Challenge 13 - Communication of key trends and messages associated with climate change impacts at the national - regional scale supporting the European Climate Data Explorer

[RubenRT7]

Challenge 13 - Communication of key trends and messages associated with climate change impacts at the national - regional scale supporting the European Climate Data Explorer

Stream 1 - Data Visualization and visual narratives for Earth Sciences applications

Goal

The goal of the challenge is to provide users of the European Climate Data Explorer application with information related to key trends and messages associated with the impacts of climate change at the national – regional scale. The challenge is to ensure that this info is presented in an appealing, useful, and interactive way within the developed applications. The use of ML language modules can be a useful technology to meet this goal.

Mentors and skills

• Mentors:
  • EEA: Christina Photiadou
  • ECMWF: Samuel Almond, Edward Comyn-Platt, Chiara Cagnazzo
• Skills required:
  • Knowledge of handling geospatial data and time series
  • Processing and visualization of climate data - C3S used netcdf files, so experience in handling and interpretation of this data is required
  • Experience with Javascript and React which are both required for the development of the applications in the new applications framework
  • UX and communication of climate information
  • Machine learning could be a viable technology to implement this challenge

---

Challenge description

The European Climate Data Explorer is a joint activity between C3S and the EEA (European Environment Agency). The ECDE brokers climate impact indicators developed in C3S into a dedicated front end embedded in Climate Adapt (EEA) - https://climate-adapt.eea.europa.eu/en/knowledge/european-climate-data-explorer. The ECDE is based on a series of interactive applications that allow users to explore key climate impact indicators and hazards at the European scale. At present the applications provide a series of plots and charts that depict the measured (from reanalysis) and projected changes in these indicators up to the end of the century. The European Environment Agency (EEA) have a requirement from users that they required a summary of changes / highlights in an appealing / useful/ interactive way within the applications. Detection, estimation of trends, and associated statistical significance are important in climate risk assessments and the communication of the impacts of climate change.

This challenge would provide users with a quick, yet robust assessment of the expected changes as indicated by the ECDE applications. The ideal solution will provide an assessment of these changes, but also provide some supporting narrative. The development of ML language modules provides an opportunity to provide users with some context that describes the changes and trends in the climate impact indicators expected in their area of interest.

The ECDE present key climate hazards at the National to sub-national scale utilising NUTS regions (https://ec.europa.eu/eurostat/web/nuts/overview) and transnational boundaries. The ECDE contains numerous climate impact indicators that are produced from C3S ERA5 reanalysis and ensemble of CORDEX regional climate projections (bias adjusted) for both RCP 4.5 and 8.5. The ECDE is being updated to include relative and absolute values to allow reference period (derived from reanalysis to be compared to projections.

It would be useful for the non-technical users to be able to:

• identify significant trends associated with climate change in a visual, easy to interpret way
• The ECDE is for intermediate users, who request a visual assessment of significance in and trends associated with climate change. The assessment should be made for any region the user clicks on, from national (NUTS 0) to regional (NUTS 2)
• Consider significance of any trends in the context of the CORDEX ensemble (uncertainly – model spread)

• What data / system do you plan to use?
The ECDE will serve as the main platform to broker climate hazard information from C3S to EEA. Therefore the challenger must consider technologies that are compatible with the new C3S applications. The C3S web-applications will use JS-react for the front-end and any backend processing done with Python. The applications will be deployed as DockerImages within a Kubernetes framework. The applications should follow the guidelines set by ECMWF which includes instructions on the components libraries to use. Any new components developed should be written generically such that they could be added to an ECMWF components library. The applications should be appropriately documented with a view that this documentation may also be published on user facing web-sites.
The challenger should consider the UX, and is encouraged to explore use of supplementary narrative to support the trend analysis – ML language modules could be appropriate tool to support this goal.