One-dimensional equilibrium model and source parameter determination for soil-column experiment (Q2383873): Difference between revisions

The authors present one-dimensional equilibrium soil-columns experiment models with source (sink) reaction terms. Basing on the constructed model in 1-dimensional space they consider soil-column experiments. Then they apply an optimal perturbation algorithm for determining the source coefficient with the limited additional data. They show that their approach could be more efficient than regularization strategies, which are also used in their paper. They present some pictures of computational results for \(\beta(T)\) -- the source coefficient functions reflecting some chemical physical recations in the soil-column. More precisely, they consider the following mathematical model for soil-column experiment \[ R{\partial C\over\partial T}={1\over P}{\partial^2C\over\partial Z^2}- {\partial C\over\partial Z}+\beta(T)Z,\quad 0<Z<1,\;\;0<T<\overline T \tag \(*\) \] with initial-boundary value conditions: \(C(Z,0)=0\), \(C(0,T)=1\), \(C_Z=(1,T)=0\), where \(C=c/c_0\), \(Z=x/L\), \(T=vt/l\), \(P=lv/D\), \(\overline T=vT_1/l\) \(c\) -- denotes the solute concentration in the liquid phase, \(c(0,x)=c_0\) -- initial infiltrating concentration, \(v\) -- steady flow velocity, \(D\) -- dispersion coefficient, \(l\) -- length of the column, \(a_L\) -- dispersivity \((D=a_L\times v)\), \(T_1\) -- total infiltration time with polluted water. Basing on the \(BTCs\) data of solute they carry out numerical computations to determine the source coefficient of \(\beta(T)\) based on the model \((*)\) applied to an optimal perturbation algorithm with Matlab tools. It should be emphasized that the paper presents some initial investigation on the equilibrium model and the inverse model determining source dependency for one-dimensional soil-column experiments. A real problem needs more investigation (particulary in the nonequilibrium situation).