5 Q’s with Hansi Singh, CEO of Planette – Center for Data Innovation
The Center for Data Innovation recently interviewed Dr. Hansi Singh, CEO of Planette, a San Francisco-based company pioneering AI-accelerated long-range weather forecasting. Singh explained how Planette bridges the gap between short-term weather forecasts and long-term climate projections, offering businesses high-resolution predictions of key environmental variables to support proactive decision-making.
This interview has been edited
David Kertai: What does Planette do?
Hansi Singh: At Planette, we specialize in long-range weather forecasting by integrating advanced AI technologies with comprehensive environmental data. We use coupled atmosphere-ocean Earth system models that consider multiple interacting components of the Earth’s climate system. By harnessing these existing models, we can extend our weather forecast beyond the typical 2-week short-term limit. We enhance weather predictability by combining observational data and physics-based models with machine learning techniques. By integrating our in-house AI learning model with existing weather models from NOAA and the European Center for Medium-Range Weather Forecasts, we create forecasts that blend AI-driven insights with traditional dynamical modeling. This approach enables us to provide long-term forecasts to businesses, NGOs, and governments, helping them make better operational and financial planning decisions, while managing environmental risks related to extreme weather events and climate variability.
Kertai: Why does Planette emphasize the use of ocean data for forecasting?
Singh: Due to its vast heat capacity, the ocean offers greater climate stability, making it a more reliable predictor for long-term trends than atmosphere-based data. By analyzing ocean temperature trends, currents, and heat distribution, our AI machine-learning models can predict weather volatility months or even years in advance. Oceanic phenomena like El Niño and La Niña also significantly shape global weather patterns, often having a greater impact than atmosphere-based factors by driving events such as droughts, hurricanes, and heavy rainfall. Integrating oceanic data sets with atmospheric and land data allows our AI models to identify complex patterns and correlations, resulting in more accurate sub-seasonal, seasonal, and annual weather predictions.
Kertai: What challenges do AI face in weather prediction accuracy?
Singh: Traditional weather prediction models focus on atmospheric conditions, with forecast apps typically providing predictions for up to 10 to 14 days. While AI has improved short-term forecasting, long-range predictions remain challenging due to the complexity of the Earth system. These forecasts require integrating the atmosphere, ocean, and land surface, making them far more complex than traditional two-week predictions. Additionally, AI models rely on large datasets for training, but the limited availability of accurate long-term data makes producing reliable forecasts difficult.
To address these issues, we are heavily focused on R&D to integrate advanced AI techniques with traditional physics-based models. This hybrid approach combines the data-driven power of AI with the physical accuracy of traditional models, allowing us to better capture complex physical relationships between the atmosphere, ocean, and land. Our aim is to create an AI machine-learning model that can effectively, efficiently, and effortlessly predict near-future weather patterns.
Kertai: Can you share examples of how different industries use Planette’s forecasts?
Singh: Various industries rely on our forecasts to make data-driven decisions and mitigate risks. For example, the Metropolitan Water District of Southern California, serving around 20 million people in the Los Angeles and San Diego areas, uses our forecasts to proactively manage water resources, prepare for droughts, and optimize reservoir operations.
Agriculture and finance are other sectors impacted by climate volatility. Companies trading energy, power, and agricultural goods use our forecasts to predict price shifts in commodities like cocoa and soybeans. Similarly, NGOs working in agriculture use our forecasts to guide small-scale farmers in deciding when to plant and harvest crops, reducing weather related losses.
Kertai: How does Planette plan to help predict the impacts of climate change?
Singh: Climate change is no longer a distant threat; it is already reshaping our world. Extreme weather events that we once considered unusual are now becoming the new normal. At Planette, we help businesses, governments, and communities adapt to this reality.
We don’t just provide climate data to industries seeking profit; we commit to making our insights accessible to organizations focused on disaster preparedness, public safety, and humanitarian efforts. By delivering accurate, high-resolution, AI-powered forecasts, we empower decision-makers to anticipate and mitigate the adverse effects of climate change. As climate change accelerates globally, we equip decision-makers with the knowledge needed to navigate and adapt to an unpredictable, rapidly changing world.
