A warning level would be good, but only if it's dependable. To the extent that it raises false alarm, it is counter productive. This is the problem these weather services have. It seems that the science of predicting weather is, well, more intuitive than quantitative. But it's also administrative. In Pingtung County, for example, there are many micro climates. Where we used to live, there was nearly a 90% chance of daily afternoon showers between July and October, by my observations. And yet the CWB would list Pingtung as having 0% chance of rain because, I believe, they lumped Pingtung into the Kaohsiung weather report. These were isolated yet daily events, and so in that sense the error was administrative.

Nalgae is different, obviously, being an isolated event in all respects, but the pattern developing is one that should raise alarm, or at least a warning of some sort. The two images below are from 6:30am Taiwan time today and from just now, 9:00pm. You can see how Nalgae was breaking up in the earlier one, and how the system that has been feeding us since has been stationary. Should the CWB have been able to predict this? I have no idea. But it looks like we're going to get at least another 100mm before it's over.