Recent research has demonstrated the potential for the remote identification of human altered lithic material in underwater contexts (Hermand et al. 2011). The underlying principle of this method is the ability of low frequency sound waves to resonate within lithic materials of an ideal shape, making the material vibrate. These vibrations disrupt sound waves in a manner that can be consistently observed in SONAR profiles. These signals appear as “haystacks:” parabolic and wavy anomalies in the water column (Grøn et al. 2018). The underlying geophysics of this method were replicated in the Wellbore Acoustic Lab at Texas A&M University in November, 2019. Grøn and colleagues (2018) conducted experimental field tests of this method and tested a known submerged Neolithic site off the coast of Israel. The tests and Neolithic site both produced the haystack anomalies. Interestingly, scans of an unquarried bedrock chert exposure in Denmark did not produce the haystack signature, indicating that this method may only detect human-altered chert. This poster details an attempt to rigorously evaluate the potential of this method, and test its pitfalls.
As a proof of concept and replicative study, Smith and Kim conducted two experiments on a piece of Coastal Plains chert. In the first experiment, the chert was placed 1 m away from a speaker (emitter) and a microphone (receiver) was placed 2 cm away from the speaker. The speaker was programmed to emit sine burst signals with center frequencies from 4 to 13 kHz. The microphone measured the reflected excitation acoustic signals. The acoustic waves travel from the speaker to the chert and are then reflected back to the microphone. Our microphone detected two auto-correlation peaks at approximately 0 m and 2 m. The peak at the wave travel distance of 2 m indicates the detection of the lithic material 1 m away from the speaker, demonstrating that chert can be detected with acoustics. In the second experiment, the stone was affixed to a sound-neutral rod and then structurally excited by a sharp blow from an impact hammer. A microphone then measured the acoustic waves that radiated from the excited lithic. The calculated acoustic transfer function further refined the peak structural resonance of Coastal Plains chert to 9.572, 13.830, and 23.140 kHz.
Field Testing and Future Plans
In January of 2019, we used an EdgeTech 3100 Sub-Bottom Profiler to scan several submerged prehistoric sites ahead of Phase 2 and 3 testing scheduled for May-July of 2019. Acoustic signatures similar to “haystack” features were observed at all locations, including documented sites both buried and exposed, in both fresh and saltwater. Two of these localities, one believed to represent a buried lithic assemblage and one that likely represents a surface artifact scatter, will be tested this summer to evaluate the reliability and accuracy of this method in the field.
Further lab tests are also underway. More raw material and artifact material types will be tested to determine if other materials (bone, ceramics, etc.) also have peak excitation frequencies. Lab models will also determine what size, shape, and amount of chert are most ideal for excitation, and how environmental variables (depth, salinity, etc.) affect the method. Currently, this method is a presence absence indicator of the material. Additionally, we intend to equip a custom sub-bottom sonar system with multiple hydrophones to triangulate the exact location (depth) of submerged lithic materials.