Weather Forecasting and Climate Patterns

🌍 The latest World Meteorological Organization El Niño/La Niña Update brings important news for climate-sensitive sectors: 🔹 Current cooler-than-average Pacific sea surface temperatures are expected to return to neutral conditions (neither El Niño nor La Niña) between March and May 2025 — with a 60% probability, rising to 70% by April-June 2025. 🔹 The risk of El Niño redeveloping in this period is negligible, but uncertainty remains high due to the spring predictability barrier, a well-known challenge for long-range climate forecasts. Why should businesses care? 👉 Seasonal climate forecasts like this are not just scientific updates — they are powerful risk management tools that translate into economic savings worth millions of dollars for industries like #agriculture, #energy, #transport, #supplychains, and #insurance. 👉 Understanding how ENSO patterns will evolve helps businesses: ✅ Plan for #climaterisks and supply chain disruptions ✅ Anticipate shifts in #wateravailability, #energydemand, and #cropyields ✅ Align operational strategies to climate-sensitive markets (#commodities, #foodsecurity, #renewableenergy production) ✅ Enhance disaster preparedness and reduce costly damage from climate extremes Even in a warming world, businesses can no longer rely solely on historical weather patterns — they must actively integrate climate intelligence into decision-making. 📊 The latest forecast also comes in the context of record-breaking heat: January 2025 was the hottest January ever recorded, despite the presence of weak La Niña conditions since December 2024. What’s next? WMO’s Global Seasonal Climate Updates (GSCU) provide even broader insights — covering key climate drivers like the North Atlantic Oscillation, Indian Ocean Dipole, and tropical Atlantic temperatures — all critical for regional weather and climate risks that matter to business continuity and resilience. ✅ Takeaway for leaders: Climate intelligence is a competitive advantage. Use it to future-proof your business strategy and build climate resilience into your operations. 💡 Want to stay ahead of #climaterisks and opportunities? Follow #WMO for ongoing updates and insights. https://lnkd.in/e-hXFXPC

How far in advance can we reliably predict the weather? The answer isn't simple. It depends on many factors: the weather model used, the time of year, geographical location, and—most importantly—the type of weather pattern we're trying to predict. Stable weather regimes (like a persistent high-pressure system bringing clear skies) are generally much easier to forecast than dynamic or unstable situations. While a deep dive into forecast verification would require several posts (it’s a complex field involving statistics and mathematics), I’d like to show a recent example: the development of a strong—though fortunately short-lived—heatwave over Northern Europe. In the animation below, I’ve visualized temperature forecasts at about 1.5 km altitude, using outputs from a weather prediction model initialized at different times, from June 3 to June 12. Each forecast includes multiple lines, representing different ensemble members (scenarios with equal probability). I’ve also added the climatology to help highlight the heatwave signal. The forecast is for Hamburg, Germany. What stands out is that as early as June 6, the model began hinting at a heat spike around June 13—more than a week in advance. From that point on, the intensity of the heatwave became clearer, while the timing and duration continued to adjust until the final forecast. Had this been a low-pressure system or a moving front, the forecast would have looked much more chaotic. Looking at the early frames of the animation, you might wonder: is there any value in such long-range forecasts, when they seem so uncertain? The answer is yes. The spread of scenarios helps us quantify uncertainty and assess predictability. Saying how likely something is—even with uncertainty—is far more informative than saying nothing at all. Finally, a good model isn’t one that shows no change from run to run—it’s one where the actual evolution falls within the range of predicted scenarios. In the end, every forecast is just one of many possible paths the atmosphere might take.

A new era for climate forecasting: real-time 3D weather models in your browser 🌍 Earth observations are becoming sharper, faster and more accessible than ever before. Gaia3D Inc., led by SANGHEE SHIN, has launched a web-based 3D weather visualisation system for the Korea Meteorological Administration. It’s already fully operational and was recently used to track Typhoon Cỏ May in real time, directly through a browser. Why does this matter? Because the more precisely we can model the atmosphere, the better we can prepare for the impacts of climate change on people and nature. 🌪️ Extreme weather → faster, more accurate forecasting saves lives 🌱 Biodiversity under stress → visualisation highlights shifting habitats and risks 🌊 Rising seas & floods → instant, actionable data for planners and conservationists Not long ago, this kind of analysis required expensive high-end graphics workstations. Now it’s web-native, scalable and collaborative, making advanced Earth observation available to everyone who needs it. As climate impacts intensify, tools like this will become the backbone of environmental modelling, helping us anticipate change, safeguard ecosystems, and make smarter decisions for the planet. Watch the video below to see the system in action. #NatureTech #ClimateAction #Biodiversity #EarthObservation #FutureOfForecasting

Thrilled to unveil our latest work: multi-modal machine learning to forecast localized weather! We construct a graph neural network to learn dynamics at point locations, where typical gridded forecasts miss significant variation. Paper: https://lnkd.in/eBmfsJin Weather dataset: https://lnkd.in/ejCG8bKs Code: https://lnkd.in/eQg-JzQJ AI weather models have made huge strides, but most still emulate products like ERA5, which struggle to capture near-surface wind dynamics. The correlation between ERA5 and ground weather station data is low due to topography, buildings, vegetation, and other local factors. In this work, we forecast near-surface wind at localized off-grid locations using a message-passing graph neural network ("MPNN"). The graph is heterogeneous, integrating both global forecasts (ERA5) and historical local weather station data as different nodes. What do we find? First off, ERA5 interpolation performs poorly, failing to capture local wind variations, especially in coastal and inland regions with complex conditions. An MLP trained on historical data at a location performs better than ERA5 interpolation, as it learns from the station's past observations. However, it struggles with longer lead times and lacks the spatial context necessary to capture weather patterns. Meanwhile, our MPNN dramatically improves performance, reducing the error by over 50% compared to the MLP. This is because the MPNN incorporates spatial information through message passing, allowing it to learn local weather dynamics from both station data and global forecasts. Interestingly, adding ERA5 data to the MLP does not improve its performance significantly. The MLP struggles to integrate spatial information from global forecasts, while the MPNN excels, highlighting the importance of combining global and local data. Large improvements in forecast accuracy occur at both coastal and inland locations. Our model shows a 92% reduction in MSE relative to ERA5 interpolation overall. This work showcases the strength of machine learning in combining multi-modal data. By using a graph to integrate global and local weather data, we were able to generate much more accurate localized weather forecasts! Congrats to Qidong Yang and Jonathan Giezendanner for the great work, and thanks to Campbell Watson, Daniel Salles Chevitarese, Johannes Jakubik, Eric Schmitt, Anirban C., Jeremy Vila, Detlef Hohl, and Chris Hill for a wonderful collaboration. Thanks also to our partners at Amazon Web Services (AWS) for providing cloud computing and technical support!

🌡️ 𝗠𝗼𝘀𝘁 𝗽𝗲𝗼𝗽𝗹𝗲 𝘁𝗵𝗶𝗻𝗸 𝗰𝗹𝗶𝗺𝗮𝘁𝗲 𝗺𝗮𝗽𝘀 𝗰𝗮𝗻 𝘀𝗵𝗼𝘄 𝗷𝘂𝘀𝘁 𝘁𝗲𝗺𝗽𝗲𝗿𝗮𝘁𝘂𝗿𝗲 𝗢𝗥 𝗽𝗿𝗲𝗰𝗶𝗽𝗶𝘁𝗮𝘁𝗶𝗼𝗻. 𝗧𝘂𝗿𝗻𝘀 𝗼𝘂𝘁, 𝘆𝗼𝘂 𝗰𝗮𝗻 𝘃𝗶𝘀𝘂𝗮𝗹𝗶𝘇𝗲 𝗕𝗢𝗧𝗛 𝘀𝗶𝗺𝘂𝗹𝘁𝗮𝗻𝗲𝗼𝘂𝘀𝗹𝘆, 𝗿𝗲𝘃𝗲𝗮𝗹𝗶𝗻𝗴 𝗽𝗮𝘁𝘁𝗲𝗿𝗻𝘀 𝗶𝗻𝘃𝗶𝘀𝗶𝗯𝗹𝗲 𝗼𝗻 𝗮 𝘁𝗿𝗮𝗱𝗶𝘁𝗶𝗼𝗻𝗮𝗹 𝗺𝗮𝗽. Just finished creating this 3D bivariate climate visualization of Italy. The technique maps two variables onto a single color scheme: temperature drives the color's hue, while precipitation controls its saturation. This approach reveals Italy's incredible climate diversity in a single glance: 🏔️ 𝗧𝗵𝗲 𝗔𝗹𝗽𝘀: Cool temperatures with high precipitation (the blues and teals) 🌿 𝗧𝗵𝗲 𝗣𝗼 𝗩𝗮𝗹𝗹𝗲𝘆: A distinct climatic pocket with moderate temperatures and varying moisture. ☀️ 𝗧𝗵𝗲 𝗦𝗼𝘂𝘁𝗵 & 𝗜𝘀𝗹𝗮𝗻𝗱𝘀: The classic Mediterranean signature of hot, dry conditions (the rich magentas). ⛰️ 𝗧𝗵𝗲 𝗔𝗽𝗲𝗻𝗻𝗶𝗻𝗲𝘀: A beautiful climate divide running down the peninsula's spine. The 3D elevation isn't just for looks. It clearly shows how topography drives these climate patterns. This is the power of data visualization; complex relationships become instantly understandable. The data comes from 𝗪𝗼𝗿𝗹𝗱𝗖𝗹𝗶𝗺, and the visualization was made possible by an excellent tutorial from Milos Popovic, PhD. 𝗪𝗵𝗮𝘁 𝗰𝗼𝘂𝗻𝘁𝗿𝘆 𝘄𝗶𝘁𝗵 𝗮 𝗱𝗶𝘃𝗲𝗿𝘀𝗲 𝗰𝗹𝗶𝗺𝗮𝘁𝗲 𝘀𝗵𝗼𝘂𝗹𝗱 𝗜 𝘃𝗶𝘀𝘂𝗮𝗹𝗶𝘇𝗲 𝗻𝗲𝘅𝘁? 👇 #DataVisualization #ClimateScience #Italy #BivariateMapping #GIS #Cartography #R #Geography #Rayshader

You might have seen news from our Google DeepMind colleagues lately on GenCast, which is changing the game of weather forecasting by building state-of-the-art weather models using AI. Some of our teams started to wonder – can we apply similar techniques to the notoriously compute-intensive challenge of climate modeling? General circulation models (GCMs) are a critical part of climate modeling, focused on the physical aspects of the climate system, such as temperature, pressure, wind, and ocean currents. Traditional GCMs, while powerful, can struggle with precipitation – and our teams wanted to see if AI could help. Our team released a paper and data on our AI-based GCM, building on our Nature paper from last year - specifically, now predicting precipitation with greater accuracy than prior state of the art. The new paper on NeuralGCM introduces 𝗺𝗼𝗱𝗲𝗹𝘀 𝘁𝗵𝗮𝘁 𝗹𝗲𝗮𝗿𝗻 𝗳𝗿𝗼𝗺 𝘀𝗮𝘁𝗲𝗹𝗹𝗶𝘁𝗲 𝗱𝗮𝘁𝗮 𝘁𝗼 𝗽𝗿𝗼𝗱𝘂𝗰𝗲 𝗺𝗼𝗿𝗲 𝗿𝗲𝗮𝗹𝗶𝘀𝘁𝗶𝗰 𝗿𝗮𝗶𝗻 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻𝘀. Kudos to Janni Yuval, Ian Langmore, Dmitrii Kochkov, and Stephan Hoyer! Here's why this is a big deal: 𝗟𝗲𝘀𝘀 𝗕𝗶𝗮𝘀, 𝗠𝗼𝗿𝗲 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆: These new models have less bias, meaning they align more closely with actual observations – and we see this both for forecasts up to 15 days, and also for 20-year projections (in which sea surface temperatures and sea ice were fixed at historical values, since we don’t yet have an ocean model). NeuralGCM forecasts are especially performant around extremes, which are especially important in understanding climate anomalies, and can predict rain patterns throughout the day with better precision. 𝗖𝗼𝗺𝗯𝗶𝗻𝗶𝗻𝗴 𝗔𝗜, 𝗦𝗮𝘁𝗲𝗹𝗹𝗶𝘁𝗲 𝗜𝗺𝗮𝗴𝗲𝗿𝘆, 𝗮𝗻𝗱 𝗣𝗵𝘆𝘀𝗶𝗰𝘀: The model combines a learned physics model with a dynamic differentiable core to leverage both physics and AI methods, with the model trained directly on satellite-based precipitation observations. 𝗢𝗽𝗲𝗻 𝗔𝗰𝗰𝗲𝘀𝘀 𝗳𝗼𝗿 𝗘𝘃𝗲𝗿𝘆𝗼𝗻𝗲! This is perhaps the most exciting news! The team has made their pre-trained NeuralGCM model checkpoints (including their awesome new precipitation models) available under a CC BY-SA 4.0 license. Anyone can use and build upon this cutting-edge technology! https://lnkd.in/gfmAx_Ju 𝗪𝗵𝘆 𝗧𝗵𝗶𝘀 𝗠𝗮𝘁𝘁𝗲𝗿𝘀: Accurate predictions of precipitation are crucial for everything from water resource management and flood mitigation to understanding the impacts of climate change on agriculture and ecosystems. Check out the paper to learn more:  https://lnkd.in/geqaNTRP

Google DeepMind just unveiled GenCast, an AI forecasting model capable of predicting weather patterns up to 15 days in advance. With severe weather events on the rise, this tool could be life-saving—providing critical lead time for disaster preparedness, agriculture, and energy planning. One researcher described GenCast’s impact as “decades’ worth of improvements in a single year.” Its outpu is impressive: it outperformed a leading forecasting model 97.2% of the time. Why it matters for #sustainability professionals: • Better #weather predictions can inform #climaterisk assessments and #resilience strategies. Historical weather models just don't cut it anymore. • AI models like GenCast demonstrate how data and machine learning can help tackle weather-related challenges, paving the way for #innovation in areas like agricultural optimization and supply chain logistics. • These breakthroughs remind us that the intersection of #AI and sustainability isn’t just about tech—it’s about driving impactful solutions for people and the planet. To accelerate collaboration, DeepMind has made GenCast open source, sharing its code and weights to empower researchers worldwide: ➡️ Announcement - "GenCast predicts weather and the risks of extreme conditions with state-of-the-art accuracy" https://lnkd.in/eEVB7v6q ➡️ Academic paper in Nature Magazine - "Probabilistic weather forecasting with machine learning" https://lnkd.in/e7zsmXbZ ➡️ GenCast model code on GitHub - https://lnkd.in/en7DikWk and weights on Google Cloud - https://lnkd.in/eunX7BHC

Why is NOAA Research climate science important to forecasting tornados? Because there is a strong connection between certain climatic patterns and the number of Spring tornados in the United States. The NOAA: National Oceanic & Atmospheric Administration graphic below shows two very important things. First, the number of tornados we experienced this year is far above the 45-year average. Second, tornado frequency is correlated with the El Niño-Southern Oscillation (ENSO) status, with more tornados in the US in February-April during strong La Niña events and fewer during strong El Niño periods. These data come from National Weather Service's Storm Prediction Center and Climate Prediction Center. ENSO is a recurring climate pattern involving changes in the temperature of waters in the central and eastern tropical Pacific Ocean. Surface waters across a large swath of the tropical Pacific Ocean warm or cool by anywhere from 1°C to 3°C, compared to normal, creating either La Niña or El Niño conditions. In turn, these patterns influence the frequency of days that are ripe for tornado outbreaks in the US. When we understand current and likely future climate patterns, we better understand the likelihood of deadly Spring tornado outbreaks. #science #NOAA #research #climate #tornado

Nature has just published Microsoft Research's Aurora, the first foundation #model of the #earth system. Aurora outperforms operational #forecasts in predicting #air quality, #ocean waves, tropical #cyclone tracks and high-resolution weather, all at orders of magnitude lower computational cost. Aurora first learns how to generate forecasts through training on #weather patterns from over one #million hours of data. These data are derived from satellites, radar and weather stations, simulations, and forecasts. The model can then be fine-tuned to perform a variety of specific tasks such as predicting wave height or air quality. When #Typhoon Doksuri hit the Philippines in July 2023, the damage was devastating. As reported in Nature, Aurora accurately predicts Typhoon Doksuri’s landfall in the Philippines using measurements from four days in advance of the event (image below). Official predictions at that time mistakenly placed the storm off the coast of Northern Taiwan. Results like this show how #AI is paving the way toward democratizing high-quality climate and weather prediction.  Learn more here: https://lnkd.in/gNiM5tsQ Try it here: https://lnkd.in/gn9DZsry