This seminar is a collaboration between RISE and Climate AI Nordics – climateainordics.com.
Title: AI in the wild: How Neural Networks Help Us Understand Our World Through Sound
Speaker: Jeppe Rasmussen, University of Copenhagen
Abstract: Bioacoustics, the study of nature’s sounds, has long been a powerful tool for studying wildlife. With the rise of artificial intelligence, particularly deep learning, the potential of this field has expanded dramatically. By applying advanced AI algorithms to bioacoustic data, researchers can now identify and monitor species with greater accuracy, even in environments where visual observation is difficult, such as dense forests or deep oceans.
This capability is especially critical as we face the sixth mass extinction. AI-enhanced monitoring offers new hope for conservation by providing deeper insights into the presence, behavior, and well-being of endangered species. Beyond detection, AI also opens doors to understanding animal communication and emotional states, thanks to its ability to autonomously identify and prioritize key acoustic features.
In this talk, I will present a series of case studies spanning multiple species and ecosystems to illustrate how cutting‑edge AI research can meaningfully advance our understanding of the living, complex world around us—and how these methods can help mitigate the biodiversity crisis we face and discover the surprisingly rich inner life of the animals surrounding us.
About the speaker: Jeppe Have Rasmussen received his PhD in bioacoustics from the University of Southern Denmark (SDU) in 2018 and has since worked at the intersection of bioacoustics and artificial intelligence in Germany, the USA, and Norway.
His research spans a wide range of species—from farm animals such as pigs to marine mammals like whales—and he was featured in the documentary "If Pigs Could Talk," recently broadcast across Europe, including Sweden.
2024 he published an article in Science exploring how AI and bioacoustics can revolutionize the monitoring of animal populations. He is currently based at the University of Copenhagen working on detecting and classifying vocalizations from narwhales in Greenland.
Location: Lindholmsallén 10, Gothenburg, or online using Zoom.
Date: 2026-03-26 15:00
This seminar will have an in-person presence at RISE office in Gothenburg located at Lindholmsallén 10, third floor. Make sure that you arrive in good time and ring the bell at the door.
Upcoming seminars:
- 2026-05-28: Kelsey Doerksen, Oxford Applied and Theoretical Machine Learninch Group (OATML)
- All seminars are 15:00 CET.
More information and coming seminars: https://ri.se/lm-sem
– The Learning Machines Team
]]>This seminar is a collaboration between RISE and Climate AI Nordics – climateainordics.com.
Title: Machine learning based classification of tree crops of Syrian Arab Republic
Speaker: Purnendu Sardar, Lund University
Abstract: Accurate mapping of tree crops is vital for regional resource management, ecosystem service assessment, and the support of local livelihoods within the Syrian Arab Republic. Despite their socio-economic importance, tree crops are frequently misclassified or omitted in global and regional cropland products due to their complex spectral signatures and structural similarities to natural vegetation. This study proposes an integrated machine learning framework that combines the computational power of Google Earth Engine (GEE) with Python to enhance classification precision of tree crops across Syria.
The methodology evaluates the integration of Global Ecosystem Dynamics Investigation (GEDI) LiDAR data with Sentinel-2 multi-spectral imagery to facilitate robust tree crop mapping. By utilizing GEE for the large-scale preprocessing of Sentinel-2 time-series data, the workflow generates high-dimensional, machine-learning-ready datasets that incorporate both structural and phenological variables. A Convolutional Neural Network (CNN) is subsequently trained in Python, chosen for its proficiency in processing time-series remote sensing data where temporal spectral patterns are more diagnostic than spatial textures. This approach allows the model to capture the distinct phenological cycles of various tree species, overcoming the limitations of traditional pixel-based or purely spatial classifiers.
The findings underscore the efficacy of the CNN in distinguishing tree crop cover with high efficiency, demonstrating that the fusion of LiDAR-derived structural metrics with multi-temporal satellite data significantly reduces classification errors. The resulting high-resolution tree crop map provides an essential tool for sustainable agricultural planning and resource allocation in Syria.
About the speaker: Dr. Purnendu Sardar is a Postdoctoral Research Fellow at the Department of Physical Geography and Ecosystem Science, and at the Centre for Advanced Middle Eastern Studies at Lund University. His research focuses on studying vegetation and land-use changes in the Middle East using remote sensing techniques. Originally trained in animal biology, Dr. Sardar holds a PhD from the Indian Institute of Technology (IIT) Dhanbad, where he conducted research on the impact of climate change on the mangrove ecosystems of Sundarbans, India. He has published in the fields related to geospatial applications for addressing environmental challenges. Following his PhD, Dr. Sardar worked in the climate-action industry, leading a team of geospatial experts to execute large-scale agroforestry projects in India. Additionally, he has served as a geospatial consultant for international companies. Dr. Sardar is interested in understanding conflicts, land-use changes and ecological processes at landscape level using geospatial tools.
Location: Scheelevägen 17, Lund, or online using Zoom.
Date: 2026-04-23 15:00
This seminar will have an in-person presence at the RISE office in Lund, located at Scheelevägen 17, IDEON Science park, house Beta5, fourth floor. Make sure that you arrive in good time and knock at the door.
Upcoming seminars:
- All seminars are 15:00 CET.
More information and coming seminars: https://ri.se/lm-sem
– The Learning Machines Team
]]>This seminar is a collaboration between RISE and Climate AI Nordics – climateainordics.com.
Title: Bridging data gaps in the Global South: Harnessing AI and Earth observation in the humanitarian sector
Speaker: Isabelle Tingzon, RISE
Abstract: This talk highlights how AI and Earth Observation (EO) are transforming data availability for two global challenges: climate resilience and universal school connectivity. In the Caribbean, where extreme climatic hazards like hurricanes, floods, and landslides can devastate over 90% of built infrastructure across Small Island Developing States, comprehensive and up-to-date housing stock information is critical to enable resilient housing operations but is often incomplete, inaccessible, or completely non-existent. The “Digital Earth for Resilient Caribbean” initiative addresses these gaps by integrating EO data such as drone, LiDAR, satellite, and street-view imagery with AI workflows to classify roof types, materials, and building conditions. These efforts aim not only to enhance disaster response and recovery but also strengthen long-term resilience through government capacity building and knowledge exchange. In parallel, the presentation showcases UNICEF Giga’s mission to connect every school to the Internet by 2030. Reliable school location data is critical for planning telecommunications infrastructure, yet often inaccurate or incomplete in many low- and middle-income countries. By leveraging ML and satellite imagery, we’ve developed scalable pipelines to map schools across Africa on a nationwide scale, helping to ensure that no child is left behind in accessing digital learning opportunities. Together, these projects demonstrate how Geospatial AI can help bridge critical data gaps in the Global South, supporting both climate adaptation and inclusive education.
About the speaker: Isabelle Tingzon is a doctoral researcher at RISE and PhD student at KTH. Previously, she was a Geospatial Data Scientist at the World Bank, where she spearheaded AI-driven climate adaptation projects in support of resilient housing operations. She has also worked as a Data Science Consultant at UNICEF, where she led the development and deployment of UNICEF Giga’s AI-enabled school mapping pipeline across low- and middle-income countries. Her other projects span high-resolution poverty estimation, informal refugee settlement detection, and disaster- fragility, conflict, and violence assessment, in partnership with humanitarian organizations and non-profits in Southeast Asia, Latin America, the Middle East, and the Caribbean.
Location: Scheelevägen 17, Lund, or online using Zoom.
Date: 2026-03-19 15:00
This seminar will have an in-person presence at the RISE office in Lund, located at Scheelevägen 17, IDEON Science park, house Beta5, fourth floor. Make sure that you arrive in good time and knock at the door.
Upcoming seminars:
- 2026-03-26: Jeppe Rasmussen, University of Copenhagen, digital and physical: Lindholmsallén 10, Gothenburg
- 2026-05-28: Kelsey Doerksen, Oxford Applied and Theoretical Machine Learninch Group (OATML)
- All seminars are 15:00 CET.
More information and coming seminars: https://ri.se/lm-sem
– The Learning Machines Team
]]>As extreme events following climate change are getting ever more frequent, humanity needs to tackle the new reality on a changing planet on many fronts. Many of the approaches which are in use or under development can be strengthened using advanced analytics tools and optimization techniques. The 2026 Nordic Workshop on AI for Climate will gather researchers from the Nordics. This one-day, in-person workshop, will take place in Copenhagen, Denmark, June 26th 2026. The workshop will feature a mix of keynotes, oral presentations, and posters around the topics of AI for climate change, including AI for biodiversity and the green transition. The workshop will be a meeting point for a wide range of researchers from (primarily) around the Nordic countries, but other interested people will be welcome too.
This workshop is the 2nd annual workshop in this series of workshops organized by Climate AI Nordics. For a summary of the 2025 edition click here.
Schedule
COMING SOON!
Keynotes
Céline Heuzé, University of Gothenburg
Céline is docent and Senior Lecturer in climatology at University of Gothenburg. She uses in-situ observations, climate models, AI and remote sensing to study the interactions between deep ocean, sea ice / glaciers, and atmosphere, mostly in the Arctic for now. She is a receiver of the 2022 Ocean Science division Outstanding Early Career Scientist award of the European Geosciences Union (EGU), and the 2024 Royal Society of Arts and Sciences in Gothenburg (KVVS) Birger Karlsson science prize.
TBD - more keynotes to be announced!
Invited talks
Joakim Bruslund Haurum, University of Southern Denmark
Joakim is an Assistant Professor at the University of Southern Denmark, and also affiliated with the Pioneer Centre for Artificial Intelligence. He works on both foundational and applied aspects of machine learning and computer vision, including the application of these methods in domains such as biodiversity monitoring.
Frida Berry Eklund, Klimatkollen
Frida is a Swedish climate strategist, author, and activist specializing in climate communication and civic engagement. She serves as a co-founder and spokesperson of the Swedish climate organisation, Klimatkollen – an independent, AI-powered platform designed to make climate data—such as corporate and municipal carbon emissions—accessible and transparent for the general public. Her role focuses on bridging the gap between complex data and public understanding to hold policymakers accountable. Frida is also appointed European Climate Pact Ambassador, promoting civic climate action across Europe.
TBD - more invited talks to be announced!
Poster presentations
The workshop will also have a poster session, more info will come later.
Topics
The workshop will feature presentations about Earth observation, computer vision, soundscape analysis, natural language processing, geometric machine learning, optimization, natural language processing, and multimodal modeling for climate related applications. Applications in green infrastructure, species distribution modeling, biodiversity monitoring.
Further topics of interest: The role of policy change, the use of AI to tackle climate change within built environments, transport, material, and product development. AI method development that benefits AI applications within climate change mitigation and adaptation.
Workshop information
Date and time: 2026-06-26, 09:00-18:00 + Social dinner until late
Place: TBD, Copenhagen, Denmark
Organization
The workshop is organized by Climate AI Nordics in collaboration with The Pioneer Centre for AI and Climes, The Swedish Centre for Impacts of Climate Extremes. The day will end with a poster session and dinner.
Registration and pricing
The deadline for registering to this workshop is TBD. See registration link here: TBD.
The fee for registration is TBD, which includes lunch and coffee. Invoices will be sent out to attendees subject to the registration fee.
]]>Welcome to the February 2026 edition of the Climate AI Nordics Newsletter. Since our launch in October 2024, we have together built a community of researchers and practitioners, currently comprising 225 members. Together, we are focused on cultivating a collaborative ecosystem dedicated to the intersection of artificial intelligence and climate science, spanning mitigation strategies, adaptation frameworks, and advanced environmental monitoring.
We welcome colleagues across academia, the public sector, and industry who share these research and applied interests to join us at climateainordics.com/join.
In this month’s issue, we highlight recent community developments, upcoming interdisciplinary events, career opportunities, and our featured researcher profile.
Community spotlight
Featured Project: Open-Insect: Benchmarking Open-Set Recognition of Novel Species in Biodiversity Monitoring
We propose a new open-set recognition dataset, Open-Insect, and evaluate 38 algorithms for new species detection on geographical open-set splits with varying difficulty.
Read more!
Featured member, February 2026: Paola Vesco
Paola Vesco is a Senior Researcher at the Peace Research Institute Oslo (PRIO), where she forecasts armed conflicts and studies how conflict and climate extremes shape societal vulnerability. She applies machine learning methods for theory testing in her research.
Read more!
News
Two new Nordic networks/centers related to biodiversity, sustainability, AI
2026-02-26 During 2026, two new Nordic networks/centers related to biodiversity, sustainability, AI will be initiated, with funding from
Nordforsk. Check them out
here and
here.
Read more!
Coming events
The Material Cloud Film Festival
Event date: 2026-03-05.
The Marie Skłodowska-Curie project LIBRA proudly announce The Material Cloud Film Festival, a three-night film festival exploring the often unseen material realities behind artificial intelligence, including labour conditions, extractive supply chains, and power concentration.
Read more!
Machine learning based classification of tree crops of Syrian Arab Republic
Event date: 2026-03-05.
Webinar with Purnendu Sardar, Lund University. Accurate mapping of tree crops is vital for regional resource management, ecosystem service assessment, and the support of local livelihoods within the Syrian Arab Republic. Despite their socio-economic importance, tree crops are frequently misclassified or omitted in global and regional cropland products due to their complex spectral signatures and structural similarities to natural vegetation. This study proposes an integrated machine learning framework that combines the computational power of Google Earth Engine (GEE) with Python to enhance classification precision of tree crops across Syria. The methodology evaluates the integration of Global Ecosystem Dynamics Investigation (GEDI) LiDAR data with Sentinel-2 multi-spectral imagery to facilitate robust tree crop mapping. By utilizing GEE for the large-scale preprocessing of Sentinel-2 time-series data, the workflow generates high-dimensional, machine-learning-ready datasets that incorporate both structural and phenological variables. A Convolutional Neural Network (CNN) is subsequently trained in Python, chosen for its proficiency in processing time-series remote sensing data where temporal spectral patterns are more diagnostic than spatial textures. This approach allows the model to capture the distinct phenological cycles of various tree species, overcoming the limitations of traditional pixel-based or purely spatial classifiers. The findings underscore the efficacy of the CNN in distinguishing tree crop cover with high efficiency, demonstrating that the fusion of LiDAR-derived structural metrics with multi-temporal satellite data significantly reduces classification errors. The resulting high-resolution tree crop map provides an essential tool for sustainable agricultural planning and resource allocation in Syria.
Read more!
International Conference: Climate Impacts in a Changing World 2026
Event date: 2026-03-09.
The Swedish Centre for Impacts of Climate Extremes (CLIMES) invites abstract submissions for the international conference Climate Impacts in a Changing World 2026, held in Uppsala on March 9–11, 2026. The event fosters interdisciplinary dialogue on the wide-ranging impacts of climate extremes on human and natural systems.
Read more!
Bridging data gaps in the Global South: Harnessing AI and Earth observation in the humanitarian sector
Event date: 2026-03-19.
Webinar with Isabelle Tingzon, RISE. This talk highlights how AI and Earth Observation (EO) are transforming data availability for two global challenges: climate resilience and universal school connectivity. In the Caribbean, where extreme climatic hazards like hurricanes, floods, and landslides can devastate over 90% of built infrastructure across Small Island Developing States, comprehensive and up-to-date housing stock information is critical to enable resilient housing operations but is often incomplete, inaccessible, or completely non-existent. The “Digital Earth for Resilient Caribbean” initiative addresses these gaps by integrating EO data such as drone, LiDAR, satellite, and street-view imagery with AI workflows to classify roof types, materials, and building conditions. These efforts aim not only to enhance disaster response and recovery but also strengthen long-term resilience through government capacity building and knowledge exchange. In parallel, the presentation showcases UNICEF Giga’s mission to connect every school to the Internet by 2030. Reliable school location data is critical for planning telecommunications infrastructure, yet often inaccurate or incomplete in many low- and middle-income countries. By leveraging ML and satellite imagery, we’ve developed scalable pipelines to map schools across Africa on a nationwide scale, helping to ensure that no child is left behind in accessing digital learning opportunities. Together, these projects demonstrate how Geospatial AI can help bridge critical data gaps in the Global South, supporting both climate adaptation and inclusive education.
Read more!
Swedish Climate Symposium
Event date: 2026-05-20.
The call for abstracts is now open for the Swedish Climate Symposium (Lund, May 20–22, 2026). The event focuses on “Science, Society, and Actions,” connecting researchers with policymakers. If you are applying AI or data science to climate challenges, don’t miss the chance to present your findings. Registration opens in January.
Read more!
Climes interdisciplinary summer school 2026
Event date: 2026-06-15.
The Climes Summer School 2026 at Uppsala University offers doctoral, postdoc, and advanced master’s students an interdisciplinary curriculum focused on climate extremes, public health, and societal impacts. The program features a specialized AI component where participants use deep learning and natural language processing to automate the extraction of climate data from texts. While the school is free to attend, applicants must submit their motivation and support letters by March 22nd, 2026, and are generally responsible for their own travel and lodging.
Read more!
Recent events
SBDI Days 2026: Artificial Intelligence in Ecology and Biodiversity Research
This event took place 2026-02-10. A conference hosted by SBDI in partnership with SciLifeLab’s Planetary Biology Strategic Area, exploring the transformative partnership between AI and biodiversity research.
Job openings
PhD in AI-Driven Circular Material Recovery for Energy Applications
RISE Research Institutes of Sweden is recruiting a PhD student to develop AI-driven approaches for recovering and reusing composite materials for energy applications, in collaboration with Swedish universities and industrial partners.
Deadline: 2026-02-27
Doctoral student in Predicting Hydrological Extremes with Earth Data
KTH Royal Institute of Technology is recruiting a PhD student to develop Earth observation–based indicators and machine learning models to improve forecasting of consecutive drought-to-flood events.
Deadline: 2026-03-01
Associate Senior Lecturer (Assistant Professor) in Data-Driven Evolution and Biodiversity
The Swedish University of Agricultural Sciences (SLU) is recruiting a tenure-track fellow to develop data-driven research in evolution and biodiversity, applying machine learning and computational methods to aquatic or terrestrial systems.
Deadline: 2026-04-17
Doctoral student in accelerated solar cell materials exploration
Chalmers University of Technology is recruiting a PhD student to optimize the stability of perovskite solar cells using automated and data-driven approaches.
Deadline: 2026-03-08
Machine learning engineer for computer vision and controlled pesticide-use.
Dimensions Agri Technologies (DAT) are recruiting a machine learning engineer. The selected candidate will help reduce and optimize the use of pesticides by developing targeted schemes through machine learning model assisted computer vision.
Deadline: Rolling
Your news in the newsletter!
Make sure to share your work with us, by sending us an email ( [email protected]), posting in our Slack or some other channel, and we’ll add it to the news feed! Take the chance of showcasing your work or your events to the community!
Also be sure to follow us on LinkedIn and BlueSky. Climate AI Nordics will have the most impact if you repost and like our stories!
Climate AI Nordics is a network of researchers working to harness AI in tackling the climate crisis through both mitigation and adaptation.
We promote the development of AI-based tools and optimization methods that support sustainable decision-making—helping reduce emissions, restore ecosystems, and build climate resilience.
]]>As AI expands into nearly every sector of society, it is frequently framed as clean, efficient and intangible. Yet behind its sleek interfaces lie vast infrastructures, hidden workforces and resource-intensive systems that raise urgent ethical and political questions. Through film screenings and expert-led discussions, the festival invites audiences to ask: What lies behind the cloud?
Each evening features a carefully selected film exploring a different dimension of AI. Topics range from human labour and bias in training data to mineral extraction, contamination, and the environmental and emotional costs of sustaining algorithmic systems. All events are open to the public and free of charge.
- Humans in the Loop
- Thursday 5 March, 2026, 17.15-20.00, screening starts 17.30
- Location: SKH Biosal, Valhallvägen 189, Stockholm
- Donde los niños no sueñan & Paradise Island
- Wednesday 18 March, 2026, 17.15-20:00, screening starts 17.30
- Location: ForumCiv, Alsnögatan 7, Stockholm
- In the Belly of AI
- Tuesday 14 April, 2026, 16.00-21.00, screening starts 17.00
- Location: Bio Skandia, Drottninggatan 82
More information: https://www.kth.se/ehl/ehl-events/the-material-cloud/the-material-cloud-film-festival-1.1456604
]]>Participants apply these concepts in practical exercise sessions, where they use AI tools for the automated extraction of climate impact data from various texts. Beyond technical skills, the school offers training in “soft skills” like science communication and engaging with policymakers. The program is free to attend, though participants must generally cover their own travel and lodging unless they are awarded limited available sponsorship. Applications are due by March 22nd, 2026, and must include a motivation statement and a letter of support. The daily schedule will typically run from 9:00 to 16:00, concluding with participant presentations and a social dinner.
]]>Authors: Yuyan Chen, Nico Lang, B. Christian Schmidt, Aditya Jain, Yves Basset, Sara Beery, Maxim Larrivée, David Rolnick
Project page: https://yuyan-c.github.io/open-insect-project/
TL;DR: We propose a new open-set recognition dataset, Open-Insect, and evaluate 38 algorithms for new species detection on geographical open-set splits with varying difficulty.
Global biodiversity is declining at an unprecedented rate, yet little information is known about most species and how their populations are changing. Machine learning has recently emerged as a promising tool to facilitate biodiversity monitoring, including algorithms for fine-grained species recognition from images. However, such algorithms typically are not designed to detect examples from categories unseen during training – the problem of open-set recognition (OSR) – limiting their applicability for poorly studied taxa such as insects. To address this gap, we introduce Open-Insect, a large-scale, fine-grained dataset to evaluate unknown species detection across different geographic regions with varying difficulty.
Open-Insect consists of closed-, open-set, and auxiliary splits for three geographical regions: Northeastern North America, Western Europe, and Central America, utilizing geographical metadata to study local and non-local semantic shifts. Local open-set species may not only be harder to detect, but also reflect challenges encountered when deploying species-recognition models in practice. For each region, we also include a realistic auxiliary dataset for OSR methods that benefit from training with such data.
We evaluated 38 OSR methods, and following OpenOOD v1.5, we categorized these methods into 1) post-hoc; 2) training-time regularization; and 3) training with auxiliary data. Post-hoc methods can be combined with a trained classifier and do not require further training. For training-time regularization, these methods require further training, but they only utilize the closed-set during training time. Finally, for methods that require training with auxiliary data, they require both the closed-set and the auxiliary set.
Though no method consistently outperforms the others across datasets, we observe that simple, efficient post-hoc methods, for example, maximum softmax probability, remain a strong baseline. Local open-set species are substantially more difficult to detect than non-local and non-moth open-set species. However, for all regions, the best performing method can achieve over 85% AUROC, indicating strong potential for automating species discovery with OSR in computer vision.
One common concern of domain experts to adopt ML-based tools for species discovery is that models only use background features instead of species-level fine-grained features to determine whether the species are from the open-set or not. We conducted an experiment to empirically verify that background features are not enough to achieve good OSR performance on Open-Insect.
We hope that the Open-Insect benchmark will draw attention to the problem of species discovery and enable further work within the ML research community on OSR and OOD detection methods for biodiversity. Such work stands to benefit the preservation of ecosystem services on which humanity depends.
]]>During 2026, two new Nordic networks/centers related to biodiversity, sustainability, AI will be initiated, with funding from Nordforsk:
Nordic BiodiversityAI Network: Strengthening Wildlife Animal Monitoring through Image-based Automation, led by Junwen Guo, Umeå University (Junwen is also a member of Climate AI Nordics).
NextGen Nordic Geotechnics: Bridging Sustainability, Engineering and Emerging Research Fields, led by Stefan Ritter, Norwegian Geotechnical Institute (Stefan is also a member of Climate AI Nordics).
Both of these networks include topics which partially overlap with and are included within Climate AI Nordics: AI for biodiversity/wildlife monitoring (e.g., 78 members have “Biodiversity” as one of their keywords on our people page, as of Feb 2026), and AI for tackling climate and sustainability challenges (e.g., 86 members have “Climate Modeling”, and 63 have “Sustainable Cities”, as a keyword as of Feb 2026).
As such, we at Climate AI Nordics look forward to future collaborations with these to-be-launched networks, and are excited to see more Nordic activities and initatives within these critically important areas!
]]>This seminar is a collaboration between RISE and Climate AI Nordics – climateainordics.com.
Title: Machine learning based classification of tree crops of Syrian Arab Republic
Speaker: Purnendu Sardar, Lund University
Abstract: Accurate mapping of tree crops is vital for regional resource management, ecosystem service assessment, and the support of local livelihoods within the Syrian Arab Republic. Despite their socio-economic importance, tree crops are frequently misclassified or omitted in global and regional cropland products due to their complex spectral signatures and structural similarities to natural vegetation. This study proposes an integrated machine learning framework that combines the computational power of Google Earth Engine (GEE) with Python to enhance classification precision of tree crops across Syria.
The methodology evaluates the integration of Global Ecosystem Dynamics Investigation (GEDI) LiDAR data with Sentinel-2 multi-spectral imagery to facilitate robust tree crop mapping. By utilizing GEE for the large-scale preprocessing of Sentinel-2 time-series data, the workflow generates high-dimensional, machine-learning-ready datasets that incorporate both structural and phenological variables. A Convolutional Neural Network (CNN) is subsequently trained in Python, chosen for its proficiency in processing time-series remote sensing data where temporal spectral patterns are more diagnostic than spatial textures. This approach allows the model to capture the distinct phenological cycles of various tree species, overcoming the limitations of traditional pixel-based or purely spatial classifiers.
The findings underscore the efficacy of the CNN in distinguishing tree crop cover with high efficiency, demonstrating that the fusion of LiDAR-derived structural metrics with multi-temporal satellite data significantly reduces classification errors. The resulting high-resolution tree crop map provides an essential tool for sustainable agricultural planning and resource allocation in Syria.
About the speaker: Dr. Purnendu Sardar is a Postdoctoral Research Fellow at the Department of Physical Geography and Ecosystem Science, and at the Centre for Advanced Middle Eastern Studies at Lund University. His research focuses on studying vegetation and land-use changes in the Middle East using remote sensing techniques. Originally trained in animal biology, Dr. Sardar holds a PhD from the Indian Institute of Technology (IIT) Dhanbad, where he conducted research on the impact of climate change on the mangrove ecosystems of Sundarbans, India. He has published in the fields related to geospatial applications for addressing environmental challenges. Following his PhD, Dr. Sardar worked in the climate-action industry, leading a team of geospatial experts to execute large-scale agroforestry projects in India. Additionally, he has served as a geospatial consultant for international companies. Dr. Sardar is interested in understanding conflicts, land-use changes and ecological processes at landscape level using geospatial tools.
Location: This is an online seminar. Connect using Zoom.
Date: 2026-03-05 15:00
Upcoming seminars:
- 2026-03-19: Isabelle Tingzon, RISE
- 2026-03-26: Jeppe Rasmussen, University of Copenhagen
- All seminars are 15:00 CET.
More information and coming seminars: https://ri.se/lm-sem
– The Learning Machines Team
]]>