What does creating an operational service entail?

An operational climate or air quality service provides users with climate information tailored to their needs and updated on a regular and continuous basis, that can be used for decision making. There are several steps and methodologies involved in order to transform climate data to an operational service.

First, climate data from different climate prediction systems and databases are obtained. These raw data are then post-processed (e.g. bias adjustment, downscaling) to improve the reliability of the predictions. Forecast quality assessment is also performed to verify the forecast, as well as inform the user of the forecast quality and whether using this for decision making provides an added value compared to using past observations. Climate variables, indicators and risk indices tailored to the needs of users from specific sectors can then be computed and delivered to the user in the form of a climate service product or platform.

How do we do this?

Our team specialises in developing operational climate and air quality services using diverse methodologies to provide a continuous service or product to users. We mainly provide sub-seasonal to seasonal and decadal climate predictions, air quality and dust forecasts, as well as tailored information for a number of sectors, such as renewable energy, agriculture and retail. We are dedicated to working with users throughout the process of developing these services in order to co-produce products and tailored information that meet their specific needs.

A more detailed description of how we operationalise climate and air quality services is provided below:

  • Prediction data: We retrieve raw data, like model predictions and observational/reanalysis data, from various databases and climate institutions (e.g. Climate Data Store, NOAA). We use various prediction systems and models, including but not limited to NCEP CFSv2 for sub-seasonal forecasts, SEAS5 ECMWF for seasonal forecasts and CALIOPE for air quality forecasts.
  • Post-processing: Our team develops and uses automated processes for the download and post-processing of data in real time, and assessment of forecast quality.
  • Tailored indicators: A number of indicators are computed to focus the service on the needs of the user and specific sector. For example, capacity factors are calculated for the energy sector, which transform the essential climate variables (e.g. temperature, wind speed, solar radiation) into useful information to support energy sector decisions, such as how much energy can be produced in the next weeks and months. We have experience providing indicators for different sectors, such as renewable energy, agriculture, water management, health or retail.
  • User-centred services: Our team works to translate complex information into easy-to-understand visualisations and final products. In this process, it is important to identify the user's needs and understanding of different concepts, and find common ground in terms of the language used, as well as how data are presented.
  • Operational platform or product: We co-produce with the user the final product or platform where the climate and air quality information is presented, and provide tailored services, including outlooks for the upcoming weeks and months. We continuously monitor the service and provide automated workflows that produce forecast updates in near real-time. Quality controls and checks are implemented throughout the process to ensure the continuity of the service.