Science blog

Science blog banner image

How can we make ensemble forecasts more skilful?

Mark Rodwell discusses innovative diagnostics which are identifying atmospheric flow types that may lead to poor forecast performance. Such diagnostics can help to prioritise modelling efforts on aspects which might be expected to lead to the largest forecasting improvements.
Aeolus blog banner, Simulated L2B wind product

Improving forecasts with new wind data from ESA's Aeolus mission

At ECMWF we are very excited about the prospect of using the novel space-based wind profile information of Aeolus to improve our weather forecasts. We are proud to take a significant role in the operational Ground Segment processing and in the expert teams that will work on achieving a useful wind product.
Blog banner image; processor

New computer architectures and next-generation numerical weather forecasting

The next few years will see a more diverse range of computer architectures and vendors than we’ve enjoyed for over a decade. Simon McIntosh-Smith, a Professor of High Performance Computing from the University of Bristol who collaborates with ECMWF, explains the challenges and opportunities this will bring to the numerical weather forecasting community.
Image showing river network across Europe

Extending forecasts of global river flow out to seasonal timescales

Rebecca Emerton discusses her work to develop new GloFAS seasonal forecasts of river flow which have the potential to give even earlier warnings of flood and drought risk around the globe.
Sandu science blog banner, polar prediction

Improving prediction and climate monitoring of polar regions – challenges and priorities

As WMO's Year of Polar Prediction gets well under way, Irina Sandu and Peter Bauer discuss the challenges of forecasting in the polar regions and highlight the priorities for ECMWF.
HPC image, science blog banner

The 2018 ECMWF HPC workshop – making progress on scalability

The issue of scalability will feature as the headline topic at this year's ECMWF high-performance computing (HPC) workshop in September. Here Peter Bauer, head of ECMWF’s Scalability Programme, talks about what is perhaps the biggest revolution in numerical weather prediction since computer-based forecasts began.
Andersson science blog banner; excerpt from C3S multimodel seasonal forecast

Reaching out to the Asian meteorology and hydrology community

Erik Andersson, Deputy Director of the Forecast Department, will be presenting ECMWF products and what they can deliver at InterMet Asia, the region's largest assembly of meteorology professionals.
Fossil fuel emission maps obtained from global/national inventories

Earth surface processes, human impact and predictability

Gianpaolo Balsamo discusses the importance of two-way coupling between the atmosphere and Earth’s surface, and how humans are altering these interactions. He shares his drive and excitement about bringing recent research developments into operational use.
Wildfire

Wildfires – a hazard like no other

Francesca Di Giuseppe discusses some of the unique challenges involved in predicting wildfires and how ECMWF is contributing to European and global fire information systems.
Flooding in Europe

Improving forecasts of European flooding

David Lavers talks about his work on extreme precipitation and the development of effective flood forecasting tools.
Meteosat visible channel image showing evening convection over central Africa, 16:00 UTC, 7 May 2016

The complexities of assimilating cloud- and precipitation-sensitive observations

Alan Geer explains how new observations sensitive to cloud and precipitation can improve forecasts.

ERA5: the new reanalysis of weather and climate data

Hans Hersbach talks about ERA5, a new reanalysis of weather and climate to replace ERA-Interim.
Science blog 2017-09-04 - speeding up weather forecasts

Speeding up weather forecasts by using single precision computer calculations

Peter Dueben looks at whether single precision can help to improve our computational efficiency.