Subseasonal predictions for climate services, a recipe for operational implementation

Climate Services (30), 2023

by Chihchung Chou; Nube González-Reviriego; Núria Pérez-Zanón (Barcelona Supercomputing Centre – BSC); Raül Marcos (Barcelona Supercomputing Centre – BSC; Department of Applied Physics, University of Barcelona); Marta Teixeira; Sara Silva; Natacha Fontes; Antonio Graça (Sogrape Vinhos); Alessandro Dell’Aquila; Sandro Calamanti (ENEA, SSPT-MET-CLIM)


The implementation of operational climate service prototypes, which encompasses the co-design and delivery of real-time actionable products with/to stakeholders, contributes to efficiently leveraging operational climate predictions into actionable climate information by providing practical insight on the actual use of climate predictions. This work showcases a general guideline for implementing an operational climate service based on subseasonal predictions. At this timescale, many strategic decisions can benefit from timely predictions of climate variables. Still, the use of subseasonal predictions is not fully exploited. Here, we describe the key aspects considered to set up an operational climate service from the conception to the production phase. These include the choice of the subseasonal systems, the data sources and the methodology employed for post-processing the predictions. To illustrate the process with a real case, we present the detailed workflow design of the implementation of a climate service based on subseasonal predictions and describe the bias adjustment and verification methodologies implemented. This work was developed in the H2020 S2S4E project, where industrial and research partners co-developed a fully-operational Decision Support Tool (DST) providing 18 months of real-time subseasonal and seasonal forecasts tailored to the specific needs of the renewable energy sector. The operational workflow can be adapted to serve forecast products to other sectors, as has been proved in the H2020 vitiGEOSS project, where the workflow was modified to provide downscaled subseasonal predictions to specific locations. We consider this a valuable contribution to future developments of similar service implementations and the producers of the climate data.