Using neural network modeling to improve the detection accuracy of land subsidence due to groundwater withdrawal

Despite the high efficiency of remote sensing methods for rapid and large-scale detection of subsidence phenomena,this technique has limitations such as atmospheric impact and temporal and spatial decorrelation that affect the accuracy of the results.This paper proposes a method based on an artificial neural network to improve the results of monitoring land subsidence due to groundwater overexploitation by radar interferometry in the Aliabad plain(Central Iran).In this regard,vertical ground deformations were monitored over 18 months using the Sentinel-1A SAR images.To model the land subsidence by a multilayer perceptron(MLP)artificial neural network,four parameters,including groundwater level,alluvial thickness,elastic modulus,and transmissivity have been applied.The model's generalizability was assessed using data derived for 144 days.According to the results,the neural network estimates the land subsidence at each ground point with an accuracy of 6.8 mm.A comparison between the predicted and actual values indicated a significant agreement.The MLP model can be used to improve the results of subsidence detection in the study area or other areas with similar characteristics.

Depreciation factor equation to evaluate the economic losses from ground failure due to subsidence related to groundwater withdrawal

Subsidence due to groundwater withdrawal is a complex hydrogeological process affecting numerous cities settled on top of fluviolacustrine deposits. The discrete spatial variation in the thickness of these deposits, in combination with subsidence due to groundwater withdrawal, generates differential settlements and aseismic ground failure (AGF) characterized by a welldefined scarp. In cities, such AGF causes severe damages to urban infrastructure and considerable economic impact. With the goal of arriving to a general criterion for evaluating the economic losses derived from AGF, in the present work we propose the following equation: ELi = PVi*DFi. Where PVi is the value of a property “i”, and DFi is a depreciation factor caused by structural damages of a property “i” due to AGF. The DFi is calculated empirically through: . This last equation is based on the spatial relations of coexistence and proximity of property polygons and the AGF axis. The coexistence is valued as the quotient of the affectation area divided by the total area of the involved property;and the proximity to the AGF axis is expressed as the inverse of the perpendicular distance from the centroid of the property polygon to the AGF axis. The sum of these terms is divided by two to determine the percentage that affects the property value (PVi). These equations are relevant because it is the first indicator designed for the discrete assessment of the economic impacts due to AGF, and can be applied to real estate infrastructure from either urban or rural areas.

Study on the Estimation of Groundwater Withdrawals Based on Groundwater Flow Modeling and Its Application in the North China Plain

The amount of water withdrawn by wells is one of the quantitative variables that can be applied to estimate groundwater resources and further evaluate the human influence on groundwater systems. The accuracy for the calculation of the amount of water withdrawal significantly influences the regional groundwater resource evaluation and management. However, the decentralized groundwater pumping, inefficient management, measurement errors and uncertainties have resulted in considerable errors in the groundwater withdrawal estimation. In this study, to improve the estimation of the groundwater withdrawal, an innovative approach was proposed using an inversion method based on a regional groundwater flow numerical model, and this method was then applied in the North China Plain. The principle of the method was matching the simulated water levels with the observation ones by adjusting the amount of groundwater withdrawal. In addition, uncertainty analysis of hydraulic conductivity and specific yield for the estimation of the groundwater withdrawal was conducted. By using the proposed inversion method, the estimated annual average groundwater withdrawal was approximately 24.92×10^9 m^3 in the North China Plain from 2002 to 2008. The inversion method also significantly improved the simulation results for both hydrograph and the flow field. Results of the uncertainty analysis showed that the hydraulic conductivity was more sensitive to the inversion results than the specific yield.