For the case of 30 October 1999 from the field phase of the Mesoscale Alpine Programme (MAP), advection of cold air around the eastern edges of the Alps and warm air around the western edge of the Alps ahead of a synoptic ridge, set up a reservoir of colder air on the south side of the Alps and a reservoir of warmer air to the north. The depth to where the air was colder on the southern side was sufficient for a shallow feohn flow through the Brenner Pass. After the passage of the ridge axis, synoptic cold air advection provided another source of colder air, this time from the southwest, growing deeper with time and having a synoptically imposed cross-barrier flow component. The maximum depth to which the air upstream was colder than downstream extended just above the peaks of the highest mountains. An analysis of the detailed measurements across Brenner Pass showed that this depth was also the top of the layer that descended and accelerated down the lee slopes of the Wipp Valley. Upstream, air above the foehn layer had an even stronger cross-barrier component yet did not descend because it did not have colder potential temperatures than the downstream side at that level. Deep foehn never developed.
Additional measurements from aircraft, dropsondes, an instrumented car and automatic weather stations are presented for a detailed stratified hydraulic study of the foehn flow across Brenner Pass. An examination of other well-documented MAP foehn cases confirmed the conclusion from the 30 October event that shallow and deep foehns are mostly a response to differences in air masses between the upstream and downstream side. A cross-barrier component of the flow was a modification, but in itself not sufficient to cause the flow to descend down the lee slope, unless potential temperatures on the upstream side were colder in this layer than on the downstream side.