapeddy. University of Cambridge, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/4cd6efdb-fd22-4a1c-a326-df49dfc1f398/16038226-eb66-43de-b4ab-dea24471e834-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/aukfwbzvrxoUacYdmcAft3832022025-04-16T23:17:37.845ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/haaCcKVrb3bPdYNFXiLuNyhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5a1579f5-b21a-4686-9ecc-5ee7861d3883.jpgLOCO-2024. Rebound GHG Effects in AgriTech. Matthew Broadbent & Oliver Bates. SRUC; Lancaster Uni.Rebound GHG Effects in AgriTech. Matthew Broadbent and Oliver Bates. Scotland’s Rural College (SRUC) + Lancaster University, UK. 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/82d9d7f6-74c1-4160-aaec-36b4bd06af8c/00b296c4-b581-4ebc-8710-7c3da62d2280-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/haaCcKVrb3bPdYNFXiLuNy658002025-04-16T23:17:40.393ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/szVBBfJvdPY6vQ2ynPcC3bhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4578e242-8ccd-419c-b115-2484ea6bffa2.jpgLOCO-2024. Keynote talk by Ayse Coskun. Boston University, USAKeynote talk by Ayse Coskun. Boston University, USA. (Session chair: Philipp Wiesner) 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/d74a30b9-4293-4e7d-a8a2-aea0ada3ddd6/dc6bdbcf-7db6-4b43-8b4d-d792c1e97bed-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/szVBBfJvdPY6vQ2ynPcC3b1787002025-04-16T23:17:49.243ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/iq7srZXfB9Y4cfsscyf7Vohttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/cf6e742b-1baa-4d6c-92eb-7bfca9e40b3d.jpgLOCO-2024. Keynote discussion with Anne Currie. Strategically Green, UKKeynote discussion with Anne Currie. Strategically Green, UK. (Moderator: Andres La Riva Perez) 1st International Workshop on Low Carbon Computing. A hybrid event hosted in Glasgow, Scotland, UK, 3 December 2024. https://sicsa.ac.uk/loco/loco2024/https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8d092fb4-e49a-4d6c-9d37-2169330b4480/3ee7eac7-4608-4298-91df-78897a22c831-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/iq7srZXfB9Y4cfsscyf7Vo1583042025-04-16T23:17:55.545ZYESInternational Workshop on Low Carbon ComputingNOhttps://watch.eeg.cl.cam.ac.uk/w/gohsjWasx7SGdbCyiDhMyRhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/2820f030-01d4-4aef-9a45-598b55781f49.jpgIs there Hope for the ClimateSrinivasan Keshav is the Robert Sansom Professor of Computer Science at the University of Cambridge, focusing on the intersection of computer science and sustainability. He earned his PhD from UC Berkeley and has held roles at Bell Labs, Cornell University, and the University of Waterloo. A Fellow of the Royal Society of Canada, ACM, and IEEE, Keshav is recognized for his contributions to networking and sustainability. His research includes innovations in energy systems, carbon footprint reduction, and forest conservation using remote sensing. Keshav emphasizes practical applications of computer science to global challenges, fostering collaborative solutions in smart grids and biodiversity conservation.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/7c95953a-274e-42be-bbf0-fd3e94e58035/d6bb8fbd-bbcf-411e-ba14-fe70e0195fe3-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/gohsjWasx7SGdbCyiDhMyR3652062025-05-01T18:00:36.174ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/dwMbyPnsrcBXtTrUKGGVishttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/666e735f-ba18-4032-b86a-5c7710f82683.jpgModelling Building Thermal Dynamics – From Data Generation to Transfer LearningAbstract: Building operations contribute approximately one-third of global CO₂ emissions. Advanced control strategies can reduce these emissions by up to 30%. Such control requires accurate mathematical models that capture the building’s thermal dynamics. Data-driven modeling has emerged as the most scalable approach for this purpose. However, the availability of high-quality building data remains limited. To address this challenge, we propose two methods: (1) a data generation framework that synthesizes realistic building operation data, and (2) a general Transfer Learning model that serves as an effective initialization for modeling new target buildings. Bio: Fabian is a second-year PhD student in the Department of Energy Management Technologies at the Technical University of Munich, supervised by Prof. Dr. Christoph Goebel. His research focuses on using Machine Learning to model building thermal dynamics. Such models are necessary for enabling Model Predictive Control of the building, which can reduce CO₂ emissions by up to 30%.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/6579d24f-314c-4af9-b998-4b7f94813f88/c1b7d8f3-1562-4aeb-b84b-ea5dfa2c7175-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/dwMbyPnsrcBXtTrUKGGVis2473022025-05-15T16:22:19.447ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/j2WWKaVRTKRwMWn4xCzoxKhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/41272bb3-8819-4eaf-92c4-047fa0ee0389.jpgAn Introduction to Self-supervised LearningAbstract: I will present a short tutorial on some approaches to self-supervised learning (SSL), assuming no background in machine learning. If time permits, I will present examples of the use of SSL for problems in energy systems. Bio: Srinivasan Keshav is the Robert Sansom Professor of Computer Science at the University of Cambridge, focusing on the intersection of computer science and sustainability. He earned his PhD from UC Berkeley and has held roles at Bell Labs, Cornell University, and the University of Waterloo. A Fellow of the Royal Society of Canada, ACM, and IEEE, Keshav is recognized for his contributions to networking and sustainability. His research includes innovations in energy systems, carbon footprint reduction, and forest conservation using remote sensing. Keshav emphasizes practical applications of computer science to global challenges, fostering collaborative solutions in smart grids and biodiversity conservation.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/920a199a-58e0-42e2-a1b7-58dc95074351/6c657b2d-31eb-4b8c-9289-3b454b3a01e6-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/j2WWKaVRTKRwMWn4xCzoxK3449072025-06-12T15:01:08.564ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/k2Nkty1Rap5TYoMwyxS8dShttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/bdb32ba5-b5de-4103-bc61-5f3e3dc83374.jpgReal-time Wildlife MonitoringAbstract: This research introduces an AI-based alert system to reduce human-wildlife conflicts in the Romanian Carpathian Mountains. Globally, conflicts between people and wildlife are rising due to population growth, shifting land use patterns and climate change. In Romania, mountain communities are impacted by bears and wild boars, which damage livestock, crops and property. These conflicts can undermine conservation efforts and may result in the killing of problematic animals. In collaboration with Fundația Conservation Carpathia, this research supports Rapid Intervention Teams who respond to wildlife activity in mountain villages. Six years of camera trap data are used to train and test AI models to detect and classify European mammals. These models are integrated into an alert system and deployed in three locations. The new pipeline improves on the state-of-the-art for detecting and classifying bears and wild boars. Preliminary results from the field deployment show a positive impact on conservation efforts. This is the first known study to use remote processing of 4G-enabled camera trap images to operate a human-wildlife conflict alert system, with potential wider applications as cellular connectivity expands to more remote locations. Bio: Tom is an MRes student on the AI for Environmental Risk Centre for Doctoral Training at the University for Cambridge. He previously spent 10 years working for the UK's Foreign, Commonwealth and Development Office, where he designed and managed sustainable development projects while on postings in DRC, Sierra Leone and Tanzania.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/9a1def0a-04d3-4fa8-860c-33cf50b779a6/aba4c2f6-b1b7-40ea-9c58-18a692149ee4-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/k2Nkty1Rap5TYoMwyxS8dS2272052025-11-06T13:12:29.505ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/vb5T4XGfdbzTWERAcBozh6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ffbeb4ed-2fe7-41aa-ae08-e07acfc1b884.jpgCOP30: How Cambridge is helping to tackle the climate crisisCambridge is combining world-leading research, innovation and education to equip the next generation to lead with knowledge and vision. With COP30 highlighting the urgency of climate and biodiversity action, Cambridge's mission is to enable engaged citizens to think critically, navigate mis- and disinformation and act with clarity and purpose. Find out more about how we're doing this: https://www.cam.ac.uk/stories/education-climate-justicehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ec41776f-a03b-42f7-beee-75a9cf21f8c9/10effed5-1021-4bf2-8ce5-dc28e30fa4b7-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/vb5T4XGfdbzTWERAcBozh684032025-11-08T14:13:12.350ZCambridge UniversityCambridge researchYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/da6KXrX3T4NkhSzH3ekJighttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/aae6b575-35ca-4510-8270-2d152bcb8d47.jpgFrom Perception to Prediction: Modelling Pastoral (Im)Mobility in the West SahelAbstract: Modelling climate-induced mobility is one of the most pressing challenges in the Central Sahel and neighbouring regions, where environmental stressors such as droughts, extreme temperatures, and erratic rainfall are increasingly displacing vulnerable communities. This is especially true for pastoralists, a type of herders that move across landscapes with their livestock in search of fresh forage and water, while allowing the land to recover. Changing climatic circumstances disrupt the availability of these resources, diminishing crop yields and forcing pastoralists to adjust their traditional routes in search of suitable grazing areas. Although climate models provide reasonably reliable projections of future climate conditions such temperature and precipitation patterns, they do not directly reveal how such environmental shifts shape human decisions. Similarly, many migration models exist, but few capture the combined environmental, socio-cultural and behavioural mechanisms that shape pastoral mobility. Bio: Aline is a 2nd year PhD student co-supervised by Emily So and Emily Shuckburgh in the Departments of Architecture and Computer Science at the University of Cambridge. Her research sits at the intersection of data science, urban planning and climate adaptation, exploring how remote-sensing and data-drive approaches can support responses to humanitarian crises.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/6272e800-8bdb-46de-9384-97bb691a8a99/99f22141-a0af-4858-ad0d-ad4a94f13b99-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/da6KXrX3T4NkhSzH3ekJig1988002025-11-07T17:21:19.939ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/9YS2ocTb41FcLdtz11twkqhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/7ca1f17d-7c01-48dd-9ca4-5740cab714cc.jpgTESSERA Temporal Embeddings of Surface Spectra for Earth Representation and AnalysisAbstract: Optical satellite Earth-observation (EO) time series are often obscured by clouds, resulting in sparse and temporally irregular observations. Compositing addresses these issues, but is insensitive to changes in vegetation phenology, which is critical for downstream tasks. Instead, we present TESSERA, a pixel-wise foundation model for multi-modal (Sentinel-1/2) EO time series that learns robust, label-efficient embeddings. During model training, TESSERA uses Barlow Twins to enforce invariance to the choice of cloud-free observations randomly sampled from the time series, so that the generated embeddings interpolate missing observations. We employ two key regularizers: global shuffling to decorrelate spatial neighborhoods, and mix-based regulation to improve invariance under extreme sparsity. We find that for diverse classification, segmentation, and regression tasks, TESSERA embeddings deliver state-of-the-art accuracy with high label efficiency, often requiring only a tiny task head and minimal computation. To democratize access, adhere to FAIR principles, and to simplify use, we release global, annual, 10m, pixel-wise int8 embeddings together with open weights/code and lightweight adaptation heads, thus providing practical tooling for large-scale retrieval and inference at planetary scale. Bio: Frank Feng is a second-year Ph.D. student in the Department of Computer Science and Technology at the University of Cambridge. His research interests lie at the intersection of machine learning and earth sciences, with a particular focus on developing self-supervised learning methods in remote sensing.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/48b996b4-40f9-4f22-8c39-fa281c1c6296/26cb5f17-fddf-4f2c-87cc-198fc7a13762-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/9YS2ocTb41FcLdtz11twkq3494062025-11-14T17:41:52.081ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/hanDLh2nzjXY4yL1akVEYhhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/56ea97f4-2ea6-431d-a3b8-707cd96a2c57.jpgTESSERA (Workshop on Foundational AI to forecast ecosystem resilience)Profs Srinivasan Keshav and David Coomes present TESSERA at the Workshop on Foundational AI to forecast ecosystem resilience at Pembroke College. 24th Nov 2025. https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/82e14165-298e-48e1-9cc4-b9035f9c5930/aa31cdb0-22b5-4ae3-824a-f7ac1e76a1b1-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/hanDLh2nzjXY4yL1akVEYh211452042025-11-24T15:24:41.040ZYESFoundational AI to Forecast Ecosystem ResilienceNOhttps://watch.eeg.cl.cam.ac.uk/w/tZVGowQRtNY9FsBCubzSMPhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/eb200f85-08bd-4a74-840d-a40dbafcb44c.jpgConcordia: Generative Agent Based ModelingSasha Vezhnevets from Google DeepMind presents Concordia at the Workshop on Foundational AI to forecast ecosystem resilience at Pembroke College. 24th Nov 2025. https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/e2bd3af2-84e9-4066-9963-7175572ffb31/01598b27-8001-4534-8316-3859454efce2-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/tZVGowQRtNY9FsBCubzSMP20070612025-11-24T22:38:24.096ZYESFoundational AI to Forecast Ecosystem ResilienceNOhttps://watch.eeg.cl.cam.ac.uk/w/vwUDfmeLPVoB3AfK4mium6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/eba079de-4c5f-49dc-9c05-86b3746066bd.jpgForecasting ecosystem resilience: what this means and why we need itJulia P. Jones discusses issues around causal forecasting at the Workshop on Foundational AI to forecast ecosystem resilience at Pembroke College. 24th Nov 2025. https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ef29c6b1-800d-4272-9b9f-fb4626c65905/825bc6f4-2b05-421c-a315-64dfb60ae394-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/vwUDfmeLPVoB3AfK4mium622720772025-11-24T22:38:46.286ZYESFoundational AI to Forecast Ecosystem ResilienceNOhttps://watch.eeg.cl.cam.ac.uk/w/hdWwEQ6NGxyHfVQU8t9iUnhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/7bb035d3-dade-481a-8695-7daec078805d.jpgDr Becks (Rebecca) Spake: Virtual EcologyDr Becks Spake discusses virtual ecology at the Workshop on Foundational AI to forecast ecosystem resilience at Pembroke College. 24th Nov 2025.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8360bf94-521f-4f6d-a337-5a232bdd6691/3c143de1-e070-45af-b1c0-888db19a7ee5-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/hdWwEQ6NGxyHfVQU8t9iUn12940892025-11-29T08:09:55.807ZYESFoundational AI to Forecast Ecosystem ResilienceNOhttps://watch.eeg.cl.cam.ac.uk/w/k5eyLUNQemCQ1gyoRJm8Wrhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/86ec3364-1174-4ec0-8a8b-fdc47f75e4dc.jpgSam Reynolds: Conservation and AIDr Sam Reynolds discusses his horizon scan on the impact of AI on conservation at the Workshop on Foundational AI to forecast ecosystem resilience at Pembroke College and the Cambridge Conservation Initiative. 27th Nov 2025. You can read his [paper here](https://linkinghub.elsevier.com/retrieve/pii/S0169534724002866)https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/9a74f82a-33cd-403a-94f0-76b797376431/f17fd4ce-a78f-49d5-826a-4acecd9b393d-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/k5eyLUNQemCQ1gyoRJm8Wr16770292025-12-01T17:41:47.608ZYESFoundational AI to Forecast Ecosystem ResilienceNOhttps://watch.eeg.cl.cam.ac.uk/w/gVh9Po8U6PhVcFdNM619bbhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5af8f36f-06fb-4934-92eb-f43a1594987e.jpg16. Diogo Veríssimo - Oxford UniversityBehaviour change without evidence: we ignore people at our perilhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/80e97428-85e1-4381-bd53-e29acc28e05e/95b1f3c4-dab7-4019-a084-394b4e737aaf-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/gVh9Po8U6PhVcFdNM619bb798002026-01-22T04:08:48.543ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/mU72fwT8QrNsG8Vds5EhSNhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/75bc032f-4f4e-44cf-a476-118cd043ed24.jpg15. Mark Burgman - University of HawaiʻiEvidence and reasoning in conservation decisionshttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/a93d2900-2b1a-4e6f-b896-726e061a5cb2/73efccd3-180d-441b-8a80-f3573f920a97-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/mU72fwT8QrNsG8Vds5EhSN982002026-01-22T04:08:59.472ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/ncyLeVLFL6hYd7XXoiAk5Whttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/33e697be-e89a-414b-9042-cfbe98f9f9d4.jpg14. David Righton - CefasEvidence to support nature recovery in the marine environmenthttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/abad48ba-ad37-4181-9870-b9b6d432952a/9da9cf8c-0215-497f-9632-911746683b06-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ncyLeVLFL6hYd7XXoiAk5W968002026-01-22T04:09:07.071ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/tihwvUx1aHKs4Mzucijk83https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/182225b9-4b05-4b95-a658-03046e8e3b73.jpg13. Sallie Bailey - Natural EnglandCreating evidence-based practicehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/dd10928f-5c21-4b9e-8d3d-de5595b95284/e6b44f64-c911-4f94-8302-d25a3485fe8e-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/tihwvUx1aHKs4Mzucijk831013002026-01-22T04:09:10.657ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/i75xA11FuTfU9gb8tKAad6https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/6a63062e-9353-4959-8f75-fe88916e60e2.jpg12. Jake Fiennes - Holkham EstateEmbedding monitoring and learning into conservation practichttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8a849edb-e4ca-45ab-beec-d46374a94fa1/b74a2337-f879-4f8a-8f5d-6e9a37387f74-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/i75xA11FuTfU9gb8tKAad6946002026-01-22T04:09:14.364ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/eEcsGx9WN1LvvyQPSC4Y71https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/93a27b95-075d-4011-8b32-17ff7c4314c4.jpg11. E.J. Milner-Gulland - Oxford UniversityEvidence for business actionhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/6e9c20e6-79c4-4b6c-b2d8-6e982c7f42f4/9776257b-4ed8-4613-91a0-514d14065eed-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/eEcsGx9WN1LvvyQPSC4Y71954002026-01-22T04:09:20.685ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/neHk5D7FVQDDxmh2W6RnNyhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/bb5c4253-49d9-4fcd-a895-abc7b4b2f99d.jpg10. Jeremy Wilson - RSPBHow RSPB builds on evidencehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/abfa0cfd-eefb-43fe-aa1e-369cb0583578/673aacdd-5629-449b-a109-b46bccb6aee7-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/neHk5D7FVQDDxmh2W6RnNy923002026-01-22T04:09:23.967ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/8u77echCyzEmxToRFfN6pthttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/1cf7f568-d4bc-4c3b-9756-a6d0df52b5d1.jpg9. Craig Bennett - The Wildlife TrustsThe Wildlife Trust’s Evidence Emergency programmehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3c9c8f5a-e63a-4596-9f30-5e50a0007eb5/63db5632-6931-456e-8044-a8a9a69b7da6-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/8u77echCyzEmxToRFfN6pt918002026-01-22T04:09:26.991ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/gizYNxhJfJpxvUeAkfSkabhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/588fcbb4-0170-4460-87a2-86deb3d059f4.jpg8. Sally Hayns - CIEEMChanging the culture of evidence use amongst environmental practitionershttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/7bedaa1d-49f2-45cb-821d-99c4f71ee050/69a032fd-a1e2-4823-af14-4b0b781661ca-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/gizYNxhJfJpxvUeAkfSkab853002026-01-22T04:09:29.411ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/r6iTXZTWxcJKQZsBtFexGbhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c12ffed5-477d-4d16-a37b-e38ac9917102.jpg7. Nancy Ockendon - Endangered Landscape & Seascape ProgrammeLessons from embedding evidence in the ELSPhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/cb3236d0-e35c-4fee-8b14-4003d6abca8e/e56eb805-857f-4b36-a974-b1b75cdf65fc-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/r6iTXZTWxcJKQZsBtFexGb915002026-01-22T04:09:40.831ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/cheQBkBfSsZwGZP6F3jtaahttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5c7fbdd7-3c42-42d7-8347-7607523956c8.jpg6. Anil Madhavapeddy - University of CambridgeHow AI might revolutionise conservation practicehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/5b58f41f-8d71-48ba-92d1-b34a49ee76ef/29278e03-d9d3-4e4d-b52a-d7a24e477bfa-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/cheQBkBfSsZwGZP6F3jtaa9580182026-01-22T04:09:47.340ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/5gKMas7rqfegBZgSTsunmMhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5ef43ba6-7871-43f2-85a0-0820fa0cf129.jpg5. Julia Jones - Bangor UniversityBiodiversity conservation’s causal revolutionhttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/2297b0d5-4d5b-4125-94b5-9db0d1d67889/c90830ed-3027-4e21-b77f-5e4315671a8c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/5gKMas7rqfegBZgSTsunmM812002026-01-22T04:09:51.993ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/oUQHgVF7Xg31PHdCiqfALohttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/9e5e9d07-e124-4c79-839f-8bec99607606.jpg4. Tiago de Zoeten - Mossy EarthLearning from funding tests of conservation actionshttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b989bf04-ea29-4dd5-8979-502d27c16e26/d18d7319-03ba-4a7c-a68b-5a101a7434a5-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/oUQHgVF7Xg31PHdCiqfALo743002026-01-22T04:09:56.287ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/sb766Jo5EzYE1dcgJ7vggnhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/3501c19b-78b7-4075-bd5a-d9c49ee40246.jpg3. Bill Sutherland - University of CambridgeEvidence-based conservation: progress and challengeshttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/d3f6cb6f-59c9-4691-bf0f-33c49a7efeff/7a2bb4cd-e46b-4586-a631-b2644245229c-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/sb766Jo5EzYE1dcgJ7vggn881042026-01-22T04:09:58.961ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/taKeGNMSfQk9C2wLeRxrKBhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/151c2627-3641-4b33-bb41-648ddd52c7e7.jpg2. Caroline Fiennes - Giving EvidenceHow to be more effectivehttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/dc031542-8e2d-446d-8173-6d2fd4a8a56d/57123286-1b58-435e-b906-a5a886f176dc-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/taKeGNMSfQk9C2wLeRxrKB949002026-01-22T04:10:01.586ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/wNY1enRFrG3Cy4A1Q2LX3Khttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/a6f7852e-6945-4567-9976-539a41fdd5de.jpg1. Anjali Goswami - DefraMaking effective environmental policieshttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f980e30b-a57a-4274-8e67-0cce3350154b/f6577108-d927-4178-96b9-720855db3957-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/wNY1enRFrG3Cy4A1Q2LX3K1038002026-01-22T04:10:07.999ZYESDelivering Effective Conservation EvidenceNOhttps://watch.eeg.cl.cam.ac.uk/w/qYKfndaWrJrfSiSRFEQu5Mhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/f4aa0a26-0a77-411e-9807-337bdb99b9c7.jpgCalibrated Probabilistic Interpolation for GEDI BiomassAbstract: Mapping global forest biomass from NASA's GEDI mission requires interpolating sparse LiDAR observations across diverse landscapes. Standard machine learning approaches like Random Forest and XGBoost fail to produce calibrated uncertainty estimates, as they conflate ensemble variance with true predictive uncertainty and ignore spatial context. We introduce Attentive Neural Processes (ANPs), a probabilistic meta-learning framework that conditions predictions on local observations and geospatial foundation model embeddings. ANPs learn flexible spatial covariance functions, expanding uncertainty in complex landscapes and contracting it in homogeneous areas. Validated across five biomes from tropical Amazonian to boreal and alpine ecosystems, ANPs achieve competitive accuracy with near-ideal uncertainty calibration. The framework also enables few-shot adaptation, recovering most cross-region transfer performance with minimal local data. This provides a scalable, principled alternative to ensemble methods for continental-scale biomass mapping. Bio: Robin Young is a first-year PhD student in Computer Science at the University of Cambridge.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/ca47a19c-5c71-4129-8cce-1fabca6f7885/67bd8666-4acf-4801-8f40-15b8cc8039a3-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/qYKfndaWrJrfSiSRFEQu5M3049002026-01-23T21:17:49.033ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/j9JMLJJkiKSA58F9kAv4kchttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/fa7c3388-7b0b-43ce-8625-c00205ae6740.jpg[PROPL'25] Bridging Disciplinary Gaps in Climate Research Through Programming Accessibility and(…)Bridging Disciplinary Gaps in Climate Research through Programming Accessibility and Interdisciplinary Collaboration (Video, PROPL 2025) Cristian Urlea, Ana Denisa Urlea, Wim Vanderbauwhede, Adriana Laura Voinea, and Syed Waqar Nabi (University of Glasgow, UK; Romanian Air Traffic Services Administration, Romania; University of Glasgow, UK; University of Glasgow, UK; University of Glasgow, UK) Abstract: Climate change research relies on complex computational tools to model environmental processes, analyse large datasets, and inform policy. Current scientific computing practices pose major barriers to entry, particularly for interdisciplinary researchers and those in low and middle-income countries (LMICs). Challenges include steep learning curves, limited access to expert support, and difficulties with legacy or under-documented software. Drawing on real-world experiences, we identify recurring obstacles in the usability, accessibility, and sustainability of scientific software. Our analysis highlights that current approaches to scientific software disadvantage interdisciplinary and LMIC researchers. We propose specific mechanisms to address these inequities: improved documentation, domain-aware training, automation for diverse hardware environments, domain-specific languages and hybrid support communities. These measures should be integrated into grant funding requirements to ensure sustainability beyond initial project periods, transforming scientific software from short-lived outputs into accessible research infrastructure. By reimagining scientific programming as a shared public good, we can lower barriers to entry and foster a more inclusive, equitable climate research ecosystem. Article: https://doi.org/10.1145/3759536.3763804 ORCID: https://orcid.org/0000-0001-7851-8916, https://orcid.org/0000-0002-0716-9904, https://orcid.org/0000-0001-6768-0037, https://orcid.org/0000-0003-4482-205X, https://orcid.org/0000-0003-3835-4851 Video Tags: climate change, s...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/92fcd0d8-1338-4346-95eb-b9fbce089bcd/86cc3d53-a318-4926-9f4c-0e5ff0025ec8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/j9JMLJJkiKSA58F9kAv4kc428002026-01-28T09:49:43.272ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/4Y7jQWtqm89jJ4dfEdb8aEhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/096d62ca-19ea-4d9e-9578-311a79aef2a2.jpg[PROPL'25] Towards Modelling and Verification of Coupler Behaviour in Climate ModelsTowards Modelling and Verification of Coupler Behaviour in Climate Models (Video, PROPL 2025) Chinmayi Baramashetru and Dominic Orchard (University of Kent, UK; University of Cambridge, UK / University of Kent, UK) Abstract: Climate models and earth system models often comprise submodels composed via a 'coupler', a software component that enables interaction between submodel components. The continuous exchange of data through couplers creates the risk of subtle errors propagating across components, potentially distorting scientific conclusions. In this paper, we argue for lightweight formal verification techniques applied at the coupler interface to improve both coupler and model correctness. By enforcing formal contracts on data exchanges, the coupler can act as a checkpoint that detects and prevents certain classes of component-level errors before they propagate between models. We abstract general design principles for couplers and propose verifiable subsystems. Using an example of a real-world bug, we illustrate a hybrid verification strategy that integrates module-level contracts, verified through both static and runtime techniques. We aim to offer a practical pathway for both existing and future couplers, ultimately enabling auditable and formally verifiable couplers. Article: https://doi.org/10.1145/3759536.3763801 ORCID: https://orcid.org/0000-0001-5344-0032, https://orcid.org/0000-0002-7058-7842 Video Tags: Climate Models, Formal Methods, Couplers, Verification, splashws25proplmain-p9-p, doi:10.1145/3759536.3763801, orcid:0000-0001-5344-0032, orcid:0000-0002-7058-7842 Presentation at the PROPL 2025 conference, October 12–18, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN, ACM SIGAda,https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/2020f707-3307-4ec4-bad1-9bb60bda6abc/41c7cb9d-d2bf-47fb-806f-c2dc16b4424a-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4Y7jQWtqm89jJ4dfEdb8aE1298002026-01-28T09:49:53.448ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/wHULgdkG72tpyK9fpscszxhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4ae7be30-b3e2-431e-a59c-f24602db7e4b.jpg[PROPL'25] Welcome to the 2nd PROPLWelcome to the 2nd PROPL (Video, PROPL 2025) Anil Madhavapeddy, KC Sivaramakrishnan, and Dominic Orchard (University of Cambridge, UK; IIT Madras and Tarides; University of Cambridge; University of Kent) Abstract: Welcome to the 2nd ACM SIGPLAN International Workshop on Programming for the Planet (PROPL 25), colocated with ICFP/SPLASH 2025 and held in Singapore on October 13th 2025! This workshop is dedicated to bridging the use of principled computer science towards positive climate and biodiversity actions. This year’s edition follows a successful 1st workshop held at POPL 2024 in London, which had over 100 participants attend at the Institute of Engineering and Technology in London, UK. The chairs will open the workshop, report on the provocations received, and introduce the first session. Presentation at the PROPL 2025 workshop, Oct 13, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f8cc2bba-52c4-4622-b964-b3d8b8567c79/7e239dff-601a-45a9-a2e4-2e492f93d4b8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/wHULgdkG72tpyK9fpscszx369002026-01-28T09:49:56.233ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/oBywYf4wd1gpFGXBZDtn4Vhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/0e27777c-0cd9-4967-a577-451a7115fb42.jpg[PROPL'25] Yirgacheffe: a declarative approach to geospatial dataYirgacheffe: A Declarative Approach to Geospatial Data (Video, PROPL 2025) Michael Winston Dales, Alison Eyres, Patrick Ferris, Francesca A. Ridley, Simon Tarr, and Anil Madhavapeddy (University of Cambridge, UK; University of Cambridge, UK; University of Cambridge, UK; Newcastle University, UK; IUCN, UK; University of Cambridge, UK) Abstract: We present Yirgacheffe, a declarative geospatial library that allows spatial algorithms to be implemented concisely, supports parallel execution, and avoids common errors by automatically handling data (large geospatial rasters) and resources (cores, memory, GPUs). Our primary user domain comprises ecologists, where a typical problem involves cleaning messy occurrence data, overlaying it over tiled rasters, combining layers, and deriving actionable insights from the results. We describe the successes of this approach towards driving key pipelines related to global biodiversity and describe the capability gaps that remain, hoping to motivate more research into geospatial domain-specific languages. Article: https://doi.org/10.1145/3759536.3763806 ORCID: https://orcid.org/0009-0003-0832-4114, https://orcid.org/0000-0001-7866-7559, https://orcid.org/0000-0002-0778-8828, https://orcid.org/0000-0001-6068-7519, https://orcid.org/0000-0001-8464-1240, https://orcid.org/0000-0001-8954-2428 Video Tags: Declarative, Geospatial, Python, Biodiversity, splashws25proplmain-p63-p, doi:10.1145/3759536.3763806, orcid:0009-0003-0832-4114, orcid:0000-0001-7866-7559, orcid:0000-0002-0778-8828, orcid:0000-0001-6068-7519, orcid:0000-0001-8464-1240, orcid:0000-0001-8954-2428 Presentation at the PROPL 2025 conference, October 12–18, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN, ACM SIGAda,https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b72021da-29fb-44f2-9ec0-2ebb9fc8993f/86669be8-f7d2-41d6-b74d-329dd73791b9-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/oBywYf4wd1gpFGXBZDtn4V768062026-01-28T09:49:59.378ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/f9LyYoTx8iWJAmn1ziXtUYhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c076bd10-ac5f-42e1-b576-4c72007b9b91.jpg[PROPL'25] GPU-accelerated Hydrology Algorithms for On-prem Computation: Flow accumulation,(…)GPU-Accelerated Hydrology Algorithms for On-Prem Computation: Flow Accumulation, Drainage Lines, Watershed Delineation, Runoff Simulation (Video, PROPL 2025) Rahul Kumar, Vatsal Jingar, Abhilash Jindal, and Aaditeshwar Seth (IIT Delhi, India; IIT Delhi, India; IIT Delhi, India; IIT Delhi, India) Abstract: Critical hydrology related algorithms to trace the path of surface water flows, including flow accumulation, stream order, watershed delineation, and runoff simulation, can be difficult to execute for large aerial extents at fine spatial and temporal resolutions. Libraries like GDAL that use multi-threaded CPU-based implementations running on a single host may be slow, and distributed infrastructures like Google Earth Engine may not support the kind of computational primitives required by these algorithms. We have developed a GPU-accelerated framework that re-engineers these four algorithms and is able to process areas as large as river basins of 250,000 km2 on commodity GPU workstations. We express these algorithms in terms of easily parallelizable primitives of pixel independent (PI) and short-pixel (SP) operations, and iterative primitives of long-pixel (LP) operations. Each algorithm uses a different mix of the primitives which helps us ensure that the implementation is generic. We show that our implementation of these algorithms produces accurate outputs and is able to achieve significant performance benefits over alternative methods. Being able to execute the algorithms on a commodity GPU workstation paves the path to use on-prem infrastructure to produce national-scale outputs, and collaborate to pool multiple national-scale outputs together for global-scale analysis. Article: https://doi.org/10.1145/3759536.3763805 ORCID: https://orcid.org/0009-0006-0206-5774, https://orcid.org/0009-0001-9395-8511, https://orcid.org/0000-0002-4525-9791, https://orcid.org/0000-0001-9012-5656 Video Tags: GPU, hydrology, parallel computing, flow accumulation, watershed ...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/72995b92-a9d7-42d6-b92e-51a7c0076194/c2b20390-bcfe-4402-a00a-2542e5280e08-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/f9LyYoTx8iWJAmn1ziXtUY1231002026-01-28T08:19:21.567ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/trq77gFwc8ZbUeGqyDpzN7https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ea555e9e-6e24-40d3-bdb8-1ef32c76aa56.jpg[PROPL'25] Challenges in Practice: Building a Usable Library for Planetary-Scale EmbeddingsChallenges in Practice: Building a Usable Library for Planetary-Scale Embeddings (Video, PROPL 2025) Sadiq Jaffer, Frank Feng, Robin Young, Srinivasan Keshav, and Anil Madhavapeddy (University of Cambridge; University of Cambridge; University of Cambridge; University of Cambridge; University of Cambridge, UK) Abstract: Remote sensing observations from satellites are critical for scientists to understand how our world is changing in the face of climate change, biodiversity loss, and desertification. However, working directly with this data is difficult. For any given satellite constellation, there are a multitude of processed products, data volume is considerable, and for optical imagery, users must contend with data sparsity due to cloud cover. This complexity creates a significant barrier for domain experts who are not specialists. Pre-trained, self-supervised foundation models such as TESSERA (https://arxiv.org/abs/2506.20380) aim to solve this by offering pre-computed global embeddings. These rich embeddings can be used in-place of raw remote sensing data in a powerful “embedding-as-data” approach. For example, a single 128-dimensional TESSERA embedding for a 10-meter point on Earth can substitute for an entire year of optical and radar imagery, representing its temporal and spectral characteristics. While this could democratise access to advanced remote sensing-derived analytics, it also creates a new programming challenge: a lack of tools designed for this new approach. In this talk we will focus on our lessons learnt from the development of geotessera (https://github.com/ucam-eo/geotessera), a library designed for this new embeddings-as-data approach. We will explore key design decisions that focus on both a high-level API for accessibility and tight integration with the existing scientific Python ecosystem. The core user workflow will be demonstrated, showing how our library enables a rapid classification task on this new data paradigm. By presenting this...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/de333188-f771-4531-b9ef-12148222bbfe/d770c7af-5e33-4b1b-a270-c2e6040a34b8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/trq77gFwc8ZbUeGqyDpzN7895012026-01-28T08:19:31.259ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/4AMXNsM1bNd3gJcmj3fYB1https://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4bfeeb07-0ec7-410f-9c8f-63ba380cf6ee.jpg[PROPL'25] Authoring Tools for Transparent Climate ReportingAuthoring Tools for Transparent Climate Reporting (Video, PROPL 2025) Roly Perera, Joe Bond, Cristina David, Andrew McNutt, and Alfonso Piscitelli (University of Cambridge/University of Bristol; University of Bristol, UK; University of Bristol; University of Utah; University of Salerno) Abstract: Energy transition and decarbonisation, adaptation to climate change, risk mitigation strategies and other components of a sustainable future all require changes in public policy and behaviour. These in turn require transparent, evidence-based communication of the core issues to policymakers, other scientists, and the general public. This talk will highlight the role of software infrastructure in meeting these transparency requirements and will report on a “transparent programming languages” project called “Fluid”. Presentation at the PROPL 2025 workshop, Oct 13, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1d270780-612b-4c7f-bdae-93fc31505dde/8927a0f3-1366-4378-8fa9-5de99e7d59e9-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4AMXNsM1bNd3gJcmj3fYB11233002026-01-28T06:56:55.251ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/fEoa7jde33i35w1Xz816fthttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b0b72d94-d121-41de-a858-c39a2905c1ea.jpg[PROPL'25] STACD: STAC Extension with DAGs for Geospatial Data and Algorithm ManagementSTACD: STAC Extension with DAGs for Geospatial Data and Algorithm Management (Video, PROPL 2025) Saharsh Laud, Saurabh Joshi, Tarun Mangla, Abhilash Jindal, and Aaditeshwar Seth (IIT Delhi, India; IIT Delhi, India; IIT Delhi, India; IIT Delhi, India; IIT Delhi, India) Abstract: Geospatial datasets have complex lineages that are crucial for reproducibility and understanding data provenance, yet current metadata standards like STAC (SpatioTemporal Asset Catalog) provide limited support for capturing complete processing workflows. We propose STACD (STAC extension with DAGs), an extension to STAC specifications that incorporates Directed Acyclic Graph (DAG) representations along with defining algorithms and version changes in the workflows. We also provide a reference implementation on Apache Airflow to demonstrate STACD capabilities such as selective recomputation when some datasets or algorithms in a DAG are updated, complete lineage construction for a dataset, and opportunities for improved collaboration and distributed processing that arise with this standard. Article: https://doi.org/10.1145/3759536.3763803 ORCID: https://orcid.org/0009-0008-4549-0556, https://orcid.org/0009-0008-2912-023X, https://orcid.org/0000-0001-9016-9931, https://orcid.org/0000-0002-4525-9791, https://orcid.org/0000-0001-9012-5656 Video Tags: Geospatial Workflows, STAC, Data Provenance, Workflow Management, Reproducibility, Metadata Standards, splashws25proplmain-p42-p, doi:10.1145/3759536.3763803, orcid:0009-0008-4549-0556, orcid:0009-0008-2912-023X, orcid:0000-0001-9016-9931, orcid:0000-0002-4525-9791, orcid:0000-0001-9012-5656 Presentation at the PROPL 2025 conference, October 12–18, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN, ACM SIGAda,https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/76bbdbf8-e65d-4967-bcb8-d6db678577cb/d4b2d918-044b-43c6-8282-ec457509d615-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/fEoa7jde33i35w1Xz816ft813002026-01-28T06:57:04.297ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/r8agKUxpRpXzgYR79EL2Qbhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/9c3ef08a-bb56-4d2a-8ce2-f2bace399fe8.jpg[PROPL'25] Spatial Programming for Environmental MonitoringSpatial Programming for Environmental Monitoring (Video, PROPL 2025) Josh Millar, Ryan Gibb, Roy Ang, Hamed Haddadi, and Anil Madhavapeddy (Imperial College London; University of Cambridge; University of Cambridge; Imperial College London; University of Cambridge, UK) Abstract: Large-scale environmental monitoring demands real-time, spatially-aware coordination across distributed networks. However, existing distributed computing models poorly capture spatial structure, hindering dynamic collaboration and fine-grained access control. We argue that space must be treated as a first-class concept in programming models for these systems based on bigraphs – a formalism that explicitly models spatial arrangements, data movement, and access policies, while supporting real-time reconfiguration and localised reasoning. This approach facilitates secure, composable, and dynamically verifiable coordination across geographically distributed nodes and organisations, paving the way for scalable, responsive environmental networks. Presentation at the PROPL 2025 workshop, Oct 13, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/cb746302-c258-4d82-b9df-cc2ae1cd664e/96f2a193-4d6f-448e-9dfe-b5895cc91497-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/r8agKUxpRpXzgYR79EL2Qb966512026-01-28T06:57:10.341ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/pMcmaRRuESvP41RrtdCAKdhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/7d7f57ce-e037-40b4-b5f4-7a2a4e263a59.jpg[PROPL'25] Scaling the Urban Forest: An Integrated Framework for Managing Cities by Fusing Raster(…)Scaling the Urban Forest: An Integrated Framework for Managing Cities by Fusing Raster and Vector Data (Video, PROPL 2025) Andrés C. Zúñiga-González, Anil Madhavapeddy, and Ronita Bardhan (University of Cambridge; University of Cambridge, UK; University of Cambridge) Abstract: Urban trees are a cornerstone of 21st-century cities, serving as the most immediate source of nature in our highly urbanised world. Their benefits are extensive; they improve public health, offer calming green spaces, and contribute to the socio-cultural fabric of city life. Ecologically, urban forests are vital in the fight against climate change, mitigating the urban heat island effect, reducing air pollution, and acting as powerful carbon sinks that can harbour more carbon per hectare than some tropical forests. These green assets also host diverse wildlife and provide essential recreational spaces for citizens. However, traditional tree mapping efforts are often costly, labour-intensive, and slow. Furthermore, manual surveys typically only capture trees in accessible areas like parks and roadsides, leading to a significant underestimation of the total urban canopy. In this short demonstration, we showcase a reproducible framework to move from disparate geospatial datasets to a cohesive, building-level description of green infrastructure at a national scale. Our approach leverages several state-of-the-art tools to overcome the limitations of traditional methods. We aimed to quantify the 3-30-300 rule for urban greening—a “rule of thumb” policy stating that every citizen should see 3 trees from their home, live in a neighbourhood with 30% canopy cover, and be within 300 metres of a public park. Presentation at the PROPL 2025 workshop, Oct 13, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/c091a6f2-4775-462b-a171-e0b2e5f58b92/32936723-ad30-4501-883b-171c1b1ddd9d-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/pMcmaRRuESvP41RrtdCAKd639002026-01-28T06:57:16.887ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/aYXqXLtgQawYMjVXjSQtjxhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/7c6ac547-689b-4f2a-8235-513fdd15d3d3.jpg[PROPL'25] What we talk about when we talk about scientific programmingWhat we talk about when we talk about scientific programming (Video, PROPL 2025) Patrick Ferris (University of Cambridge, UK) Abstract: Programming for the planet undoubtedly involves programming scientifically, but what kind of programming are we talking about and what makes it scientific? In what ways does it differ from other forms of programming, if at all? Is scientific programming, data science or machine learning fundamentally different to constructing a compiler or building a high-throughput web server? By considering how the scientific method (with its falsifiable hypotheses and repeatable and reproducible experiments) relates to scientific programming, I hope to explore how computer science and traditional programming techniques are coming up short in meeting the requirements of scientific programmers. Presentation at the PROPL 2025 workshop, Oct 13, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/50d60f2a-521f-4a45-990c-1f93caa1891f/43d88c89-2681-4fe0-bfca-54eb2ce42fc9-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/aYXqXLtgQawYMjVXjSQtjx1046552026-01-28T06:57:21.567ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/hHN9CnfsTZZTvRZNKUgGpohttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/7f45626b-ac24-4c94-a822-0423aa4e24a9.jpg[PROPL'25] Precision Action Towards Climate and Health (PATCH)Precision Action Towards Climate and Health (PATCH) (Video, PROPL 2025) Dr. Angela Chaudhuri, Nitish Kumar Venkatesan, Prerakkumar Mukeshkumar Shah, and Sabhimanvi Dua (Swasti; Catalyst Management Services Pvt. Ltd.; Catalyst Management Services Pvt. Ltd.; Swasti) Abstract: Climate change presents multifaceted public health challenges, from heat-related mortality and vector-borne disease expansion to water contamination and respiratory ailments. The 2022 Lancet Countdown Report demonstrates a host of health effects of climate change ranging from heat-related illness and mortality to the spread of vector-borne and water-borne pathogens, to rising food insecurity as cropping patterns change. Current public health systems lack integrated, real-time data capabilities to identify vulnerable populations and coordinate timely responses to these climate-induced health threats, particularly in resource-constrained settings. Precision Action Towards Climate and Health (PATCH) addresses this gap through a comprehensive digital platform that integrates multiple surveillance streams like environmental surveillance (air quality, weather patterns, water quality), participatory surveillance (community-reported health indicators), and media surveillance (disease outbreak signals) with human health, animal health, and climate data to create actionable insights for government stakeholders. The platform transforms raw data into information and ultimately into insights that inform multiple stakeholders: government departments for policy decisions, community organizations and NGOs for targeted interventions, private sector for risk mitigation actions, and the general public through media communication. The system provides three core functionalities: - a customizable state-level health intelligence platform for real-time climate-health risk assessments - GIS-based vulnerability mapping for targeted interventions - tailored risk communication strategies across stakeholder groups. Imple...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/8768273e-d123-4bf6-95b6-e163d834e164/63abd41e-a6f4-4abc-9c11-0b8b9ece0241-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/hHN9CnfsTZZTvRZNKUgGpo1100012026-01-28T06:57:25.200ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/4BmakZzSLcdphQEcJFeKkRhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e7d7e117-cbd9-4aaf-9efc-e24fdcde4fb8.jpgEilean: Self-hosted digital islandsOur digital lives are increasingly fragmented across numerous centralised online services. This model concentrates power, leaving us with minimal technical control over our personal data and online identities. The long-term permanence of these platforms is uncertain, and their commercial incentives are often misaligned with user interests. We propose inverting this model: instead of centralising our data in proprietary silos, let’s centralise our presence under our own control using open, federated services. We introduce the concept of ‘digital islands’, or Eileans – self-hosted hubs for an individual’s or community’s online presence. By hosting services ourselves, we regain autonomy and control. Eilean is a project designed to simplify the creation and management of these digital islands. The core idea is to parameterise a complete operating system deployment by a domain name and a desired set of services. This allows users to easily deploy their own instances of federated services like Matrix, Mastodon, and E-Mail. We utilise NixOS to enable declarative, reproducible configuration and deployment of these services. This provides strong guarantees about the system’s state.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1d3adf66-92d9-4a2a-9537-7d267ff9a703/5c4b5f7d-47e1-4c91-a127-32b7fd7c6ccd-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4BmakZzSLcdphQEcJFeKkR361062026-02-04T17:55:53.258ZYESFOSDEMNOhttps://watch.eeg.cl.cam.ac.uk/w/iPX1Xx8LAZnuhojVYbyvABhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/3e6bd170-4542-4d2a-a65a-38a9554b8aa8.jpgPackage managers à la carte: A Formal Model of Dependency ResolutionPackage managers are legion. Every language and operating system has its own solution, each with subtly different semantics for dependency resolution. This fragmentation prevents multi-lingual projects expressing precise dependencies across language ecosystems, means external system and hardware dependencies are implicit and unversioned, and obscures security vulnerabilities that lie in the full dependency graph. We present the Package Calculus, a formalism for dependency resolution that unifies the core semantics of diverse package managers. Through a series of formal reductions, we show how real-world package manager features reduce to our core calculus. We define the language Pac to translate between distinct package managers and show we can perform dependency resolution across ecosystems.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/905d3833-a890-4ece-8ba2-cf6dbf5e2dcb/e049a886-930d-44ae-916f-8be8a657b433-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/iPX1Xx8LAZnuhojVYbyvAB177351962026-02-04T18:44:10.469ZYESFOSDEMNOhttps://watch.eeg.cl.cam.ac.uk/w/mcGduiez8scvNGwtWuFAfLhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/9839fb04-fe44-4942-ad09-155a5f578e29.jpgOpam's Nix system dependency mechanismThe OCaml language package manager, Opam, has support for interfacing with system package mangers to provide dependencies external to the language. Supporting Nix required re-thinking the abstractions used to interface with traditional package managers, but enables using Opam for development easy whilst benefitting from Nix's reproducible system dependencies. This provides one example of how Nix interfaces with other software development and deployment technologies.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/a398bcfc-e8fb-4ce0-a132-90c1235a3f78/b32f3a5b-e026-4aa6-995a-8e53ee355688-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/mcGduiez8scvNGwtWuFAfL3500152026-02-04T18:50:50.457ZYESFOSDEMNOhttps://watch.eeg.cl.cam.ac.uk/w/8UkzArqR37uD9q44rHHaqthttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/4dde3e1e-b23b-4dd0-819d-f4a1db113640.jpgFinding a Job after your PhDBio: Madeline Lisaius received BS and MS degrees in Earth Systems with a focus on environmental spatial statistics and remote sensing from Stanford University, Stanford, California, USA as well as MRes degree in Environmental Data Science from the University of Cambridge, Cambridge, UK. She is working towards the PhD in the Department of Computer Science and Technology at the University of Cambridge. She is focused on topics of food security and environmental justice, remote sensing, and machine learning.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/3ffeb678-406a-4729-9c0a-5c4a1a11cea7/5a2e3f6c-f7b0-49aa-ae7c-d05e67b45d02-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/8UkzArqR37uD9q44rHHaqt3392032026-02-06T17:50:31.530ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/ppe8LV3MAaYCAmeDMAfspihttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/13ad6d1a-95fc-4175-af9a-ece538e7bc0d.jpgCan Agentic AI Accelerate IUCN Red List Assessments?Abstract: The IUCN Red List of Threatened Species is one of the world's most important conservation resources–often referred to as a "Barometer of Life". It provides a standardised, evidence-based assessment framework for grouping species into extinction risk categories (from Least Concern to Critically Endangered) using quantitative criteria. The Red List is thus a critical indicator of the health of the world’s biodiversity, and guides policy and conservation action worldwide. However, Red List coverage is constrained by funding and availability of trained assessors. This results in significant data gaps (for example, fewer than 2% of invertebrates have been assessed), and a long tail of outdated assessments (over 25% are at least 10 years old). In this talk, I'll share initial research into how agentic AI could support the Red List workflow. I'll present results showing that AI coding agents can reliably pass the official Red List assessor training exam–and, crucially, explain their answers with citations to official guidelines. I'll also demonstrate how I leverage agentic coding to rapidly develop and maintain a real-time "evidence-base" dashboard, integrating live citizen science observations with relevant scientific literature. I'll close by outlining plans for next steps and future research directions, and then open the floor to questions and highly-welcomed feedback. Bio: Shane Weisz is a first-year PhD student in Computer Science at the University of Cambridge, supervised by Professor Anil Madhavapeddy. His research focuses on AI to support global biodiversity conservation.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/bd8063d9-0572-4d12-ba0e-8c2ac2614a1f/e71bcbf4-1601-4aef-b189-6a7a2cceee90-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ppe8LV3MAaYCAmeDMAfspi27450322026-02-13T19:51:41.509ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/eesGyr3GVuPEC6FuaptZ2khttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/3c3faf62-5def-4083-b04d-4d4a3b0a13e6.jpgIntroduction to TESSERA: Time-series embeddings for geospatial analysisMirror of https://www.youtube.com/watch?v=9yrpwFrwbGY Satellite remote sensing enables a wide range of downstream applications, including habitat mapping, carbon accounting, and strategies for conservation and sustainable land use. However, satellite time series are voluminous and often corrupted, making them challenging to use: the scientific community’s ability to extract actionable insights is often constrained by the scarcity of labelled training datasets and the computational burden of processing temporal data. The presentation will introduce TESSERA (Time-series Embeddings of Surface Spectra for Earth Representation and Analysis), an open foundation model that preserves spectral-temporal signals in 128-dimensional latent representations at 10-meter resolution globally. The model uses self-supervised learning to summarise petabytes of Earth observation data. TESSERA is shown to be label-efficient and closely matches or outperforms state-of-the-art alternatives. By preserving temporal phenological signals that are typically lost in conventional approaches, TESSERA enables new insights into ecosystem dynamics, agricultural food systems, and environmental change detection. Moreover, the open-source implementation supports reproducibility and extensibility, while the privacy-preserving design allows researchers to maintain data sovereignty. To current knowledge, TESSERA is unprecedented in its ease of use, scale, and accuracy: no other foundation model provides analysis-ready outputs, is open, and delivers global, annual coverage at 10m resolution using only spectral-temporal features at pixel level. This session is part of a two-session series, providing the theoretical introduction to TESSERA. The second session, a hands-on workshop, will be held on February 2nd, 2026. https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/6b27ecf5-25d0-48c8-bf55-dcca8a9ec419/f549fbf0-31ec-4940-9426-e5c32561b55e-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/eesGyr3GVuPEC6FuaptZ2k4441062026-02-14T15:48:27.241ZArtificial IntelligenceMachine LearningDeep LearningYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/4jpNbSDLvebs6ChjthkpGyhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/0ed98203-fd98-4e8f-bec8-d3b7fd09ca02.jpg3D Radiative Transfer Modeling of Heterogeneous Vegetation CanopiesFull Title: 3D Radiative Transfer Modeling of Heterogeneous Vegetation Canopies: from Reflectance to Photosynthesis Abstract: Accurate quantification of vegetation structure and function is critical for understanding the global carbon cycle and energy budget. However, the structural heterogeneity of vegetation canopies poses significant challenges to traditional remote sensing models, often leading to uncertainties in retrieving biophysical parameters. 3D Radiative Transfer (RT) modeling serves as a powerful bridge connecting remote sensing observations with the physical and physiological processes of vegetation. In this seminar, Dr. Qi will introduce the Large-Scale Remote Sensing Data and Image Simulation Framework (LESS), a new 3D radiative transfer model developed to address these challenges. LESS employs efficient ray-tracing techniques to simulate photon interactions within complex, heterogeneous 3D scenes. The talk will demonstrate how LESS simulates multi-modal data—including visible/multispectral images, thermal infrared, and LiDAR signals—with high fidelity. Furthermore, the presentation will extend beyond optical reflectance to discuss the modeling of photosynthesis and Solar-Induced Chlorophyll Fluorescence (SIF). By linking 3D structural information with physiological processes, the model provides a physics-based approach to validate satellite products and deepen our understanding of vegetation productivity under changing environmental conditions. Bio: Dr. Jianbo Qi is an Associate Professor at the Faculty of Geographical Science, Beijing Normal University, China. He obtained his PhD degree from Paul Sabatier University (University of Toulouse III) in France, and Beijing Normal University in China. His research interests lie at the intersection of remote sensing and computer graphics, with a primary focus on 3D radiative transfer modeling of complex land surfaces and LiDAR point cloud processing. Dr. Qi is the lead developer of LESS (http://less...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/1add7806-be54-4bcd-a226-dbe3645047c4/3f470c99-d344-400c-9b82-d4a61cdd1bf2-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/4jpNbSDLvebs6ChjthkpGy3501002026-02-24T03:20:55.949ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/dZNDoKiuH8sugfKLCWh8gShttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/5d839d69-9ec8-4a02-998c-900d935e068a.jpgEO-AI4GlobalChange: AI-Driven Earth Observation for Monitoring Global Environmental ChangeAbstract: Our planet is facing unprecedented environmental challenges, including rapid urbanization, deforestation, pollution, biodiversity loss, and climate change. At the same time, extreme events such as floods, heatwaves, and wildfires are increasing in frequency and severity, with far-reaching human, economic, and environmental impacts. Earth Observation (EO), combined with advances in Artificial Intelligence (AI), provides powerful tools for understanding these processes and supporting evidence-based decision-making. In this seminar, Professor Ban will discuss recent research at the intersection of EO and AI, with a focus on deep learning methods for monitoring environmental change at scale. She will present selected results from EO-AI4GlobalChange, a collaborative research project developing novel, globally-applicable deep learning approaches for analysing multi-sensor, multi-modal EO data. The talk will cover examples including 2D and 3D urban mapping, urban change detection, wildfire detection and near-real-time monitoring, flood mapping, and multi-hazard building damage detection. The seminar will also briefly introduce PANGAEA, a global benchmark for Geospatial Foundation Models, and discuss insights from the systematic evaluation of widely used foundation models across multiple geospatial domains. Finally, Professor Ban will briefly outline the objectives of the recently established AI4EO Working Group within Group on Earth Observations (GEO), which aims to advance GEO’s vision of Earth Intelligence for All through AI-driven Earth observation research, innovation, and collaboration. Bio: Dr. Yifang Ban is the Professor and Director of the Division of Geoinformatics at KTH Royal Institute of Technology, and an Associate Director at Digital Futures in Stockholm, Sweden. Before joining KTH as a full professor in 2004, Dr. Ban was a tenured Associate Professor at York University in Toronto, Canada. She received her PhD in 1997 from the University of W...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/693fcdfb-c87c-47d5-99a1-3b8deb53c874/5c0ac5a2-7451-42e4-904a-373100b12319-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/dZNDoKiuH8sugfKLCWh8gS3214032026-02-28T00:04:08.225ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/uumwtv1bbqwgscqEhwooHjhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/6bd01d77-eabf-4263-b9e6-3fee959588ed.webp[PROPL'25] Programming Opportunities for the Global Biodiversity Observation NetworkProgramming Opportunities for the Global Biodiversity Observation Network (Video, PROPL 2025) Jean-Michel Lord, Jamie M. Kass, Andrew Gonzalez, Michael Dales, and Anil Madhavapeddy (McGill University; Tohoko University; McGill University; University of Cambridge, UK; University of Cambridge, UK) Abstract: Our talk will lay out the overall state of global biodiversity observations via the global Biodiversity Observation Networks (BONs) community of practice. We describe the digital infrastructure behind them, and offer a set of provocations and opportunities towards motivating computer scientists to contribute towards some of the challenges. We believe that a wide array of contributions from programming languages, distributed systems and security all have a role to play in the meeting the urgent challenge to conserve biodiversity in the coming years. Presentation at the PROPL 2025 workshop, Oct 13, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN.https://watch.eeg.cl.cam.ac.uk/static/web-videos/b3d88a6a-d21f-43a2-8b8e-05d3a3a90d95-1080.mp4https://watch.eeg.cl.cam.ac.uk/videos/embed/uumwtv1bbqwgscqEhwooHj1464562026-03-08T08:45:24.452ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/2MVwdn8LvjKvhy2Jhj8Ehvhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/ec9934b3-cabb-42c9-856f-3e25a69a25c5.webp[PROPL'25] A FAIR Case for a Live Computational CommonsA FAIR Case for a Live Computational Commons (Video, PROPL 2025) Cyrus Omar, Michael Coblenz, and Anil Madhavapeddy (University of Michigan, USA; University of California at San Diego, USA; University of Cambridge, UK) Abstract: Scientists increasingly write software as part of large-scale collaborative workflows, but current tools make it difficult to follow FAIR principles (findability, accessibility, interoperability, reusability) and ensure reproducibility by default. This paper proposes Fairground, a computational commons designed as a collaborative notebook system where thousands of scientific artifacts are authored, collected, and maintained together in executable form in a manner that is FAIR, reproducible, and live by default. Unlike existing platforms, Fairground notebooks can reference each other as libraries, forming a single planetary-scale live program executed by a distributed scheduler. We describe the design of Fair Python, a purely functional subset of Python, and a foreign function interface for interoperating with existing code. Through three interleaved research tracks focusing on language design, interoperability, and distributed execution, we aim to create a next-generation collaborative scientific workflow system that makes best practices the path of least resistance. Article: https://doi.org/10.1145/3759536.3763802 ORCID: https://orcid.org/0000-0003-4502-7971, https://orcid.org/0000-0002-9369-4069, https://orcid.org/0000-0001-8954-2428 Video Tags: python, reproducible, scientific computing, functional, visualization, fair, reusability, splashws25proplmain-p18-p, doi:10.1145/3759536.3763802, orcid:0000-0003-4502-7971, orcid:0000-0002-9369-4069, orcid:0000-0001-8954-2428 Presentation at the PROPL 2025 conference, October 12–18, 2025, https://conf.researchr.org/home/icfp-splash-2025/propl-2025 Sponsored by ACM SIGPLAN, ACM SIGAda,https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/0e82977c-ba11-487f-bece-147fb1da104d/95c87248-f8df-4521-b4d4-0790058c0fd4-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/2MVwdn8LvjKvhy2Jhj8Ehv1546022026-03-08T17:29:34.500ZYESProgramming For the Planet (PROPL)NOhttps://watch.eeg.cl.cam.ac.uk/w/vJBmybFVSc7uS4jTd1Exyrhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/c1852c4e-8ba9-4050-aaf7-820b45b3fa7e.jpgThe Carbon at Risk Measure Can Unlock Financial Markets for Large-Scale Carbon RemovalAbstract: Meeting net-zero targets requires a rapid and large-scale increase in investment in carbon dioxide removal, and ensuring that investment is allocated efficiently across technologies with fundamentally different risk profiles. Carbon removal markets currently lack a standardised, quantitative measure of permanence risk, leaving buyers and policymakers reliant on coarse qualitative classifications that inhibit informed comparison and portfolio construction. Inspired by Value at Risk in financial markets, we propose Carbon at Risk (CaR): the additional removal that must be purchased to guarantee, at a given confidence level and time horizon, that a target quantity of carbon remains durably stored. We estimate CaR in two empirical applications with very different risk profiles: forest carbon, where Monte Carlo simulations calibrated to satellite-derived fire data yield a 95% CaR at 200 years of up to 80%, and geological storage (DACCS), where the 95% CaR ranges from 0.15% to 17% depending on regulatory regime. We then show how combining technologies in a portfolio creates a trade-off between cost and risk: the minimum cost of meeting a durability target depends on within-technology correlation and the relative price of safer alternatives. CaR provides a practical basis for calibrating buffer pools, comparing projects on a common scale, and designing cost-effective removal portfolios. Bio: Tom Bearpark is an environmental economist, whose research focuses on the impacts of climate change. He completed his PhD at Princeton University in December 2025. He is now a Post-Doctoral Research Fellow at the University of Exeter, and a Visiting Fellow at the LSE Grantham Institute.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/f0cc0c22-621b-4b05-843b-a68e32e526e5/d94844a2-2e9d-4ded-b475-80b6759f9fb8-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/vJBmybFVSc7uS4jTd1Exyr2857002026-03-09T01:23:54.635ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/sVyeYuQq4eRxfH14SWFpMqhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/0b8044c7-b268-4caa-a416-ac683de6bc81.jpgPILA: Physics-Informed Low Rank Augmentation for Interpretable Earth ObservationAbstract: Physically meaningful representations are essential for Earth Observation (EO), yet existing physical models are often simplified and incomplete. This leads to discrepancies between simulation and observations that hinder reliable forward model inversion. Common approaches to EO inversion either ignored this incompleteness or relied on case-specific preprocessing. More recent methods use physics-informed autoencoders but depend on auxiliary variables that are difficult to interpret and multiple regularizers that are difficult to balance. We propose Physics-Informed Low-Rank Augmentation (PILA), a framework that augments incomplete physical models using a learnable low-rank residual to improve flexibility, while remaining close to the governing physics. We evaluate PILA on two EO inverse problems involving diverse physical processes: forest radiative transfer inversion from optical remote sensing; and volcanic deformation inversion from Global Navigation Satellite Systems (GNSS) displacement data. Across different domains, PILA yields more accurate and interpretable physical variables. For forest spectral inversion, it improves the separation of tree species and, compared to ground measurements, reduces prediction errors by 40-71\% relative to the state-of-the-art. For volcanic deformation, PILA's recovery of variables captures a major inflation event at the Akutan volcano in 2008, and estimates source depth, volume change, and displacement patterns that are consistent with prior studies that however required substantial additional preprocessing. Finally, we analyse the effects of model rank, observability, and physical priors, and suggest that PILA may offer an effective general pathway for inverting incomplete physical models even beyond the domain of Earth Observation. The code is available at https://github.com/yihshe/PILA.git. Bio: Yihang She is a third-year PhD student in Computer Science at the University of Cambridge, supervised by Prof. Sri...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/da07e4e7-09de-44ca-88c8-d4b9fe696afe/bc365d97-2bb4-41e6-a18e-d0074c24396e-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/sVyeYuQq4eRxfH14SWFpMq2882012026-03-14T01:04:15.266ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/7eVdTkCekovjufoAK6CTXphttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/82ca3418-8a23-4819-82f0-c46f3a3a1444.jpgCambridge Evidence TAP OpenGL interactive visualiser14-year old Jens Kromdijk did a placement with Conservation Evidence in 2025 and worked on an visualizer for the knowledge graph of millions of full text papers that we have assembled for evidence synthesis. Jens worked with Sam Reynolds, Sadiq Jaffer, Will Morgan, Bill Sutherland and Anil Madhavapeddy from the University of Cambridge to build this visualizer using native OpenGL and an interactive user interface, allowing us to browse through the complex connections and metadata in the literature."https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/32884cdf-cdb1-487e-b413-9c12d0d0ad09/b266fe17-f959-419a-8880-85cc27db41b1-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/7eVdTkCekovjufoAK6CTXp1170152026-03-15T16:09:34.740ZYESEEG ResearchNOhttps://watch.eeg.cl.cam.ac.uk/w/mjMkiLBYcSSrfAjKahC9Vmhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/b9eba94b-ca98-4f99-8929-c94db1c50605.jpgReconstructing Landsat Archive 1997-2024+: Sun, Clouds, Snow, Noise and HumansAbstract: A serious obstacle to the total uptake of open Earth Observation data (Copernicus Sentinel images, NASA’s Landsat and similar) in daily lives is the steep data analysis curve required to get from raw images to Analysis-Ready, Decision-Ready/Relevant, not to mention Forensics ready data. The combined complexity of high data volumes, atmospheric disturbances (clouds, haze) and inconsistent coverage and diverse and complex signal physics (e.g. radar images vs optical images; sudden changes in land use) has resulted in the number of EO data applications remaining rather marginal. For example, in Europe, it is estimated that only a small fraction of farmers and forest managers use Sentinel images for decision-making. The recently generated Google DeepMind AlphaEarth (10 m global for 2017–2025) and Tessera embeddings being complete, consistent and ARD, provide an opportunity to decrease the steep data processing curve and enable thousands of applications. In our work, we have also consistently focused on making EO data more ARD and more usable, primarily by aggregating Landsat 1997–2025 values to bi-monthly (Consoli et al., 2024). In the current approach (Landsat ARD global mosaics V2 monthly) developed a 4–step process to derive improved quality mosaics: (1) first, we aggregate monthly reflectances across the whole time-frame (cca 30 years) and use these normalized values to detect outliers, (2) we then derive monthly median values with filtered reflectances (already significantly reduces clouds, snow and noise), (3) we then gap-fill values using convolutional filter and consistent land cover classes, and (4) we finally gap-fill all remaining values using modeling. For these steps we use a data fusion approach with annual ensemble land cover data at 30 m, together with MODIS EVI monthly (complete, consistent) and geometric temperature (a function of latitude and day of the year) as covariate layers to help improve gap-filling. Although using embeddings seems...https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/a4961932-e077-477d-857a-fb6ef3fb2236/114d91a0-9a29-40c4-970d-e21ed593d5ba-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/mjMkiLBYcSSrfAjKahC9Vm3489022026-03-20T19:54:54.795ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/ocp7Mq1R63GM66SPuHBRCXhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/e9fa9642-3dca-49a2-8b16-90675b9baaaf.jpgExploring Tessera EmbeddingsExploring Tessera Embeddingshttps://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/b3c07a72-df87-479d-a7d8-83ca6ec2c28b/74b7cc9a-3de7-45fd-bf56-1dd3937362ee-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/ocp7Mq1R63GM66SPuHBRCX3090272026-03-22T19:07:14.560ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/w/kYz3F77ihXrtuF3mkRt52Vhttps://watch.eeg.cl.cam.ac.uk/lazy-static/thumbnails/80f2b0d3-a568-46ef-8a4f-1512db8ebff5.jpgQuantifying Tropical Forest Degradation with Spaceborne LidarAbstract: Tropical forest degradation is an important contributor to forest carbon fluxes, but it is difficult to quantify its effects using existing remote sensing. As a result, degradation is difficult to incorporate into national reporting, with no internationally agreed-upon definition or consistent monitoring mechanism. This talk will cover ongoing international efforts to define and monitor forest degradation, highlighting mismatches between the policy and remote sensing communities. It will propose a novel method that takes advantage of occasional repeat measurements from a spaceborne lidar sensor (GEDI) to capture the impacts of forest degradation. Finally, based on this methodology, it will present results on the carbon emissions and forest structural changes induced by multiple drivers of tropical degradation. Bio: Amelia Holcomb is a postdoctoral researcher at the University of Maryland on NASA's GEDI mission team. Her work focuses on measuring dynamic frontiers of degradation, deforestation, and regrowth in dense tropical forests using spaceborne lidar and other remote sensing instruments. She is particularly interested in bridging disciplinary divides, bringing computer science techniques to large-scale remote sensing problems and ensuring that scientific outputs translate to policy outcomes. She earned her PhD from the University of Cambridge studying computer science and plant sciences. She also holds a masters degree from the University of Waterloo and a B.A. in mathematics from Yale University. Prior to her life as a research scientist, she worked as a software engineer at Google.https://watch.eeg.cl.cam.ac.uk/static/streaming-playlists/hls/a1c3a6ab-0554-469f-9eb3-76243a2fc727/36e2aef3-1079-4c87-8ba2-99ab57940668-master.m3u8https://watch.eeg.cl.cam.ac.uk/videos/embed/kYz3F77ihXrtuF3mkRt52V3514022026-03-27T19:09:47.956ZYESEEG SeminarsNOhttps://watch.eeg.cl.cam.ac.uk/c/eeg_seminars/videoshttps://watch.eeg.cl.cam.ac.uk/c/eeg_research/videoshttps://watch.eeg.cl.cam.ac.uk/c/4c/videoshttps://watch.eeg.cl.cam.ac.uk/c/propl24/videoshttps://watch.eeg.cl.cam.ac.uk/c/loco/videoshttps://watch.eeg.cl.cam.ac.uk/c/ecoresilience/videoshttps://watch.eeg.cl.cam.ac.uk/c/ce/videoshttps://watch.eeg.cl.cam.ac.uk/c/fosdem/videoshttps://watch.eeg.cl.cam.ac.uk/a/eeg/video-channels