Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Scott Collis Applicant's institution: ANL Applicant's division: EVS Project Name: rainfall Project title: Using NEXRAD radar retrievals to test validation and scale dependance of rainfall and subsurface moisture in the NOAA HRRR model Associated funding: DoE-BER-ARM Other Systems: Assorted ANL systems (my own 32 core machine). Systems at OLCF used for formal ARM work (open cluster with no quota system..) Science: ARM is moving to a model as a product paradigm. NOAA both already has a very large operational network (NEXRAD) and fine scale model (3km, High Resolution Rapid Refresh, HRRR). This project will perform rainfall rate retrievals from NOAA radar and do a scale analysis comparison to HRRR rainfall fields. Furthermore we wish to extend this in collaboration with the University of Chicago to investigate the reaction of the subsurface rainfall stimulus and see if the same relations we see in nature hold true for that observed in climate models. Project description: The project will use the Python-ARM Radar Toolkit (Py-ART) to ingest, correct and retrieve rainfall rates from dual polarimetric NOAA NEXRAD radars. Starting in Oklahoma and Chicago and move to investigating other regions. Since time steps (~10 minutes) are independent the methodology is pleasantly parallel. As part of the project we will be installing Py-ART (https://github.com/ARM-DOE/pyart) on Fusion/Blues and will likely need some support. The code is fairly efficient with higher demand components written in C and fortran. It is, however, memory intensive with 2-4GB per job needed. Fortunately this is perfect for the configuration of Blues. The project will begin in an exploratory phase, ensuring that the processing systems tested on smaller machines works well with the SGE. We will also experiment with Python based methods for job control (multi-proccessing module, iPython's cluster tools, Joblib) to see if we can further improve cluster usage. After this point we will scale to a full year of data from one radar (~51,000 radar volumes/files taking ~ 10 minutes of CPU) and then extend as our allocation allows. In addition we will be using our allocation for analyzing the results including comparison to gauge data. The team consists of myself and Jonathan Helmus. We are likely to involve a UChicago collaborator (Liz Moyer) as we have a proposal to the UChicago/ANL strategic partnership call (we will proceed with the above analysis regardless of the outcome). In addition we expect to bring on some summer students to work on this project. NOTE: The PI was on a previous project that failed to use its allocation. This project was to work on ARM programmatic work. It was decided by ARM management that this work should be performed at ORNL thus making it difficult to use our allocation. While the current proposal is in support of ARM Science it is tangential enough not to run into the same issue. Industry partnership: None as yet. We have, several times, tried to build an LDRD based on utility impacts. This is part of an overall demonstration exercise which we hope to build tangible progress to interest electrical utilities etc.. Project URL: Requested allocation: 70000 Q1: 10000 Q2: 10000 Q3: 20000 Q4: 30000 Justification: Storage requirements: We have our own 300+TB disk just installed. The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System