Search

The hydraulic connection between aquifers directly determines whether deep coal mining activities affect water quality in shallow aquifers. This study was conducted to investigate the hydrogeochemical characteristics and identify the solute sources of groundwater (including surface water, Quaternary water, water from Cretaceous and Jurassic coal strata aquifer) in the Yuhengbei mining area, Western China, using conventional, multivariate statistical analysis, and dissolved organic matter (DOM) analysis. All water samples were slightly alkaline, and the water from the shallow aquifers (surface water, Quaternary and Cretaceous aquifers) was predominantly of the HCO3-Ca type. The groundwater in deep Jurassic aquifer was mostly of the SO4-Na type. Major ion and DOM (total organic carbon and ultraviolet absorbance) concentrations decreased with increasing aquifer depth. Water-rock interaction and cation exchange were the main mechanisms controlling the chemical constituents of groundwater. The DOM was mainly derived from autogenous biological sources. Based on the indicator values, the hydraulic connection between the deep aquifers and the upper shallow aquifers was generally weak. The knowledge of hydrogeochemical characteristics and formation mechanisms will provide a technical basis for local groundwater management and support the sustainable use of water resources.

To improve the inversion results of roadway or tunnel transient electromagnetic detection data, particle swarm optimisation (PSO) and damped least squares (DLS) algorithms are combined. Thi...

Quantitative determination of the coal seam thickness distribution within the longwall panel is one of the primary works before integrated mining. In-seam seismic (ISS) surveys and interpolations are essential methods for predicting thickness. In this study, a new quantitative method that combines ISS and Bayesian kriging (BK), called ISS–BK, is proposed to determine the thickness distribution. ISS–BK consists of the following six steps. (1) The group velocity of Love waves is plotted by using the simultaneous iterative reconstruction technique under a constant frequency value. (2) An approximate quantitative relationship between the thickness and the group velocity is fitted based on sampling points of the coal seam thickness, which are measured during the process of entry development. (3) The group velocity map is translated into a primary thickness map according to the above-mentioned fitted equation. (4) By subtracting the ISS prediction result from the actual thickness at a sampling point, the residual variable is created. (5) The residual distribution is interpolated within the whole longwall panel by applying BK. The residual map establishes the interconnection between the ISS survey and BK. (6) A refined thickness distribution map can be obtained by overlapping the primary thickness map and the residual map. The application of this method to the No. 2408 longwall panel of Yuhua Coal Mine using ISS–BK showed a considerable improvement in thickness prediction accuracy over ISS. The residuals of ISS and ISS–BK mainly lie in the intervals (− 3.0, 3.0 m) and (− 1.0, 3.0 m), respectively. The accurate prediction rates [where the residual lies in the interval (0, 0.1 m)] of ISS and ISS–BK are 9.39% and 50.28%, respectively, and the effective prediction rates (where the residual is less than 1.0 m) of ISS and ISS–BK are 61.88% and 77.90%, respectively. All the above statistics reflect a considerable improvement in the ISS–BK method over the ISS method.

Exploration and development of the deep part coal resource have important strategic meanings for Chinese economic. However, the seismic of strata in deep part is always suffering from the decline of resolution. To overcome this problem, we propose a method to improve resolution of seismic data by spectrum whitening in time-frequency domain. Compared with conventional algorithm, our method enhances every point of a seismic signal according to the spectrum in a local region. The application results of the real data show our method could broaden bandwidths of seismic data and make strata in deep part more distinct and continuous.