6.6
Development and implementation of an automated system for analysis of mesoscale phenomena
Stephen Saleeby, North Carolina State University, Raleigh, NC
This current research undertook the task of developing and successfully testing the first automated near-real-time system for the detection of mesoscale phenomena. Five-minute data from the NWS Automated Surface Observing System (ASOS) network provide the primary source of input to the analysis system.
The system downloads high-resolution time-series data, performs data quality control, bandpass filters pressure and winds, objectively analyzes the surface observations over the data domain, and creates surface contoured fields available for viewing over the internet. The objective analysis is a modification of the Barnes scheme that incorporates Time-to-Space Conversion (TSC) for interpolating between ASOS stations. This scheme relies upon a TSC advection vector, which is estimated from current sounding data within the affected region. Given a typical average station spacing of 75km, the TSC process is able to resolve mesoscale features with wavelengths as short as 150km.
Sensitivity tests performed on the TSC objective analysis scheme reveal the best wave coherence for the following parameter choices: 1) 4 or more advection vectors from the data domain, 2) a time response parameter of 50 min that provides station overlap in TSC, 3) a "recommended" station spacing value that lies between the actual value and the uniform spacing value, and 4) a numerical convergence parameter of 0.5 that forces moderate convergence of the gridded to observed values on the second pass.
The effectiveness of the system is demonstrated through analysis of two recorded events. The first event was a mesohigh and wake-low event that tracked across central Alabama on 14 March 1999. Stations in Alabama reported pressure falls of 6mb in 10-15 minutes with corresponding winds of 30-40 kt. The second event involved a gravity wave train that propagated to the northeast through eastern Indiana and western Ohio on 12 March 2000. This series of waves was associated with multiple bands of snow that appeared in both the satellite and radar imagery.
Session 6, Testing and Simulation of Observing Systems: Part 2
Thursday, 18 January 2001, 8:30 AM-11:30 AM
Previous paper Next paper