Page 1 of 1

About spatial discretization

Posted: Tue Apr 16, 2019 7:32 am
by hhfang
Dear all,
I am testing different spatial discretization to save computation time for MC. I have tested
different setups of nodes so that I can use the least nodes but still get similar profile of soil

All the setups are listed here, but I know n73 and n61 can be problematic because the great
change of the sizes between the neighboring elements. So based on the note of the manual
(The ratio of the sizes of two neighboring elements is not recommended to exceed about 1.5.),
I introduced the transition part between the smaller grid (0.5 cm) and the large one (6 cm).
I have bottom free drainage BC, and the top BC is the daily precipitation with hCritA 100000 and
hCritS 0. No root uptake was considered.

I suppose the profile of n201 is the correct one and compare the others to it. I want to obtain
a profile with as less nodes as possible, but it should look similar to the one of n201. I thought
it is a simple setup, so the profile of n97 and n60 (with transition grid) should be similar to the
profile of n201. But it is not the case... Does it make sense? Or did I miss something?
123.png (30.8 KiB) Viewed 685 times
n101 1 cm (depth 0 ~-100 cm)
n201 0.5 cm ( 0 ~-100 cm )
n41 2.5 cm ( 0 ~-100 cm )
n73 0.5 cm ( 0 ~-20 cm ) ; 2.5 cm ( -20 ~-100 cm)
n61 0.5 cm ( 0 ~-20 cm ) ; 4.5 cm ( -20 ~-100 cm)
n60 0.5 cm ( 0 ~-20 cm ) ; transition ( -20~ -44 cm) ; 6 cm (-44 ~-100 cm)
n97 0.5 cm ( 0 ~-40 cm ) ; transition ( -40~ -64 cm) ; 6 cm (-64 ~-100 cm)

All comments are welcome. Thanks in advance for your time.

Re: About spatial discretization

Posted: Tue Apr 16, 2019 3:10 pm
by Jirka
I suggest that you look at the following webpage: ... filtration. Here I did the sensitivity analysis of simulation results for infiltration as a function of soil texture, initial saturation, spatial discretization, etc. J.

Re: About spatial discretization

Posted: Thu Apr 18, 2019 7:59 am
by hhfang
Thank you very much!