Abstract:

Long-term exploitation of thermal groundwater in deep-seated aquifers has led to a continual falling in water levels at geothermal wells. Periodic fluctuations in the falling water levels due to exploitation of thermal groundwater in a pumping period followed by a non-pumping period can also be observed. On the other hand, temperature may affect the water level at geothermal wells. The observed water levels can not reflect the realistic water levels of thermal groundwater in deep-seated aquifers. Calibration of water levels is necessary. The calibrated water levels show complex periodic changes and trending changes. In this paper, mathematical models containing the sum of a periodic term and a trending term are established to describe the changes in water levels of thermal groundwater in the Tianjin area. The trending term is fitted by using a linear function, and the periodic term, by Fourier series. The periodic term is analyzed with the spectrum method and determined with the least square method. The fitting is verified with the mean square error of the calibrated and computed water levels and reasonable fitting is obtained. The mathematical models developed can be used to predict the changes in water levels of thermal groundwater in the deep-seated aquifers and good results are also attained.

Key words: thermal groundwater, spectrum analysis, fitting, prediction, Tianjin area