National Radar Mosaic and QPE Project

by J.J. Gourley and Jian Zhang

NSSL's Worldwide Integrated Sensor Hydrometeorology (WISH) team has taken on an ambitious project to seamlessly mosaic all 130 NWS and Department of Defense (DOD) radars across the U.S. The mosaic will provide the first high-resolution depiction of storms and quantitative precipitation estimation (QPE) products from coast-to-coast in real-time. Additional users for the national mosaic will include FSL and NCEP.

Mosaic

The ability to integrate data from multiple radars at their fullest resolution into a single framework was not without significant challenges. The 3D multi-radar reflectivity analysis and mosaic scheme is the result of nearly five years of research and development. The resulting scheme includes quality control, an adaptive objective analysis scheme, and a distance weighted mosaic technique. It takes into account different spatial scales of convective and stratiform weather echoes and reconstructs 3D reflectivity analyses using different interpolation techniques. The code has gone through extensive optimization to assure a high computational efficiency and operational applicability.

The 3D mosaic has been running in real-time in the FAA Corridor Integrated Weather System (CIWS) domain since August 2002. The CIWS 3D mosaic has 1km x 1km resolution and 21 vertical levels ranging from 1km to 17km above mean sea level. The mosaic combines base level reflectivity data from 32 radars with a five minute update cycle. Other product development teams at NCAR and MIT/LL have been using the CIWS mosaic data in convective and wintertime numerical weather modeling.

QPE SUMS

Challenges in radar-based QPE were discovered by NSSL scientists by examining precipitation products for cool season cases over the complex terrain of Arizona. Issues with beam blockages by terrain and bright band effects limited the value ofthe products. The Quantitative Precipitation Estimation and Segregation Using Multiple Sensors (QPE SUMS) algorithm, which runs in conjunction with the 3D mosaic, was designed to address these issues by integrating satellite and model data in its scheme. Today, it continues to provide improved basin-wide rainfall and snow-water equivalent estimates to the Salt River Project for the purpose of managing large-scale watersheds. In recent years, the 3D mosaicking and QPE SUMS software has been deployed in several other regions in the U.S. and around the world. Data quality control and estimation techniques have been improved from these experiences to accommodate precipitation from typhoons, hurricanes, extratropical cyclones, and mixed-phase events.

The next phase

The WISH group is taking its entire project to the next phase this spring. Up to now, the high-resolution radar data (i.e. level II data) that the algorithms use to operate were available for only select radars in the U.S. In 2004, the Collaborative Radar Acquisition Field Test (CRAFT) network will include all NWS and several DOD radars, thus enabling the WISH group to extend its software deployment to the whole U.S. The National Mosaic and QPE SUMS (NMQ) products will be available every 5 minutes on a 1 x 1km grid covering the conterminous U.S. This project has potential benefits to government, university and private sector users. The resolution and accuracy of the precipitation products will be a natural fit-in to NWS's Flash Flood Monitoring and Prediction (FFMP) project, for which NSSL provided delineated basins for the entire U.S. Moreover, the QPE SUMS products will be useful for verifying quantitative precipitation forecasts that are being produced at higher and higher resolution.

This project provides the WISH group an impetus to partner with other agencies and develop improved products collaboratively during this expansion. NSSL will be working with the Bureau of Reclamation in order to adopt QPE strategies they've developed for snowfall estimation. In addition, satellitebased techniques are being integrated through a collaborative effort with the NOAA National Environmental Satellite, Data, and Information Service. The 3D mosaic will become 4D through development of a new temporally weighting scheme that retains high-resolution storm structure from temporally-offset radar observations and minimizes any dampening effects on storm intensities. Several severe storm monitoring and QPE SUMS products will operate off these new mosaicked reflectivity fields. The WISH group plans to make the NMQ products available to NWS forecasters at local offices and river forecast center. They say, "we believe we can get the most valuable feedback by making the products available to forecasters on an experimental basis in their display systems. This will be a conduit for allowing forecasters to evaluate not just the NMQ products, but all NSSL severe weather monitoring algorithms." The WISH team has always been at the forefront of technologies related to high-resolution radar data mosaicking and QPE, and this new project is the next step towards the NMQ goal