To my knowledge, a discussion of body orientation is curiously absent from the literature (cf. Hooker and Fahlman 2016) on lung (alveolar) collapse at depth (two exceptions, see below).
The experimental studies I have come across, simulated dives with animals (or excised respiratory tracts) in a (presumably) horizontal (Ridgway et al. 1969, Kooyman et al. 1970, Denison et al. 1971, Moore et al. 2011 ) or 20°-head-up (Kooyman et al. 1972, Kooyman and Sinnett 1982) position. None mentioned experiments on specimens in a more natural head-down position (whichever angle) when simulating the dive descent and analyzing the sequence of compression events of the different respiratory tract compartments/the shifting of alveolar air into the non-absorptive conducting airways, the compressibility of the trachea, the depth of lung collapse at a specific diving lung volume, etc.
Moore et al. 2011 (Figures 2 and 3) and Garcia Párraga et al. 2018 (Figure 2) showed and discussed differential alveolar collapse during simulated dives of horizontally oriented marine mammals. At depth, the (compressed) volume of lung air is always found in the absolutely uppermost portion of the lungs.
Going one step further, i.e. not seeing the lungs in isolation but in connection with the conducting airways, I would think that, underwater, the body position of the breath-hold diving animal is as important a factor to consider as the diving lung volume (McDonald and Ponganis 2012) and the relative sizes and different degrees of compressibility of the trachea/the bronchial tree/the alveoli when one wants to estimate or model (cf. Bostrom et al. 2008, Fahlman et al. 2009) lung collapse depth in an effort to e.g. assess the risk of decompression sickness in marine mammals.
McDonald and Ponganis (2012) found the lung collapse depth of a free-diving sea lion to be greater than expected on the basis of pressure chamber studies. The authors‘ explanation centered on varying diving lung volumes (larger when diving deeper). I think that the difference in animal body orientation between this free-dive study and above pressure chamber studies also plays a critical role here.
An analogy: Imagine the respiratory tract of a marine mammal as a complex made of connected lifting bags (as used in scuba diving), each made of a different material; and therefore, exhibiting different degrees of compressibility (lungs>trachea>bronchi (Kooyman et al. 1970)). Underwater, depending on the orientation of the oblong model, air will always travel to the absolutely highest portions. This must lead to different sequences of compartmental compression with depth depending on the orientation of the model; i.e. whether it is submerged in a horizontal or vertical (lungs-up), or any angle-in-between, position.
Body orientation must ultimately affect the compressibility of the trachea and lungs (cf. predicted collapse depths; Murphy et al. 2012, Davenport et al. 2013); the more air in one of these compartments, the longer this compartment will resist collapse.
In comparison with the "classic" pressure chamber experiments, during a more natural head-down descent, tracheal compression must set in earlier/be more severe because the shrinking tracheal air will be pressed upward into the lungs to join the shrunken volume of original lung air. (-Provided that the trachea and the lungs communicate freely and there are no counteracting forces from the expiratory muscles). Hence, the onset of lung collapse, the ultimate forcing-out of lung air into the non-absorptive conducting airways, must be deferred. In other words, all things (i.e. diving lung volume, dive depth, size of the respiratory tract compartments) being equal, lung collapse sets in at a greater depth than predicted on the basis of chamber experiments with horizontally oriented animals.
Since large enough hyperbaric chambers exist to orient an experimental animal/carcass head-down, I am curious if anyone has ever tried it? And if not, why not? (I don’t get the logic of orienting an animal slightly head-up when trying to simulate a natural dive descent.)
[References attached.]