Searching for humpback whale vocalizations in the middle of the Pacific Ocean is sort of like searching for a needle in a haystack. So, how do you search for a needle in a haystack? You build a sophisticated audio payload and keep your fingers crossed. Luckily our needle is a loud one!
The construction of our acoustic payload required a lot of software and hardware engineering because we wanted our hydrophone to detect whales from miles away, and keep data from being lost in case of a mishap. Therefore, we acquired a “smart” digital hydrophone (HP) called the Ocean Sonics icListen 200 kHz, which can record a wide range of frequencies and target most cetaceans (Image 1).
All sorts of things could happen to a dangling HP in the middle of the ocean, so we attached the HP to Europa’s sub to ensure we wouldn’t lose it to entanglement or predation (Image 2). However, the self-noise of the sub is relatively loud, and we found that it interfered with the acoustic sampling. Therefore, we embarked on a design to decouple as much of the sub noise as possible and enlisted the help of an acoustician, Mike Holt, who has built similar devices his entire career. He helped us make a mount that would be durable but not too heavy, biofoul resistant for the several month mission and allows sound transference to occur. Thus, we used a thick-walled 3” type K copper pipe, which is an excellent biofoul deterrent and provides mounting strength. We mounted it to the center of the sub so that it did not add drag or change the pitch of the sub (Image 3).
To mask the mechanical noise, Mike first designed an inner casing tube out of low-density polyethylene into which the HP would slide. To keep it from hitting the bottom or the sides we manually cut 15 donuts of 10 pores per inch by 0.75” thick polyurethane open cell foam to encase the HP. We then sealed the tube and filled it with medical grade castor oil, which is well matched to the impedance of seawater. After testing it for leaks, we slid the inner casing into the outer copper pipe modified with 250 drilled holes. The holes enabled the interference to be no greater than 12th lambda of the highest frequency of interest (Image 4).
Overall, the entire process was painstaking but worth the effort to boost our chances of finding our needle in a haystack.
For question, please email HUMPACS@jupiterfoundation.org