| The Hongshaquan mining area is located in a semi-arid region with abundant coal resources but limited water resources.Therefore,finding an accurate method to detect the distribution of soil water in the waste dumping sites is helpful for the land reclamation and ecological restoration of the mining area.Ground-penetrating radar（GPR）technology can provide large-scale in-situ water content distribution data without damaging the soil,and it has been widely used in soil water content detection.Among the existing methods for measuring soil water content using GPR,the Average Envelope Amplitude（AEA）method can directly establish a relationship model between water content and the reciprocal of the average envelope amplitude of the radar wave’s early signal,thus avoiding the limitation of difficult detection due to the lack of obvious reflection interfaces.During the ecological restoration of the waste dumping site,the AEA method was used to detect the water content distribution.However,the results showed that there were differences in soil water content along the GPR profile.Therefore,for the specific conditions of the study area,it is of practical significance to study the influence of soil physical properties and layered soil structure on the early signal amplitude of the radar wave,and clarify the influence of various factors on the early signal amplitude in the AEA method for inverting soil water content.This study obtained field data on the physical properties and layered soil structure parameters of the spoil disposal site.A numerical model was established based on these parameters to investigate the influence of various physical properties and layered soil structure on the early signal amplitude through numerical simulation experiments.The correlations between the influencing factors and the early signal amplitude were determined.The AEA-1 value was corrected using significant factors,and a revised relationship model between the AEA-1 value and moisture content was established.The specific research is as follows:（1）Firstly,in the areas where abnormal soil moisture content was detected using the AEA method,the horizontal and vertical variations of various physical properties in the spoil disposal site were analyzed.The soil bulk density exhibited low variation,while soil particle content and salt content showed significant horizontal variation,with moderate variation in moisture content.（2）Bulk density and sand content exhibited a positive correlation with the AEA-1value,while silt content showed a negative correlation.Clay content was not correlated with the AEA-1 value,and the AEA-1 value exhibited a negative correlation with moisture content under different salt content conditions.When the surface undulation depth was less than 3cm,the AEA-1 value increased gradually with increasing undulation depth.However,when the undulation depth exceeded 3cm,the AEA-1 value started to decrease.As the field surface undulation depth in the spoil disposal site was within 6-9cm,the impact of surface undulation on the AEA method’s application in soil moisture detection in the spoil disposal site was relatively weak.Similar results were observed for homogeneous and heterogeneous media in the upper and lower layers.The AEA-1 value reached its minimum when the surface layer thickness was 17-18cm,indicating an effective detection depth of the AEA method within the range of 17-18cm.As the upper soil thickness in the spoil disposal site ranged from 40-60cm,the different lower-layer soil media did not affect the AEA-1 value.（3）Under the combined influence of various factors,the impact of different soil physical properties on the AEA-1 value was ranked as follows:salt content>bulk density>sand content>silt content.The correlation between the AEA-1 value and soil sand content was the highest,followed by soil bulk density and salt content.The soil moisture prediction model established using the AEA-1 value corrected by the aforementioned influencing factors exhibited an overall error of less than 1%when applied in the engineering demonstration area.Furthermore,the computed moisture distribution curve appeared smoother compared to the previous calculations. |
