Water Flow - Boundary Conditions

The HP1 code incorporates modules simulating
(1) transient water flow in variably-saturated media,
(2) transport of multiple components, and
(3) mixed equilibrium/kinetic geochemical reactions.
Post Reply
Ecaterina
Posts: 5
Joined: Mon Feb 01, 2010 9:20 am
Location: Netherlands

Water Flow - Boundary Conditions

Post by Ecaterina » Thu Feb 04, 2010 3:17 pm

Dear Sir,

I would like to ask why after I choose no water flow in the Main Process window, I still get the window “Water flow- boundary conditions”? Do they matter in this case?

Greetings,

Ecaterina

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Post by Jirka » Thu Feb 04, 2010 3:19 pm

Ecaterina,

No they don't. Simply select constant head on both ends of the domain.

Jirka

Ecaterina
Posts: 5
Joined: Mon Feb 01, 2010 9:20 am
Location: Netherlands

Post by Ecaterina » Thu Feb 11, 2010 3:58 pm

Dear Sir,

Thank you very much for answering my questions. In the meantime, I have another one :-)


While simulating solute transport using HP1 I have observed that when I have changed the geometry (depth from 3000 to 5000 centimeters) my calculation went wrong. I had actually no solute at all on the entire domain. I have adjusted all the data, also in the graphical editor (observation nodes, initial conditions) and still, zero concentrations in all observation points, inclusive the nodes where I have specified a certain initial solution composition. If I am building the model from scratch, it is working, no problem at all.
Do you maybe have an idea about what I am doing wrong? It is quite time-demanding to build a different model every time I want to change the depth of the layer :-(


Sincerely,

Ecaterina

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Post by Jirka » Thu Feb 11, 2010 4:15 pm

Ecaterina,

It certainly should not matter whether you develop a project from scratch or whether you modify an existing project. Just the opposity should be true. When you modify an existing project, there is less chance for making mistake. The idea behind distributing several projects and posting many more on website is that you could use these projects and modify them.

You need to make sure that once you change the depth and generate new spatial distribution, that you assign to each node initial solution compositions. Note that once you change the depth and/or discretization, all spatially variable parameters (material distribution, initial conditions, root distribution) need to be specify again.

Jirka

Ecaterina
Posts: 5
Joined: Mon Feb 01, 2010 9:20 am
Location: Netherlands

Post by Ecaterina » Thu Feb 18, 2010 3:51 pm

Dear Jirka,

Thank you very much for jour replay. I will try to be more careful when specifying the spatial parameters, probably I have missed one.
However, today I have encountered another problem. While simulating solute transport and adsorption in a 2 materials profile with HYDRUS 1D only I got the following error message “root must be bracketed for ZBRENT1”. Could you tell me what it means and where I should look for mistake?
Sincerely,

Ecaterina

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Post by Jirka » Thu Feb 18, 2010 5:02 pm

You must be doing something wrong with the dual-porosity model of Durner.

I hope that I do not have to remind you than rather than checking every possible option available in the model and then wondering where the problem is when the program does not run, you should start from the simple model and build it up in steps, especially if you do not have much experience with the model yet.

Jirka

Ecaterina
Posts: 5
Joined: Mon Feb 01, 2010 9:20 am
Location: Netherlands

Post by Ecaterina » Tue Feb 23, 2010 1:28 pm

Dear Jirka,

I use the van Genuchten-Mualem model, single porosity. I did start with a simple model which I tried to complicate a bit, just like you have advised. I had just solute transport and decay in my first model and I have tried to add adsorption. I have two layers. When I have beta = 1 for both layers, it runs, no error message, and when I try to give beta another value (0,5) for one of the layers, I get the above mentioned error.


Best regards,
Ecaterina

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Post by Jirka » Tue Feb 23, 2010 4:58 pm

Ecaterina,

Note that you posted your question at the HP1 site. In the HP1 program, you can not actually define nonlinear sorption in the HYDRUS itself, but you need to do that in PHREEQC. Anyway, now I understand that you are not actually using HP1, but a regular HYDRUS-1D (with General Solute Transport), right?

In regular HYDRUS, ZBRENT routine (for root finding of a nonlinear equation) is used in two places:
a) Dual-porosity model of Durner
b) For distributing initial solute mass into different phases (solid, liquid, and gaseous, if present).

You are likely using the b option. There should be no problem. I have tested it right now, having linear and nonlinear (with beta=0.5) sorption in different layers, and the program ran just fine. If you do encounter still problem, why don't you make your own calculations what the liquic concentration should be and specify that as initial condition?

Jirka

Post Reply