To perform data-centric research in environmental and earth sciences, researchers need effectively query, select and access data products from different research infrastructures. When providing observation data continuously, infrastructure is expected to create and deliver customised data products, e.g. for specific geo-regions, time durations or observation parameters, to enhance its ability to serve the research communities. Such kind of services often have time-critical requirements; some tasks need to be carried out within specific time windows when the data products are needed for real-time modelling or simulation frameworks.