(Click to enlarge)
The Department of Atmospheric Science at Colorado State University (CSU) has a long history of research and education in all aspects of the atmospheric sciences. Faculty, students, and staff use a wide variety of datasets in their research and teaching, from numerical models to gridded reanalyses to radar and satellite observations to measurements collected in the field and lab. However, the students in the department recognized two issues that were preventing these large datasets from being used to their full potential. First, although our department had consistently maintained a server running the Local Data Manager (LDM) to pull in real-time weather data from the Internet Data Distribution (IDD) system, that server was very limited in its capabilities and many in the department were unable to use it (or were unaware that it existed). Second, commonly used datasets such as reanalyses (ERA-Interim, NCEP-NCAR, MERRA, JRA) were scattered on various research groups' own servers, difficult to locate, and stored in numerous formats and resolutions. With these limitations in mind, the graduate students of the department initiated an effort to acquire a new server that would enhance the storage and use of both real-time and archived weather data. With a grant from the Unidata Community Equipment Awards program, along with a grant from the College of Engineering at CSU, we acquired a data storage server that makes many of these data sources readily and easily accessible to students.
The new data server provides 41 Terrabytes of total storage space, which gives us space for a number of real-time weather data feeds from the IDD alongside a number of archive datasets. We maintain archives of several datasets that are difficult to obtain after the fact because they are not archived by sources such as NOMADS; these datasets include the Short-Range Ensemble Forecast (SREF) system and the Global Ensemble Forecast System (GEFS), and the new Multi-Radar Multi-Sensor (MRMS) radar and precipitation dataset. We are building archives of up to a year in length for these datasets, as well as storing the commonly used reanalysis products. The choices of datasets to archive on this system are driven by student interest and decisions are made by a subcommittee of student representatives.
The Department of Atmospheric Science partners with the Cooperative Institute for Research in the Atmosphere (CIRA) and the Center for Multiscale Modeling of Atmospheric Processes (CMMAP), which combined hosts more than 100 research scientists, research associates, postdoctoral fellows, and visiting scientists. The data storage system is accessible to all members of this community, enhancing their ability to integrate multiple streams of data — from satellite and reanalysis products to archived GFS output — into their research.
The data storage system is also available for interactive access in our department's weather lab, which is used in synoptic and mesoscale meteorology classes and for student-led weather discussions. This lab, which features several high-resolution monitors, can easily display graphics from the Web as well as running interactive programs such as the Integrated Data Viewer (IDV).
In February 2015, several staff members from the Unidata Program Center visited our department to see how students and faculty are using this new system, and to help with installation and configuration of other tools such as RAMADDA. We now have a local RAMADDA server that offers another means for exploring and accessing the archived data.
These resources are increasingly being used in courses and research in the department, and we expect that the students in the department will identify additional opportunities for using the real-time and archived data in creative ways. We appreciate that the NSF and Unidata make these Equipment Grants available to universities to enable the full use of weather and climate data in research and education.
Have a problem with Unidata software at your site? Don't wait for someone from the UPC to come for a vist — drop us a line at support@unidata.ucar.edu to ask a question or get advice on keeping the data flowing.
