American Meteorological Society Demo

and reappearing spiral band, the user understands implicitly that the band is simply moving from one position to another. Likewise, the changing orientation of the asymmetry of an eyewall is obviously a gradual angular displacement. However, in some cases, the time gap between microwave images is so great that neither an algorithm nor a trained analyst can easily discern the pattern of change between the two images. The second main drawback is an effect we call “pulsing,” which occurs when well-defined convective cells outside of the eyewall advect rapidly and appear to regenerate in place. This occurs because the rotating fade is not the best-suited scheme for convective cells near the eyewall, which are more likely to evolve over time by shearing or entrainment into the eyewall(s). Fortunately again, when the user is aware of this effect, the artifact is more of an annoyance, and it is relatively easy to incorporate this knowledge into the interpretation of the animation. MIMIC-TC EXAMPLES AND VALIDATION. The best option for validation of MIMIC-TC is a direct comparison to ground-based coastal radars. The approach we take is through side-by-side comparison with contemporaneous radar animations to determine whether the end user will draw the same conclusions from both. The reasons for a qualitative rather than a quantitative comparison will become clear as the two products depict the same components of the TCs (eyewall dynamics, distribution of convection, etc.) but in physically different ways. The Atlantic hurricane event from 2005 with the best radar coverage was Wilma as it passed through Florida on 24 October FIG. 3. Comparison of (left) MIMIC-TC product with (right) NEXRAD 0.5° (Fig. 3). Most of Wilma elevation radar reflectivity composited from four sites for Hurricane Wilma (2005) from 0600 to 1800 UTC 24 Oct. Time resolution of MIMIC-TC is 15 remains in the short- range min; time resolution of radar composite is 7.5 min. Radar is spatially composview (<230 km) of four raited according to the nearest station to each pixel. The MIMIC-TC product is dar stations between 0600 created from 6 microwave satellite passes. Labels are as follows: (upper-left and 1800 UTC. Shown side label) NHC-reported maximum sustained winds ("Vmax"); (upper right label) by side, MIMIC-TC and the temporal separation from the microwave overpass nearest in time (either ground-level radar present before or after); (magenta perimeter) radar composite domain; (magenta the same histories of the "+") radar station positions. AMERICAN METEOROLOGICAL SOCIETY AUGUST 2007 storm structure as it crosses Florida. Both show that the strongest convection within Wilma is predominantly to the north of the eyewall. Between 0600 and 1400 UTC both the microwave scattering and radar reflectivity are stronger on the northern side of the eyewall, while a dry slot develops to the southwest. In both sequences, the eye expands and becomes less organized as it crosses into the Atlantic around 1500 UTC. Also at this time, both animations show the intensification of the eyewall on the trailing side of the storm, which was responsible for much of the property damage in eastern Florida. One significant difference between the two animations is that MIMIC-TC shows a well-defined spiral band on the eastern side of Wilma, and although the radar shows filaments of precipitation in the same area, the radar does not show a persistent eastern band like that in MIMIC-TC. The MIMIC-TC Wilma animation shows the artifacts that result from blending images in the vicinity of a coastal boundary, which in this case is the west coast of Florida. Notice how the Florida peninsula (dark blue) appears to move with respect to the coastline overlay. Fortunately, this artifact is easily distinguishable from convection, and only | 1193

Table of Contents Feed for the Digital Edition of American Meteorological Society Demo

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.