Not sure how you get that number for the phone. If at 150 dB that needs 0.005m^2 to get 5W, let's be super generous and say 10W with 50% efficiency of receive conversion, so 0.01m. An iPhone X is about 7.5 by 15cm so 0.01125m^2.
... more transmit area can be applied to compensate for that loss and maintain the same power incident on the receiver, up until the point you run out of transmit area.
I'm not sure this is true,
Assuming some level power, say 145dB, 150dB, whatever.... is deemed unsafe
The power level at the receiver side cannot exceed this safe limit, can it?
So If the effective phone size is 70cm^2, and the maximal power density that's still safe (on either side) is 150dB than 7W can really be transmitted.
But...
Assume the transmitter is actually transmitting a few watts in 0% humidity.
A day later, a receiver receives 1% of the power, can the transmitter increase it's output 100X?
It can't, for two reasons
1. To do that it must know what is the exact reason for the decline in received power is, if the reason is that my body absorbs 99% of the power I hope it's not going to increase it. Knowing the exact reason why something delivers less power, takes a human a day's work in the lab and is beyond the current capabilities of devices.
2. Such dynamic range means your costs are X100 higher, if your "dry day" cost is anywhere above $10, your wet day costs are sky high.
If someone were to follow the safety rules, I would agree that you should not have a power higher than that at any point - at transmitter, receiver, or inbetween. In that presentation, uBeam appeared to be saying there were no limits anymore. That was a surprise to me, and to those in the industry who chair that type of ultrasound safety group that I have spoken to since.
As sound travels, it loses power as heat, so as long as you do not increase the focus effect faster than you lose power in the air, you do not exceed even if you start at the limit. This is pretty standard in ultrasound medical imaging, and usually the numbers work out that the highest intensity is at the focus (and what the FDA usually worries about, in most cases that's the Mechanical Index).
And yes you're correct that at some level of transmitter area increase, you hit a hard physical limit. Is that limit within the range of normal use? What do you sacrifice as you add more transmit area? Loss of efficiency and reduction in the number of targets that can be served?
I think your numbers are a bit extreme, but yes temp and humidity do affect sound velocity and attenuation, for example here's some data for audio frequencies from NASA.
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19670007333.pdf - note responses aren't linear
And a simple tool for calculating loss (based on audio, but my recollection is that it's close enough)
http://www.sengpielaudio.com/calculator-air.htmBest to worst cases you're looking at maybe 1 to 1.5 dB/m difference. Significant, but not 100x.
As Dave and others have pointed out - distance, orientation, and being obscured by hands etc are likely to be the much more significant issues.