Add BrightSurf on Google Email

Oil spill: where and when will it reach the beach? Answers to prevent environmental impacts

A recent study used the MEDSLIK-II model to forecast the spread of an oil spill, predicting its arrival at French beaches 7 days in advance. The collaboration between CMCC Foundation and REMPEC ensured timely exchange of information, producing accurate forecasts that aided coastal preparedness.

The Arctic may influence Eurasian extreme weather events in just two to three weeks

New research suggests that the Arctic may impact Eurasian extreme weather events as quickly as two to three weeks, highlighting the need for more robust analysis techniques. The study aims to improve forecast accuracy by targeting specific observations and developing simulation experiments.

SourceInstitute of Atmospheric Physics, Chinese Academy of Sciences·JournalAdvances in Atmospheric Sciences·DateApr 2, 2020

How new data can make ecological forecasts as good as weather forecasts

A new study reveals that analyzing large-scale ecological patterns can help predict changes in ecosystems. Ecologists are using decades of weather data to forecast bird migrations and wildlife population fluctuations, which can inform conservation measures. The research aims to improve understanding of the climate-ecology link, enablin...

SourceUniversity of Wisconsin-Madison·JournalTrends in Ecology & Evolution·DateMar 9, 2020

Tropical cyclones: How they contribute to better forecast in the Maritime Continent

A new study reveals that tropical cyclones can create drying effects in the Maritime Continent due to induced zonal wind anomalies. Observational data shows that these events decrease precipitations in the region not directly affected by typhoons, highlighting their role in regional climate patterns.

SourceCMCC Foundation - Euro-Mediterranean Center on Climate Change·JournalProceedings of the National Academy of Sciences·DateFeb 10, 2020

Deep learning accurately forecasts heat waves, cold spells

Rice University engineers developed a deep learning system that uses capsule neural networks to predict extreme weather events, such as heat waves and winter storms, with high accuracy. The system achieves 85% accuracy in five-day forecasts by identifying patterns in pressure systems and regional areas.

SourceRice University·JournalJournal of Advances in Modeling Earth Systems·DateFeb 4, 2020

Wisdom of the crowd? Building better forecasts from suboptimal predictors

Researchers at the University of Tokyo and Kozo Keikaku Engineering Inc. have introduced a method for enhancing the power of existing algorithms to forecast the future of unknown time series by combining suboptimal forecasts. This approach was tested on real-world flood data and theoretical equations with chaotic behavior, resulting in...

UM-led experiment closes critical gap in weather forecasting

A UM-led experiment has shown promise in improving subseasonal weather forecasting, providing critical lead time for natural hazards like heat waves and polar vortex. The SubX project aims to provide better subseasonal forecasts to the National Weather Service, using complex computer models and large data infrastructure.

SourceUniversity of Miami Rosenstiel School of Marine, Atmospheric, and Earth Science·JournalBulletin of the American Meteorological Society·DateDec 6, 2019