Distributed velocity method of solving the convective-dispersion equation: 2. Error analysis and comparison with other methods

User`s manual for the NEFTRAN II computer code

“…In addition, large time steps also promote accuracy by limiting numerical dispersion. This is evident from the Courant number criterion above and is discussed in detail by Campbell and Longsine (1981). The apparent excessive attention to accuracy that is evident in the velocity model (Section 2.4.6), the discharge model (Section 2.4.7), the distribution of source into grid blocks (Section 2.4.8), and the legto-leg transfer model (Section 2.4.9) is required because of the large time steps that are used to limit numerical dispersion.…”

“…However, with DVM, numerical accuracy is generally improved by increasing the size of the time step. The reasons for this counter intuitive situation are explained by Campbell and Longsine (1981) The general rule concerning the CN is that it should be greater than or equal to one for all important isotopes. As discussed in previous sections, an important isotope is one that is expected to discharge from its initial inventory during the problem simulation time.…”

“…This criterion is based on the Courant number and is concerned with the numerical accuracy of NEFTRAN 11. As discussed by Campbell and Longsine (1981) In general, to minimize numerical dispersion for the DVM transport model, CN should be greater than or equal to 1.0 for all important isotopes.…”