AI-powered S2S model delivers new breakthrough levels of reliability and predictability in long-range weather forecasts
BOSTON — Aug. 22, 2023 — Salient Predictions, a leading pioneer in weather forecasting analytics, is adapting to increasing climate volatility by setting new standards in long-range weather prediction with a major upgrade to its revolutionary forecasting solution, subseasonal-to-seasonal (S2S) model, to deliver forecasts two to 52 weeks in advance. Developed by a leading team of scientists and engineers, the cutting-edge AI model employs the power of AI and calibration to create reliable probabilistic distributions and empower decision makers to navigate weather-related challenges and opportunities with greater confidence.
“Global climate change is spurring volatile weather patterns around the world,” said Matt Stein, co-founder and CEO of Salient. “This is presenting urgent challenges in weather forecasting and analytics. A substantial upgrade to our S2S model and a breakthrough for the industry, this release addresses these pressing issues, delivering major improvements in temperature, precipitation, and other forecasting variables. Salient’s new forecasting model stands out for its exceptional accuracy and reliability with new capabilities that enable confident decisions with long-range forecasts amidst unprecedented weather patterns.”
In the face of pressing global warming challenges, the new S2S model provides indispensable tools to address climate-related risks and vulnerabilities. The accuracy improvements for temperature and precipitation outperform benchmark models such as NOAA’s Global Ensemble Forecast System (GEFS), European Centre for Medium-Range Weather Forecasts (ECMWF), and climatology. Accuracy excels in sub-seasonal weekly forecasts, enhancing its value in critical decision-making scenarios for commodity trading, agronomic decisions, renewable energy production, and more. Based on a comparison of the Continuous Ranked Probability Score (CRPS) to reference models, the accuracy gain can reach up to 50%.
With its reliable probabilistic forecasts, the new model better equips stakeholders across various sectors from agriculture, energy, finance, and beyond with the knowledge to mitigate the impact of extreme weather events, optimize resource management, and prioritize climate adaptation strategies.
This upgrade brings a new dimension of understanding to weather forecasts through its user-friendly features. The integration of percentile selection enables users to assess best-case, worst-case, and most-likely scenarios, empowering decision makers to make informed choices based on a comprehensive range of possibilities. Additionally, the probability of exceedance feature enables users to evaluate the likelihood of values surpassing specific thresholds, further enhancing risk evaluation capabilities.
“As Salient’s strategic partner in Brazil over the last three years, we have been focused on co-developing a range of next generation agricultural risk management and crop index insurance solutions,” said Juan Carlos Castilla-Rubio, Chairman of SpaceTime Labs. “We are happy to report a step function increase in the accuracy and reliability of probabilistic two to 52-week weather forecasts in Brazil that can uniquely inform increasingly complex decisions in the global food and agricultural sector. This is particularly noteworthy as the climate volatility regime this year is becoming more extreme and unpredictable.”
For more information, go to: https://www.salientpredictions.com/forecasts
Salient combines novel ocean and land-surface data with machine learning and climate expertise to deliver the world’s most accurate subseasonal-to-seasonal weather forecasts and industry insights—two to 52 weeks in advance. Bringing together world-leading experts in physical oceanography, climatology and the global water cycle, machine learning, and AI, Salient helps enterprise clients improve resiliency, increase preparedness, and make better decisions in the face of a rapidly changing climate. Learn more at www.salientpredictions.com/ and follow on LinkedIn and Twitter.
All trademarks recognized.