I would take dynamical model intensity forecasts with a grain of salt at this point. Many of them have a spin-up time needed to accurately represent the storm and can thus weaken it more than realistically expected/probable in the first 12-24hr of the model cycle. With Katrina near 24hr from landfall, the model forecasts of intensity become increasingly less reliable from here on out. Granted, it should weaken somewhat before landfall,

The trough in Texas, to me at least, appears to be narrowing in scope and having a hard time approaching the storm. As it moves inland and weakens, the trough should capture it, but it's going to be a tough call as to how much it affects the storm now. Note that these narrow troughs tend to not impact storms in anything but a positive manner; it is the classic setup for rapid intensification or is a marker of such a cycle that has already passed. Here, it's likely the latter.

Needless to say, anywhere +/- 50mi of New Orleans is very much still under the gun for a direct impact. For those who missed it earlier, a summary of the potential impacts for many areas can be found on the front page in my most recent Met Blog.