PC-Progress Discussion Forums

I have since found many examples of the use of 3DADE to horizontal GW transport. Including its implementation in the SOLUTRANS program. So I guess that I've answered my own question...yes, 3DADE can be used to simulate 1D horizontal groundwater transport with 3D dispersion.

Mike

Hi All, I have been testing 3DADE for simple 1D groundwater transport of chloride. The way that the source dimensions are set out in the program are for vertical transport in a one dimensional flow field. Is there any conceptual problem with rotating the frame of reference so that the flow is modele...

I'd like to reopen the discussion on this topic. Is it possible that the Hydrus code could be modified to apportion excess water at the atmospheric boundary to runoff when the soil profile becomes saturated? I have many cases where I am examining shallow water tables 0.5 to 1 m depth (fixed pressure...

Hi Bruno, I do alot of batch runs with Hydrus. I needed to create three files. The first was a batch file, which I simply called run.bat. The run.bat file contains the following text: copy Path1.txt level_01.dir H1d_calc<return.txt copy Path2.txt level_01.dir H1d_calc<return.txt ...etc. ,where Path1...

Thanks Jirka, Enjoy your vacation! It would be be a great aid to my work if you were able to reprogram Hydrus to terminate on non-convergence. I'll contact you by e-mail at the beginning of September for an update. I was also wondering if it would be an easy task to increase the allowable number of ...

A1. What kind of variable boundary condition do you want on the bottom boundary? I have sucessfully used a variable pressure head boundary condition to represent a fluctuating water table. It that case it is important to make sure that the function used to define the boundary pressure head is smooth...

Hi Everyone, Does anyone have experience programming batch runs of Hydrus 1D so that the simulation continues past a model locked in a non-convergence loop? My problem is that everytime the model enters non-convergence in a run, the batch run gets stuck in a loop that requires manual stoppage. This ...

There are so many variations in model setup, that really all I can do is offer broad suggestions. The top two problems for convergence that I have encountered are the 1) the runoff BC, and 2) high intensity rainfall. Both seem to be related to saturating the upper BC. The model does not seem to like...

It's been awhile since your response, but I wanted to add that your advice was very useful. Many thanks. FYI for others: The use of activities in the "Gapon" type equation (which normally uses concentrations), has been referred to as being used to calculate the Vanselow coefficient. The paper "Sodiu...

It could be one of many problems. So I can only give some suggestions at this point. 1) make sure your minimum time step is small enough e.g. 1E-5 days 2) try a node spacing as small as < 1cm 3) turn off interpolation tables by setting upper and lower limit of tension interval to zero (in the Iterat...

Many thanks for the quick reply, it was very helpful. My checks on the mass balance for solute concentrations check out fine with the addition of the factor of 100. For adsorbed species, the mass units in CHEMBAL.OUT don't match up with my hand calculations. It is best illustrated with the following...

When you say "the model continues to run", what do you mean by that? Are you getting output to the screen in the DOS window? If you are getting output, is the model converging, or are you getting a convergence error message? Or is the model stopping at 1 sec. of runtime? More details would be helpful.

Jirka, I have a few questions about the mass units used in the UNSATCHEM module of HYDRUS-1D. The mass and length units I have specified in my simulation are mmol and cm. In the UNSATCHEM manual the solute concentration units are listed as mmol/l and for adsorbed species as mmol/kg (in Table 12.8) e...

Likely, your input files are buried too deep in the file structure. There is a limit to the length of the file pathway. Beyond that limit, Hydrus can't read the filename.

I would suggest putting your input files into a directory right off your C: root. For example C:\Hydrus\....

Mike

I've seen the non-convergence problem before. Are you getting through a partial run before a crash or is it crashing at the start. 1. Problem could be with your time step, node spacing or iteration criteria. Try a time step of 1E-5 days, node spacing of 10 mm, and max iterations of say 200. Set the ...