We Are Experts In

High resolution weather forecasts - Empowering decision makers

Belgingur has developed a versatile forecasting framework which can be optimized for even the most demanding conditions. Combining ensemble 3D/4D-variational and observation nudging data assimilation methods, along with novel machine learning methods for data post-processing, we can offer forecast solutions tailored to the varied needs of our customers.

Your partner in weather

Our Solutions

For over two decades Belgingur has been dedicated to research in meteorology with special emphasis on numerical simulations of the effects of orography on atmospheric flow.

Operational weather forecasts

Belgingur has created a novel weather forecasting framework, called Weather On Demand – WOD, that is deployable in the cloud and on in-house hardware and which can be customized for any location world-wide at a very short notice.

Re-analysis of past weather

Atmospheric models can not only be used to create weather forecasts but also to give us a better understanding of past weather patterns. This can be achieved by a method called dynamical downscaling.

WOD

Weather On Demand

Belgingur has created a novel weather forecasting framework, called Weather On Demand – WOD, that is deployable in the cloud and on in-house hardware and which can be customized for any location world-wide at a very short notice.

Reykjavik Iceland
4°C Feels like -2°C
991hPa 81% 75% 10m/s
São Paulo Brazil
25°C Feels like 26°C
1008hPa 65% 75% 8m/s
Wrocław Poland
1°C Feels like -4°C
1015hPa 84% 100% 5m/s

Stay updated

Latest stories

High resolution ML weather forecast for Iceland

In 2023 we started the adventure of building our own machine learning weather models. The goal of our first experiment was to create a high-resolution weather model for Iceland. As the backbone architecture we chose the ClimaX weather model proposed by Microsoft.
Read More →

Wind Resources in Iceland

The weather in Iceland 1994-2011 has been downscaled at a horizontal resolution of 3 km with the ARW-WRF atmospheric model with boundary data taken from the atmospheric analysis of the ECMWF.
Read More →

Enhancing weather forecast accuracy through variational data assimilation and deep learning

The overall goal was to enhance Belgingur’s weather forecasting system by integrating it with a 3D/4D-variational data-assimilation module and to develop tools to disseminate forecasts, from a plethora of providers, in a unified manner as well as make the forecasts available to end-users through a novel fleet management system. We ...
Read More →
Scroll to Top