MuddyTreeMan, to answer your questions...[I work with material velocities nearly every week]. Ignoring air temperature and any ram air/venturi effect, there are three things that effect air intake efficiencies, 1, the primary air intake volume pk, 2. air filter quality/through put [and its regular change to ensure maintained clean pk air supply, and 3. the air velocity, and therefore it's volumetric rate m3/hr or CFM. Importantly, it is critical for any internal combustion engine [diesel or petrol] to have matched air intake volume to it's peak air combustion capacity - this is normally completed with an air filter box which is sized at a minimum of 2-4x pk air combustion [the larger the air box/air buffer the better provided the air intake volume/velocity exceeds peak combustion][to make sense of this we need to understand that all ICE's are in fact, air pumps [yes, really...
]
The above is the basic measurement at static/idle combustion and is your diesel stoichiometric base ratio of 14.5:1. What now changes is 1. off idle and 2. air intake volume throughout the rev' range [the greater the fuel combustion the greater air volume needed/used. What determines the across the full rev range pk air volume is the total air volume entering through the air intake system at any one moment and above all, its air velocity [we'll save lamina air flow, ram air and venturi techniques for another day
]
Provided the mushroom type air filter opening is greater than air volume combustion pk across the full rev range, your static/idle air efficiency will be correct for your/our engine/s. However, even if the mushroom intake's available air volume pk is sufficient, the air velocity will significantly reduce over a given length of air intake tract [in this instance, the snorkel length] :. it is near guaranteed that all snorkels will have a reduced performance effect on our diesel engines, unless they have some form of air velocity accelerator giving greater air volume; ala ram effect and/or venturi techniques.
So, there is a firm argument for a forward facing air intake as this directly aids air velocity through the snorkel length to the head intake manifold. But...and there's always a but...would you/we notice the difference on our basic tratter engines...? Well actually, yes. But for day to day driving we'll detect very little difference.
Note: I've often thought an inline compressor turbine within the snorkel tube would make a tremendous difference and...as an anally retentive engineer...I just might get out my log tables and slide rule and spend a few nights in the shed using some groovy Heath Robinson principles