The most accurate seasonal insights. Made easy to use.
Salient delivers forecasts for precipitation, temperature, and other weather variables at up to 1/4° (25km) spatial resolution. Our models provide global coverage and can be customized to include optimized predictors and accuracy for smaller regions. Salient’s forecasts also serve as the foundation for powerful industry-specific analytics that help drive profitability and better business outcomes.Industry Analytics
Our subseasonal-to-seasonal forecasts are issued on a weekly basis. Long-range forecasts are issued every month.
1-30 days, 31-60 days, and 61-90 days
1-3 months, 4-6 months, 7-9 months, and 10-12 months
We provide forecasts for a wide range of weather variables, and a variety of ways to interact with them.
Salient shortens your time to insight by coupling our forecasts and industry analytics to decision tools including threshold alerts, recommendations, and more.
Salient’s web-based interface includes forecast maps, probability plots, historical performance graphs, and comparison forecasts at multiple spatial and temporal scales.
We provide raw forecast data in gridded NetCDF format through APIs that can be integrated into digital products, dashboards and virtually any forecasting tool.
Salient’s models are built on new insights into ocean and land data, research-based weather analysis techniques, and deep neural network architectures that we are relentlessly iterating and improving upon.
Billions of climate predictors
We employ global datasets of ocean, land, and atmospheric data to gain a unique and more robust picture of the factors that actually drive seasonal weather patterns.
Our model is continuously updated with recent climate observations, and integrates outputs of numerical models like GFS and ECMWF.
Always learning and improving
We are constantly testing and adjusting our models, and integrating the latest breakthroughs in machine learning to improve our forecasts and skill.
Quantitative evaluation of risk
Our forecasts include probabilistic distributions of outcomes to provide a comprehensive view of risks and the likelihood of extreme events.
Explaining physical connections
Salient is breaking ground in describing clear and non-obvious physical “teleconnections” in our machine learning model outputs.
We extract industry-specific forecasts to deliver meaningful insights to enterprise customers, covering a variety of formats and variables.
Rigorous backtesting for unbeatable accuracy
Salient offers a 30-year library of backtesting data that allows customers to validate our results and understand the true potential of our models. We use a rigorous cross-validation methodology to ensure our models perform well across the full spectrum of seasonal weather conditions. All forecasts are backed by a large probabilistic ensemble which reveals the range of potential weather outcomes for a given period.
Explore our research
Salient's forecasts are built on top of decades of foundational research into ocean conditions, climatology, and the global water cycle. Our team leads path-breaking research on concepts that have deepened scientific understanding and advanced numerous disciplines.
Skillful Long-Lead Prediction of Summertime Heavy Rainfall in the US Midwest From Sea Surface Salinity.
Geophysical Research Letters
Forecast of summer precipitation in the Yangtze River Valley based on South China Sea springtime sea surface salinity.
North Atlantic salinity as a predictor of Sahel rainfall
Implications of North Atlantic Sea Surface Salinity for Summer Precipitation over the U.S. Midwest: Mechanisms and Predictive Value.
Journal of Climate
The role of the subtropical North Atlantic water cycle in recent US extreme precipitation events.
Centennial Changes of the Global Water Cycle in CMIP5 Models.
Journal of Climate