M. Tsidulko

A Dust Prediction System with TOMS Initialization

Abstract A dust prediction system, developed earlier at the University of Athens within the framework of the Mediterranean Dust Experiment (MEDUSE) project, was enhanced at Tel Aviv University to support the Israeli–American Mediterranean Israeli Dust Experiment (MEIDEX) project. These enhancements include development of a dust initialization approach using Total Ozone Mapping Spectrometer (TOMS) aerosol index (AI) data and improved specification of the dust sources. The skill of the model against the TOMS AI measurements was tested during two periods in March and June 2000 using four different scores. It is shown that the TOMS-based initialization has a significant positive impact on all the scores. For instance, the average distance between the predicted and TOMS-observed dust plumes drops from 350–485 to less than 200 km. Verification of model forecasts against surface dust measurements in Tel Aviv shows correlations of up to 0.69 based on 27 predictions, for both 24 and 48 h. One example of a narrow d...

Can sensitivity studies yield absolute comparisons for the effects of several processes

Abstract The contribution of a particular process is shown to be strongly dependent upon the other processes under investigation because of synergistic contributions. In general, as the number of relevant factors being investigated increases, the role of any specific factor diminishes because the synergistic interactions with the new factors are extracted. This is illustrated with the variations of the topographic role in the impressive lee cyclone deepening event on 3–5 March 1982 during the Alpine Experiment. When latent heat release, latent heat flux, and sensitive heat flux enter into our comparative study, the topographic contribution to the surface pressure lee cyclone deepening gradually diminishes down to 50% or more.

November 2, 1994, severe storms in the southeastern Mediterranean

Abstract The available data archive over the Middle East is complemented with mesoscale fields from simulations for 1–2 November 1994 event of hazardous rains in the southeastern Mediterranean with the aim of exploring the mechanisms of this extraordinary phenomena. The associated severe weather was not predicted at all by the present operational models. A global/limited area modeling system is used for numerical simulation of the process. The system consists of the Florida State University (FSU) Global Spectral Model (FSUGSM) and the Pennsylvania State University/National Center for Atmospheric Research (PSU/NCAR) MM5 model. The accuracy of the numerical simulation is found very sensitive to the moist processes in the model. The development was a consequence of intensive non-adiabatic processes caused by northward propagation of large quantities of warm and moist tropical air. Local terrain of northeastern Africa also played a significant role in the development of a Mesoscale Convective System (MCS) in the Mediterranean region on November 2, 1994. Over Israel, the system had a character of relatively large-scale trough with a narrow frontal system accompanied by rapidly developing small cyclonic vortices in the eastern Mediterranean.

Numerical Weather Prediction on the Supercomputer Toolkit

The Supercomputer Toolkit constructs parallel computation networks by connecting processor modules. These connections are set by the user prior to a run and are static during the run. The Technions Toolkit prototype was used to run a simplified version of the PSU/NCAR MM5 mesoscale model [9]. Each processor is assigned columns of the grid points of a square in the (x,y) space. When n x n columns are assigned to each processor its computation time is proportional to n 2 and its communication time to n. Since the Toolkits network computes in parallel and communicates in parallel, then, for a given n, the total time is independent of the size of the two dimensional array or the area over which the weather prediction takes place. A mesoscale forecast over the eastern Mediterranean was run and measured; it suggests that were the Toolkit constructed from ALPHA processors, 10 processors would do a 36 h prediction in only about 13 minutes. A 36 hours prediction with full physics for the whole earth will require 2 hours for 80 ALPHA processors.