The international meteorological community is on a journey to embrace open data policies. Using our experiences from meteoblue, we give examples of how open data policies have fostered innovation, created private sector capabilities, and can “jump-start” meteorological value chains. These examples are typical and variations of them can be found in many companies and countries that implement effective open data policies.
meteoblue as a company wouldn't have been created and might not be sustainable without the benefit of open data. Already prior to forming the company the availability of WRF, GDAS assimilations, and GFS global model output enabled an initial NWP chain. Over time, the evolving processing chain provided an improved, easily accessible forecast for central Europe, in particular the Alps. Additional accuracy was achieved with running models at higher resolutions and tuning them. The development also led to an automatic post-processing and web-based visualisation chain with numerous innovative diagrams and maps.
The increased availability of open NWP data from national weather models and their further improvement allowed meteoblue to automatically evaluate an increasing number of forecasts for a given location, compare them to weather station data, compute a consensus forecast, and quantify the uncertainties of local forecasts. All information could be made available to end users in straightforward diagrams that in part of the world are used by illiterate farmers.
Availability of open weather station data allowed meteoblue to devise learning methods and further improve local forecast accuracy. Verification results are publicly available and allow users to assess how valuable the forecasts can be to them. Open radar data form the basis of a high accuracy nowcast and short term rain forecasts.
With its multi-model capability the processing chain is open to ingest additional models as well as precipitation radar and weather station data, giving their providers (and their users) instant access to all meteoblue post-processing capabilities. The multi-model processing chain is highly resilient against individual model or other data feed failures. Together, these capabilities allow partners, e.g. from small national weather services to both provide immediate access to their local models and service their communities and customers without interruptions. Therefore, meteorological value chains can be started up very quickly.
How to cite: Ramshorn, C., Müller, M. D., and Gutbrod, K. G.: Open data foster innovation and facilitate public-private engagement to create socio-economic value, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-508, https://doi.org/10.5194/ems2022-508, 2022.