Research On Acoustic Measurement Techniques For Determining The Temperature Distribution Around Seafloor Hydrothermal Vents

The hydrothermal vents are a direct result of the volcanic activity happening under ocean water. Hydrothermal fluid temperatures may be as high as 400℃. Steep temperature fields are formed between the outflowing fluids and the seawater around it. The temperature distributions around the hydrothermal vents in deep-ocean are of fundamental importance in marine environmental monitoring, precise measurement of the heat flux from the hydrothermal vents, submarine hydrothermal mineralization and our understanding of the environmental impacts on the vent organisms.However, in addition to obvious obstacles of high temperature and pressure, the corrosive nature of hydrothermal fluid and high load of particle in the fluid prevent a lot of non-intrusive methods such as infrared, laser and acoustic monitoring from being used to measure temperature of a hydrothermal vent. Placing a matrix of contact-type temperature sensors such as thermocouple sensors and resistance thermocouple sensors near the hydrothermal vents orifice is the most common practice to measure individual 1D profile of temperature around deep-sea hydrothermal vents.This dissertation is supported by the National Natural Science Foundation of China (Project title:Methodology and Technical Study of Long-term In-situ Observation System for Seafloor Hydrothermal. Grant No.:40637037) and the National High Technology Research and Development Program of China (Project title:Development of Acoustic In-situ Detector of Temperature Field and Velocity Field around the Deep-sea Hydrothermal Vents. Grant No.: 2007AA09Z213). By means of experimental observations and data analysis, as well as numerical simulation, the acoustic measurement techniques for determining the temperature distribution around seafloor hydrothermal vents are studied in this dissertation. The main contributions of this work are as follows:1. Ocean is a complicated and changeable channel which may distort the signals and degrade the performance of acoustic measurement system. The attenuation rules in suspended particles and bubble group are investigated numerically. The sound generated by the Kueishantao Island's hydrothermal activity in northeast Taiwan and the ambient noise are measured by the author. The acoustic spectrum of the recorded signals shows that vents radiate significant acoustic energy at all frequencies up to 3 kHz. The vents generate a broadband acoustic signal with power levels 10～30 dB above the ambient noise level. In this thesis we propose optimal frequencies for the acoustic temperature field measurement system around deep-sea hydrothermal vents. Higher frequencies are immune to most natural and artificial sound sources encountered during in-situ measurement. However, from an attenuation and cost perspective, lower frequency operation is more desirable. So the choice of operating frequency in acoustic in-situ detector has to be made by compromise. We choose working frequency band of 18-25 kHz for a given acoustic in-situ detection system. 2. We propose a high resolution multipath time-delay estimation scheme based on the active sound souce cross-correlation (ASSCC). This method uses a known reference waveform that neglects the negative influence caused by noise and has superior performance on time-delay estimation. We study the Cramer-Rao bound of the multipath time-delay estimation scheme, and give the variance of estimation. We also present two novel algorithms to automatically determine the center-to-center distances of underwater transducers in our system for in-situ measurements of temperature distribution around deep-sea hydrothermal vents.3. We have made a scrutiny into the reconstruction algorithm. Two novel 2-D temperature field reconstruction methods are proposed, which are based on the total least squares (TLS) and the robust least square regression (RLS) respectively. It is shown that in the presence of observation noises, comparing with the results obtained from a conventional least squares approach, a TLS solution leads to significant improvements in the quality of reconstructed images. To deal with the gross errors in measurement, we make use of the RLS method, which is based on the Huber estimate and is computed by means of the iteratively reweighted least squares algorithm. It is shown that even a single outlier can degrade a least squares solution considerably. The robust estimators, thanks to an iterative downweighting process, gradually ignore those outliers that lead to large residuals.An analysis is performed on the relevant factors affecting reconstruction quality. As our results indicate, the quality of temperature field not only depends on such parameters as the quantity and placement of the transducers, but also depends on the the area of the measurement plane, the number of grids, the accuracy of the time-of-flight measurements, gross errors and other factors.4. A new acoustic measurement system of temperature distribution around seafloor hydrothermal vents has been developed. Regarding the measuring system, hydrophones are used as transmitters and receivers. The signal correlation analysis is performed to obtain accurate acoustic wave transit time from the transmitted and received acoustic signals. Both the transmitter emission start time and the receiver capture start time are also estimated using a novel algorithm based on the time-of-flight measurements at different water temperatures. The reconstruction of the 2-D temperature field which is an ill-posed problem, is conducted by total least-squares method and the robust least square regression.5. The laboratory experiments and lake trial have been carried out to study the integrated performance of the acoustic measurement system of temperature distribution around seafloor hydrothermal vents. First we perform the system calibration, and get the response time and the center-to-center distance between a transmitter and a receiver. Next, we confirm the feasibility of the acoustic measurement techniques for determining the temperature distribution around seafloor hydrothermal vents by a tank test. Then, an experimental study is performed on acoustic imaging of underwater temperature fields in Lake Qiezishan, located in Longling Country, Yunnan Province, China. There are hot springs and craters beneath Lake Qiezishan, which lies at 24°32'33" north latitude and 98°47'44" east longitude. Experimental results confirm that acoustic tomography is a powerful tool for studying a small scale temperature field around the hydrothermal vents in seafloor, and stands as a good candidate to replace intrusive methods. The maximum relative error between the data measured by the sound probes and that measured by the thermocouples is within 1.11%.Surely, as a new and primitive studying method, it needs further improvement and perfection yet. In our opinion, a significant improvement in temperature field reconstruction can only be obtained if high precision measurement of TOF data, a low condition number of matrix A, and an effective inversion method are available. The underwater acoustic measurement system has the advantage that it can also be used in applications such as water quality monitoring, nuclear, chemical engineering and agriculture.