Piezo bender hydrophone

Hi,
My daughter asked me to design a directional hydrophone for tracking whales and dolphins.  The range had to be close to 5nm (9km). Not an easy task. Most info on the net is to make omni or towing ones. Some info about underwater communication (voice). None about directional finders.
I ended up using a normal 28mm piezo disc out of a buzzer and mounted it in a 7" household funnel with small outlet.
I used a FET (2SK596) from an electret mic. capsule, and soldered that onto the disk. These fets have an inbuild protection diode and a 25meg gate resistor. And used a 2m shielded wire from a computer mic.
I cut a 25mm disc from a piece of double sided closed cell foam tape (the tape that sticks bathroom accessories to the wall) and stuck it to the ceramic part. This gives the small air space for the disk to bend.
I covered the whole back with 5min. epoxy. After hardening, I put it all in the funnel and potted the back up with more epoxy. After hardening, I used a small amount of epoxy to "varnish" the front of the disk. Avoid air bubbles here. Mounted it all to a 32mm drainpipe with the bottom 20cm bend in a 45 degree angle.
Because she wanted to use headphones, I used the good old LM386 in a standard 46db gain configuration (cap between pin 1 and 8 ). I used 1uF here to not amplify the lower frequencies. I used a single transistor amp stage between the hydrophone and the LM386 for more bandpass options. Total gain is more then 50db. It runs on a 9volt batt and draws about 5-10mA.
My question: who has done something like this before, who knows about frequency response of various sized piezo's and what effect the load impedance has on the output signal, and the effect on covering the brass with epoxy or glass or plastic, or keeping it bare. Are there better piezo shapes for directional finders. Am I missing something?
Any thoughts.

I'm sure you are in the wrong forum; I don't know if there is a "right" forum for this.

Directional passive sonar is over 70 years old.... there's no published literature?

From basic physics: for directionality (without fussy complication), your transducer must be more than a wavelength. Speed of sound in water is high, about 4X SoS in air. So at the same frequency a directional hydrophone must be much larger, aboud 4X, than a directional speaker for air.

That suggests for good human midrange directionality a wave-collector over a yard wide.

Which does suggest one reason for little info: not easily practical.

Whales are large, make large (low-pitch) sounds. Human hearing is falling at 100Hz. That suggests a 40 foot wave-catcher!

I said 'without complication'. There IS a complicated processor available: the human brain. In addition to simple "funneling focus" with earlobe or head-size, we can note teeny differences in sound arrival left/right, and get directionality very much better than our ear/head size would suggest.

Stereo hydrophone ought to give a bearing. A fixed array gives ambiguity but a turnable array can resolve this.

As confirmation: before fancy sonar, submarines found or avoided ships with basically two rubber hoses stuck in someone's ears, mounted so that a head-turn (or control) turned the array to resolve the 180 deg ambiguity.

I'm inclined to make the hydrophones 4.3 times further apart than the operator's ears, to bring the water-to-air wavelengths together. 24"-30" is not an inconvenient thing even in a dingy.

Thanks for the reply.

Yes, I knew about most of the problems that you mentioned in your reply.
Whales have a very broad vocal range, and I was hoping to get some directivity in the upper regions of the spectrum. Remember that I am using a funnel, and the opening determens the cutoff frequency, not the piezo.
I know that piezo's of this size are used in normal hydrophones, and to pick up seismic activity in the 0-10herz range. And that people are using funnels like this to locate whales.
Love to hear about real world experiences in underwater stereo locating with a handheld device.