The DFNs currently define default fill values for some array variables (also some scalar dimension sizes) for which questions of appropriateness have been raised. E.g.
- GWF IC
strt = 1.0
- GWF DIS
nlay = 1, nrow/ncol = 2, delc/delr = 1.0, botm = 0.0
- NPF
k = 1.0
If I recall correctly, the thinking was, these make it slightly easier to create simple models, by removing the need to specify values, but there is no reason for any particular value.
Do we want default fill values for array variables that have no obvious default values? Is ease of creating a simulation sufficient reason?
Others though are justifiable on a physical basis, like
- GWE EST
density_water = 1000.0, heat_capacity_water = 4184.0
Then there is the widespread use of DNODATA (3e30) which may be a separate question.
The DFNs currently define default fill values for some array variables (also some scalar dimension sizes) for which questions of appropriateness have been raised. E.g.
strt = 1.0nlay = 1,nrow/ncol = 2,delc/delr = 1.0,botm = 0.0k = 1.0If I recall correctly, the thinking was, these make it slightly easier to create simple models, by removing the need to specify values, but there is no reason for any particular value.
Do we want default fill values for array variables that have no obvious default values? Is ease of creating a simulation sufficient reason?
Others though are justifiable on a physical basis, like
density_water = 1000.0,heat_capacity_water = 4184.0Then there is the widespread use of
DNODATA(3e30) which may be a separate question.