convergence problem

A discussion forum for Hydrus-1D users.
Post Reply
User1990
Posts: 1
Joined: Tue Apr 04, 2017 8:39 am
Location: Netherlands

convergence problem

Post by User1990 » Tue Apr 04, 2017 9:35 am

Hello everyone,
it's a week that I try to fix my model without results.
I'm trying to implement a simple water flow model but it doesn't converge.

That are the data I use:
- Simulate: water flow
- Lenght Units: meter; 1, 1, 0.4444, 2 (number of soil, layers, decline, depth);
- Time Units: Hours; Time discretization: 0, 528, 0.001, 0.00024, 1; I check "Time- variable boundary conditions and I set on 528;
- I check:
T-level every ntime steps: 1;
Screen Output;
Print fluxes;
Hit enter at end?
Print times:100 with default setting;
- Iteration Criteria: 20, 0.001, 0.01, 4, 7, 1.3, 0.5, 1e-008, 100;
- Hydraulic Model: van Genuchten-Mualem with air entry value of -2cm; no hysteresis;
- water flow parameters: 0.045 0.35 4 2 0.072 0.5;
- water flow boundary conditions:
Upper Atmospheric BC with Surface Layer(-0.1 max h at soil surface);
Bottom Deep Drainage
Initial condition: in water contents;
- Deep drainage parameters: A=0.001376 m/hour
B=2.3402 1/m
Reference ground water= 0m;
- Time variable boundary conditions:
Time: 1-528
Precipitation in the order of 0.0005 (from 0 to 0.003 m/hour)
Evaporation= 0
hCritA=-100 m
- In the graphic profile informations I have set 400 nodes, initial conditions in water content with a linear increment from 0.045 at the top to 0.35 at the bottom;
- Soil profile summary: z is from 0 to 2 m
theta=Dxz from 0.045 to 0.35
root 0
Axz=Bxz=Mat= 1

After several iterations it says "Numerical solution has been stopped since it has not converged in 10 consecutive time steps"

Please if you are not too bored by my description of the problem, tell me what do you think could be the issue.

Thanks in advance

Claudio

Post Reply