| Popis: |
This thesis presents the analysis of a series of experimental measurements from a selection of springs, geysers and wells at Te Aroha, Rotorua and Orakeikorako in New Zealand aimed at the identification of the frequencies of any characteristic variations with a period of 25 hours or less. The measurements taken from these sources collectively recorded water level, conductivity, temperature, rainfall and barometric pressure. The measurements involve the collection of data with high accuracy instrumentation and high sampling rates of 5 minutes to 1 hour over periods exceeding 1400 hours in both exploited and unexploited hydrothermal systems. The measurements are filtered using both an adaptive wavelet based filter and Fourier analysis to extract the influence of gravity tides, rainfall, and barometric pressure with the residual dominant frequencies being attributed to sub-surface thermodynamic processes. A review of the development of techniques for the determination of the pressure response of geothermal reservoirs to barometric pressure, gravity tides, and rainfall is presented. The fundamentals for the mathematical development of Fourier and wavelet analysis are discussed along with sources of error for both Fourier and wavelet analysis. A sensitivity study of wavelet and Fourier analysis is presented based on the analysis of a series of artificial signals. A system dynamics model of a spring at Orakeikorako is formulated to demonstrate fundamental heat and mass transfer processes and the effect of periodic forcing functions on spring response. This study is of importance in that it provides an understanding of periodic forcing functions derived from high accuracy measurements on geothermal features at high sampling rates over periods exceeding 1400 hours. Once identified, the amplitude and frequency of the forcing functions can be monitored to assist in identifying and quantifying any changes in reservoir characteristics. |
