| Surface shallow unsaturated soil moisture content(SMC)measurement is an important goal in many research fields.The detection and monitoring of soil moisture content plays a key role in civil engineering(embankment,dam,geotexture,etc.),landslide disaster,agricultural irrigation and soil erosion,restoration of contaminated or damaged land,environment and ecology.Soil moisture content affects almost every aspect of human production and life.Clay is widely distributed in China,and the temporal and spatial variation of its water content is one of the important factors that cause geological disasters,affect agricultural irrigation and soil erosion.GPR is a suitable method for measuring water content in medium and small scale,but most of the existing methods are based on GPR time domain signal,and the calculation is complicated.Therefore,it is of great significance to explore an accurate and rapid soil moisture content detection method based on GPR frequency analysis,which is suitable for the meso-scale soil moisture content detection in clay areas.In this paper,the forward modeling model of different water content and two physical experiment models were established to find out the relationship between the frequency offset of the echo peak and the change of water content,and the relationship model of inversion of soil water content by frequency shift method was established to realize accurate and rapid detection of soil water content in clay areas.The specific research is as follows:(1)In view of the difficulty in extracting the peak frequency of low-resolution spectrum,four extraction methods including direct reading are studied.The results of comparison and measurement show that the SINC interpolation method has the highest accuracy,followed by the parabolic method and direct reading method,and the triangle method has the lowest accuracy.(2)Through forward modeling experiments,the influence of the electromagnetic characteristics and water content of the clay with the center frequency of 100 MHz,400MHz and 900 MHz on the peak frequency of the GPR electromagnetic wave is studied.The results show that the peak frequency of GPR decreases with the increase of the permittivity,and the decreasing amplitude is more and more large.The peak frequency of high frequency antenna is more obvious than that of low frequency antenna.The peak frequency of GPR decreases with the increase of electrical conductivity.With the increase of water content,the peak frequency is shifted to low frequency and the offset antenna frequency is increased.(3)The influence of the variation of clay moisture content in two different areas with the central frequency of 100 MHz,400MHz and 900 MHz antenna on the peak frequency of GPR was studied,and the relationship model of the inversion of soil moisture content in Guiyang red clay area and Xinjiang Hongshaquan mining area by frequency shift method was established.The results show that the relationship between the water content of Guiyang red clay model and peak frequency is a quadratic curve,and the relationship between the water content of Hongshaquan clay and peak frequency is a linear function.Among the three antennas,the fitting correlation coefficient of the 900 MHz antenna is the highest,followed by the 400 MHz antenna and the 100 MHz antenna.Therefore,when the peak frequency was extracted with 900 MHz antenna and SINC interpolation method,the correlation coefficient was the highest,and the experimental model of Guiyang red clay and Hongshaquan clay were 0.9332 and0.9122,respectively.When 100 MHz antenna and triangle method were used to extract peak frequency,the correlation coefficients were the lowest,which were 0.5737 and0.4872 respectively for the two models.(4)In the field measurement of geophysical exploration,when SINC interpolation method is used to extract peak frequency,the errors of 100 MHz,400MHz and 900 MHz antennas are 7.42%,4.32% and 3.37% respectively;The errors of peak frequency extracted by parabolic method were 9.07%,7.78% and 5.24%,respectively.The errors of peak frequency extracted by triangle method were 16.32%,15.24% and 10.21%,respectively.The errors of direct reading peak frequency were 9.86%,8.03% and 5.67%,respectively.In the Hongshaquan clay area,the SINC interpolation method was used to extract the peak frequency,and the errors of 100 MHz,400MHz and 900 MHz antennas were 9.47%,8.07% and 4.41%,respectively.The errors of peak frequency extracted by parabola method were 11.37%,9.26% and 7.10% respectively.The errors of peak frequency extracted by triangle method were 18.28%,14.66% and 13.67%,respectively.The errors of direct reading peak frequency were 12.38%,10.02% and8.24%,respectively. |
PDF Full Text Request