Making sense of the data: an “all of the above” approach to retrievals from a many radar network

 

Authors

Scott Matthew Collis — Argonne National Laboratory
Jonathan Helmus — Argonne National Laboratory
Cory C Weber — Argonne National Laboratory

Category

ARM infrastructure

Description

The Atmospheric Radiation Measurement facility operates an international network of highly heterogeneous, radars using multiple bands. All of these radars are multi-purpose in nature and many have different operational modes. In order to return value to facility stakeholders and maximize the scientific impact of the facility, the Radar Products Group adds value to the data collected by applying corrections and retrivals. Unlike many agencies that run a single radar or just one radar type, (eg NOAA NEXRAD) creating a one-size-fits-all application chain is not an option. If we tried to design systems for each radar and mode, we would end up with N*R packages where N is the number of retrievals and R is the number of distinct radar systems or modes. To this end, we have focused on building a modular system that allows users to perform their own retrievals as well as being intermeshed with the ARM Data Integrator (ADI) to allow pipelines to be developed on ARM systems. This presentation will highlight some of the progress and some of the difficulties we have and are facing. We will also show some ARM and non-ARM applications including cluster and cloud based applications including the use of Amazon Web Servies (AWS).