Multiple processors, multi-platform code, and graphical user interfaces: Building an educational data assimilation tool using MATLAB

The development of a data assimilation tool that produces meteorological surface analyses is described. This tools takes advantage of the parallelization, graphical user interface (GUI) development, and multi-platform compiling features of MathWorks MATLAB software. These three software development tasks are often difficult to implement in other languages, especially in the case where the programmer has little formal computer science training. The data assimilation tool presented here, known as the Local Surface Analysis Tool (LSAT), produces a meteorological surface analysis over a limited area domain through a variational technique by assimilating observations from MesoWest over a 5-km resolution background field stored on the University of Utah THREDDS server. LSAT is controlled through an interactive GUI, making it easy to adjust parameters to optimize the analysis for a particular domain, and is platform independent.