The European Weather Cloud (EWC) – Collaboration Platform for Meteorological Development from Research to Applications

The European Weather Cloud (EWC) is the cloud-based collaboration platform for meteorological application development and operations in Europe and to enable the digital transformation of the European Meteorological Infrastructure. It consists of data-proximate cloud infrastructure established by the EUMETSAT and ECMWF. The EWC is open and partners partners can federate the access to their data or infrastructure assets.

The EWC is available for EUMETSAT and ECMWF Member and Cooperating States and EUMETSAT Satellite Application Facilities (SAFs) covering both research and operational use cases. Resources are also available for research initiatives, undertaken by one or more EUMETSAT or ECMWF Member States, via specific EUMETSAT Research and Developoperament (R&D) calls and ECMWF Special Projects. Currently, EWC hosts 16 R&D calls and Special Projects, lasting 1-3 years.

The EWC focuses very much on the community taking an iterative user needs-based approach in the development. Notably, research projects and operational applications use the very same environment, which smooths the transition from research to operations (R2O). The hosted services will also be augmented with the Software Marketplace, providing EWC users with the ability to easily share and exploit meteorological applications, algorithms, and machine-learning models. The EWC facilitates a Rocket.chat-based discussion platform for users to discuss and work together, promoting in practice the fundamental collaborative nature of this cloud offering.  

EWC hosts over 132 diverse types of use cases containing, for example, data processing, data services, application development, training, EO and weather data image production, post-processing, and experimenting with cloud technologies. To name a few examples in more detail, the FEMDI project, consisting of 11 European meteorological services, develops data services employing EWC for distributing open meteorological data to fulfil the EU Open Data directive requirements. Second, the Norwegian Meteorological Institute (MET) is piloting an infrastructure federation to create water-quality products by locating the processing chain close to the data. Lastly, numerous projects are developing machine-learning-based models in EWC, including e.g. nowcasting, medium-term weather forecasting, and feature detection from climate data.  

The main data holding accessible to the EWC users is the sum of all online data and products available at ECMWF and EUMETSAT. Services to access the data support both pull and push paradigms for long time series and time-critical access respectively. The services are supported by related functions, such as display, reformat, etc., as per applicable policies. The data offering will be augmented over time based on user needs. 

From a technological viewpoint, the initiative offers services that carry the highest benefits from cloud technology taking the users’ needs, use cases, and existing software into account. EWC looks forward to further develop the service from the current infrastructure-as-a-service (IaaS) model toward platform-as-a-service (PaaS). The plan consists of a Kubernetes engine, a high-throughput batch processing engine, function-as-a-service (serverless) capabilities, and several auxiliary services to support collaborative development and operations.