Bryan Wei Zhi Tan

Kudos to Dong Chen for championing this important initiative, which has just been published in Nature Geoscience. With support from leading voices in the #earthobservation community, the comment calls for greater global access to China’s extensive Earth observation archives. Enhanced sharing and interoperability of these valuable datasets would accelerate climate research, strengthen disaster preparedness, and expand our collective understanding of the planet’s changing systems. Through opening these archives to international collaboration - much like NASA and ESA have done over the years - we can unlock enormous scientific, societal, and economic benefits for the global community.. Read the paper here: https://lnkd.in/eFrMJ7mf #EarthObservation #ClimateChange #RemoteSensing #GlobalCooperation #OpenData #NatureGeoscience #GeospatialScience #Sustainability

84

9 Comments

📢𝗪𝗲𝗹𝗰𝗼𝗺𝗲 𝘁𝗼 𝗷𝗼𝗶𝗻: 𝟮𝟬𝟮𝟱 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽 𝗮𝗻𝗱 𝗧𝘂𝘁𝗼𝗿𝗶𝗮𝗹 𝗶𝗻 𝗚𝗲𝗼𝗔𝗜 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗺𝗲𝗻𝘁𝘀 𝗮𝗻𝗱 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀🌍 📅 14-15 July 2025 📍 The Hong Kong Polytechnic University The Research Centre for Artificial Intelligence in Geomatics and Jockey Club STEM Lab of Earth Observations at The Hong Kong Polytechnic University will hold a 𝘁𝘄𝗼-𝗱𝗮𝘆 𝘄𝗼𝗿𝗸𝘀𝗵𝗼𝗽 𝗮𝗻𝗱 𝘁𝘂𝘁𝗼𝗿𝗶𝗮𝗹. This international workshop explores cutting-edge GeoAI advancements and applications in planetary science, ecosystems, infrastructure, and cities. Participants will dive into AI technologies transforming Earth observations, environmental sciences, and urban studies, learning to solve complex geospatial problems and derive scientific insights. The hands-on tutorials enable skill development and effective application of GeoAI methods through practical experiments, exposure to unconventional applications, cutting-edge techniques, and grounded exercises bridging theory and practice. These sessions foster idea exchange, especially among young researchers. Learn more and register: https://lnkd.in/gAj7vh5Z

42

📢𝗪𝗲𝗹𝗰𝗼𝗺𝗲 𝘁𝗼 𝗷𝗼𝗶𝗻 𝘂𝘀 𝗳𝗼𝗿 𝘁𝗵𝗲 𝟮𝟬𝟮𝟱 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽 𝗮𝗻𝗱 𝗧𝘂𝘁𝗼𝗿𝗶𝗮𝗹 𝗶𝗻 𝗚𝗲𝗼𝗔𝗜 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗺𝗲𝗻𝘁𝘀 𝗮𝗻𝗱 𝗔𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀!🌍 📅 14-15 July 2025 📍 The Hong Kong Polytechnic University The Research Centre for Artificial Intelligence in Geomatics and Jockey Club STEM Lab of Earth Observations at The Hong Kong Polytechnic University will hold a 𝘁𝘄𝗼-𝗱𝗮𝘆 𝘄𝗼𝗿𝗸𝘀𝗵𝗼𝗽 𝗮𝗻𝗱 𝘁𝘂𝘁𝗼𝗿𝗶𝗮𝗹. This international workshop explores cutting-edge GeoAI advancements and applications in planetary science, ecosystems, infrastructure, and cities. Participants will dive into AI technologies transforming Earth observations, environmental sciences, and urban studies, learning to solve complex geospatial problems and derive scientific insights. The hands-on tutorials enable skill development and effective application of GeoAI methods through practical experiments, exposure to unconventional applications, cutting-edge techniques, and grounded exercises bridging theory and practice. These sessions foster idea exchange, especially among young researchers. Learn more and register: https://lnkd.in/gAj7vh5Z

45

2 Comments

Excited to share that maplibre-gl-lidar v0.6.0 is now available! This release adds support for the Entwine Point Tile (EPT) format (https://lnkd.in/ePWhQkGv), making it possible to stream and visualize billions of LiDAR points directly in the browser with ease and performance. The demo below showcases the visualization of an incredible 184 billion LiDAR points! Live Demo Viewer: lidar-viewer.gishub.org GitHub Repository: https://lnkd.in/d7DcX9YU Full Video Tutorial: youtu.be/lV3oKk2WTEo Huge thanks to U.S. Geological Survey (USGS) and the 3DEP Program (https://lnkd.in/e8FyWvA6) for acquiring nationwide LiDAR data, Hobu, Inc. for processing the point clouds into Entwine format (usgs.entwine.io), and the Amazon Web Services (AWS) Open Data Program (https://lnkd.in/exaWEejP) for hosting this massive and continually growing collection of over 350 TB of EPT data with over 75 trillion points. Jason Stoker Howard Butler Chris Stoner #lidar #opensource #opendata #geospatial #maplibre #deckgl

424

14 Comments

🚨 New publication alert! 🚨 Proud to share our latest paper in Journal of Geophysical Research: Earth Surface, led by the brilliant Christopher Hackney: 🔗 https://lnkd.in/eGbegS4r This study dives into the changing bedform morphologies of the Mekong River — and how intensive sand mining is reshaping them in dramatic ways. Using high-resolution multibeam echo sounder surveys, we compared natural and anthropogenically influenced bedforms. The findings are striking: ➡️ Anthropogenic bedforms are steeper (mean leeside angle 20.6° vs. 11.8°), taller (mean amplitude 2.8 m vs. 1.79 m), and shorter in wavelength than natural dunes. ➡️ These forms create greater hydraulic roughness (median ksf = 1.23 vs. 0.49), fundamentally altering flow and sediment transport dynamics. ➡️ Crucially, this shift in flow patterns prevents sediment from refilling mining pits, accelerating bed degradation. In short: sand mining isn’t just removing material — it’s engineering entirely new, more disruptive riverbed conditions. This has huge implications for flood risk, navigation, ecosystem health, and sediment continuity downstream. A major contribution to understanding how human activity is now a first-order control on fluvial processes in large alluvial rivers. #Geomorphology #SandMining #Rivers #Mekong #SedimentTransport #Hydraulics #EnvironmentalChange #JGREarthSurface #ResearchImpact

112

1 Comment

🌊 Flood Forecasting Using Satellite Data and Machine Learning (2025–2029) 🔍 A Self-Learning Journey from Curiosity to Application Over the past few months, I’ve been working on a fully self-initiated project to predict flood-prone areas from 2025 to 2029 across Paschim and Purba Medinipur districts of West Bengal. This project blends geospatial data, machine learning, and curiosity. 🔧 What I Did: I built flood susceptibility models using Random Forest and XGBoost trained on: 📌 NDVI & LST from MODIS (MOD13Q1, MOD11A2) 📌 Rainfall from CHIRPS (Climate Hazards Group InfraRed Precipitation) 📌 Flood Inundation using JRC Surface Water and Sentinel-1 SAR imagery 📌 All data from 2000 to 2024 were used for training and model validation 📌 Prediction horizon: 2025 to 2029 📌 Modeling tools: Python (Scikit-learn, XGBoost, RasterIO, Numpy, Pandas) 🛰️ For flood validation, I overlaid the actual flood extent of June 2025 (extracted from Sentinel-1 VV backscatter using Google Earth Engine) with predicted flood zones to analyze the model's performance in real-world conditions. 🎯 Why This Matters Floods are recurring threats in coastal districts like Medinipur. My aim was to create a spatial tool that could potentially help forecast flood-prone zones months or years in advance, based on climatic trends and land surface dynamics. 🙏 Important Note I want to openly say: I am not a modeling expert. I am a postgraduate student exploring this field out of pure curiosity and self-learning. There may be mistakes in my approach, assumptions, or data handling. 👉 If you're an expert or working in a similar field, I would be grateful for any feedback, corrections, or suggestions. Please feel free to reach out or comment. 🗺️ Study Area: Paschim & Purba Medinipur 📅 Training Data: 2000–2024 📈 Forecast Period: 2025–2029 📂 Data Sources: MODIS, CHIRPS, JRC SW, Sentinel-1 (via Google Earth Engine) 📌 This project is part of my broader work on flood monitoring and climate vulnerability mapping, which I plan to expand using deep learning and sub-basin analysis. #FloodForecasting #MachineLearning #RemoteSensing #GoogleEarthEngine #XGBoost #RandomForest #Sentinel1 #MODIS #CHIRPS #FloodRisk #ClimateChange #DisasterManagement #FloodPrediction #Medinipur #WestBengal #PythonGIS #SelfLearning #GeospatialAI #SAR #EarthObservation #NDVI #LST

197

18 Comments

"Climate change is a growing challenge in the modern world and its effects on water resources are complex and interconnected. Here at the Centre for Water Research we are dedicated to studying the impact of climate change on water treatment and developing innovative solutions to address these issues." - Professor Jiangyong HU (NUS Department of Civil & Environmental Engineering), Director at the Centre for Water Research (CWR). As part of World Water Day 2025, we highlight the amazing research conducted at CWR which focuses on dealing with the impact of climate change on Singapore's water resources and treatment processes. From preventing cyanobacterial bloom in water bodies, to maintaining efficiency in water treatment in the face of rising temperatures, CWR is committed to enhancing water conservation efforts in Singapore and contributing to the sustainability of water resources in the face of a changing climate. Read more about CWR's research here: https://lnkd.in/gqEuQwdi #CDE #NUS #ForgingNewFrontiers #CDEImpact #worldwaterday #sustainability

72

1 Comment

🌍From NASA IMERG Data to Interactive Rainfall Dashboards: A Scalable, Fully Automated Pipeline for Localized Precipitation Insights Web Dashboard - https://lnkd.in/gEQ48nGv 🔬 Why this matters Climate change is amplifying the frequency and intensity of extreme rainfall events, particularly in sensitive regions like the Himalayas. To study these events, researchers often face challenges: Downloading and handling huge NetCDF datasets. Extracting rainfall data for very specific areas of interest. Converting raw data into actionable insights with clarity and precision. To address this, I built a fully automated workflow that takes NASA IMERG hourly precipitation data and transforms it into an interactive dashboard, enabling researchers to visualize rainfall dynamics for their own regions with just a few lines of code. 🛠️ The Workflow: Step by Step 1️⃣ Data Acquisition – NASA IMERG Hourly Rainfall Downloaded IMERG Early Run hourly precipitation data (NetCDF format) from NASA’s GES DISC portal. This dataset provides high-resolution global rainfall estimates, perfect for analyzing localized hydrological extremes. 2️⃣ Defining the Study Area – Bounding Box Approach For demonstration, I chose Thunag, Himachal Pradesh (India). Using latitude–longitude boundaries, I clipped the massive global dataset into a manageable regional subset. This ensures precision: researchers can replicate this workflow for any area by simply changing the bounding box. 3️⃣ Data Processing – From Grids to Cells Extracted grid cell-wise precipitation data from the NetCDF file. Computed hourly, daily, and cumulative rainfall for each cell. Identified cells experiencing intense rainfall (potential cloudburst-like events). 4️⃣ Structuring Data – Pandas & Xarray Converted raw gridded rainfall values into a Pandas DataFrame. Each cell is enriched with: Geographic coordinates. Rainfall time series (48h, daily, cumulative). Flags for extreme events. This makes the dataset machine learning-ready for advanced hydrological modeling. 5️⃣ Visualization – Plotly Dashboards Built an interactive dashboard (dark theme for professional clarity): 🗺️ Map of Cells showing rainfall intensity across the study area. 📊 Time Series Plots to track hourly/daily rainfall. 🔄 Comparisons between Day 1, Day 2, and cumulative totals. 🔎 Zoom, hover, filter – researchers can explore their region intuitively. 🌟 Closing Note What started as a challenge to handle NASA’s massive NetCDF datasets turned into a reproducible, interactive pipeline for regional rainfall analysis. This tool can help researchers worldwide transform raw satellite data into meaningful, decision-support insights. 📌 open-source GitHub repository 👉 https://lnkd.in/gaAcuQJj 👉 Let me know if you’d like to collaborate or test it for your region! Contact Info - sayantanonfire@gmail.com

249

5 Comments

SUTD research fellow Dr Tok Yee Ching and Associate Professor Sudipta Chattopadhyay have developed an ontology framework to identify vulnerabilities in smart city infrastructure. By mapping potential threats, this approach enhances urban security and resilience. #SUTD #SUTDInnovation #DesignedAtSUTD #SmartCities #Cybersecurity #UrbanInnovation

17

Bridging Quantum Computing and Ecology: A New Frontier for Biodiversity Research At NITheCS, Prof. Cang Hui’s team pioneers computational ecology tools to tackle one of biology’s grand challenges: forecasting insect biodiversity loss in a changing climate. Key Approaches: Leveraging time-series data + AI to project species trends. Modeling insect communities as nested networks (plasticity matters!). Exploring quantum algorithms for ultra-complex systems. Why insects? Their life cycles and adaptive traits make them critical—but vulnerable—indicators of ecosystem health. This work could redefine conservation strategies globally. #InterdisciplinaryScience #BiodiversityTech #NITheCS

11

Our paper on using AI and satellite imagery to map potato fields is now published! This isn't just about maps—it's about building resilient food systems and smarter resource management through data-driven strategies. Our study highlights how red-edge and SWIR bands are game-changers for identifying crop types in semi-arid agricultural regions.  A big thank you to my co-authors and everyone who supported this research. 👩‍🌾 👩‍💻 🌱 🥔 🛰️  Read the full study here: https://lnkd.in/dhZzmEui #AgTech #AI #SatelliteImagery #PrecisionAgriculture #FoodSecurity #RemoteSensing #MachineLearning #Sentinel2 #Landsat8 #SmartFarming

92

2 Comments