Spatial Optimum Estimation Of The Concentration Variables In The Seepage Contamination Transport Simulation

With the freshwater becoming less and less, how to make use of and to protect the groundwater is calling our attention, for which simulation of water fluid and water quality is essential. Because of its storage and moving character and restricted by the actual condition, numerical method is often adopted in this kind of simulation. Since it was impossible to observe overall spots, the distributing quality in space of unknown spots were usually obtained by interpolation. As there are many methods of estimation in spatial variability, it has become so important a subject in this field to choose a method to get the best result.The study selects some representative methods of estimation in spatial variables as objects, including Inverse Distance Weighting in the traditional space estimation; Global polynomial Interpolation; Ordinary Kriging and Cokriging methods in the geology statistics; Direction Fractal, getting the interpolation in different stylebook densities. Through contrast and analysis ,we can get the precision of simulation and its applicable extension, in order to set the underground water quality model, improve precision of simulation and forecast the underground water quality.The study shows, Kriging is the best, CoKriging improves the simulation effect by considering information provided by cooperative factors, Direction Fractal depicts the spatial variation on seepage water level well as a new method, The Inverse Distance Weighting gets better simulation result in such conditions as low spatial variation, distributing orderly and high stylebook density. In the course of the experiment, the effect of Inverse Distance Weighting power selected 2 is better than selected 1.5.Global polynomial Interpolation can get better simulation effect when the spatial variable changes slowly and presents certain trend, but as the method to estimate total trend, it has lower precision and worse effect into variables with high variation such as seepage pollution on middle and little scale.