The conception of Mesher is used in the finite-difference (FD) method. Each mesher stores the discretization of one dimension. It has a array \(locations\) that stores the discretization at points \(x_{0},x_{1},..,x_{n-1}\). It also has two other arrays \(dplus\) and \(dminus\) that store the i-th elements \((x_{i+1}-x_{i})\) and \((x_{i}-x_{i-1})\) respectively.

The multi-dimensional mesh for a finite-difference model is represented by multi 1-D meshers, which build the full mesh by composing a 1-D mesh for every dimension.


Mesher creates an equally-spaced grid between two given boundaries and with the given number of points with a specific processor. 1. locations: The coordinates of location are evolved based a normal random number. 2. dplus: The argument is calculated at point \((x_{i+1}-x_{i})\). 3. dminus: The argument is calculated at point \((x_{i}-x_{i-1})\).