As promised, I have posted the updated and expanded version 2.0 of ‘Pot Lid’. It corrects some minor errors and adds several pages of new information. Here are some highlights:
— There is now a section defining the process of air column density adjustment (ACDA) and explaining in more detail why it is not modeled in a hydrostatic GCM. Version 1.2 rather flatly and incorrectly said that models do not allow net transfers of air between layers. That is not technically right, although it works out that way in practice. In version 2.0 I explain why convection (the most likely source of ACDA) is strictly ‘virtual’ in a GCM and does not transfer air, while circulation in a GCM, which does permit transfer of air between layers, has little if any net impact on the vertical density profile. Thanks to the reader who gently reminded me about convergent and divergent flow.
— I have responded to a reader question about possibly detecting the long-term density surge using pressure data. That turns out not to be practical. Historically, radiosondes were not equipped with height-finding gear. They only had temperature, pressure, and humidity sensors. Without an independent height coordinate, the pressure data cannot be converted into density data. Modern radiosondes have GPS tracking, but of course there has been very little net warming in the last decade, and thus there is not much of a surge to detect in that data set.
Traffic to the ‘Pot Lid’ download page has been steadily growing. I will be approaching more bloggers in the coming weeks, asking them to link to the essay. This experiment in online science works better with your participation! Please leave comments, and tell your friends and colleagues about the essay.