Add a modified infiltration scheme in ELM to improve surface water dynamics #5974

donghuix · 2023-10-09T17:17:03Z

In default, the infiltration rate is the same for the non-water surface and surface water storage within the grid cell. This assumption results in a significant underestimation of surface water because the infiltration in the surface water storage is not constrained.

In this PR, we introduced a modified infiltration scheme to constrain the infiltration in the surface water storage. The infiltration rate in the surface water storage is smaller than the infiltration rate in the non-water surface. More details about this new scheme and corresponding evaluations can be found in the preprint: https://doi.org/10.21203/rs.3.rs-2733749/v1 The manuscript is still under review.

The new infiltration scheme can be turned by adding use_modified_infil = .true. in user_nl_elm.

We passed all the tests in e3sm_land_developer test suite.

We also created a test for the simulation of surface water dynamics with the new infiltration scheme and calibrated parameter for NLDAS domain: surface_water_dynamics. This test was added into e3sm_land_developer test suite.

[BFB]

…torage

…s for NLDAS domain

bishtgautam

@donghuix I have a few minor comments for you to address.

bishtgautam · 2023-10-10T20:51:43Z

components/elm/src/biogeophys/SoilHydrologyType.F90

+    call ncd_pio_openfile (ncid, locfn, 0)
+    call ncd_io(ncid=ncid, varname='pc', flag='read', data=this%pc, dim1name=grlnd, readvar=readvar)
+    if (.not. readvar) then
+      this%pc(:) = 0.4


Change 0.4 to 0.4_r8

Using 0.4_r8 will result in non BFB to the master branch. That is because pc is set to 0.4 in the master (Please see next reply).

In that case, leave it as is. Thanks.

components/elm/src/external_models/emi/src/elm_stub/utils/clm_varcon.F90

components/elm/src/main/elm_varcon.F90

components/elm/src/biogeophys/SoilHydrologyMod.F90

peterdschwartz · 2023-10-11T14:54:59Z

Are all the input files for the test on the input data server? Will use_modified_infil = .true. work without needing calibrated parameters? If not, I would suggest making a compset or use case to take care of all the dependencies.

Also, urls can't go into the git log, so do you have a citation that could be used instead?

donghuix · 2023-10-16T17:10:24Z

@bishtgautam I addressed your comments in the new commit. However, I cannot add ._r8 for the assignment of pc, which will result in differences compared to baseline simulation.

@peterdschwartz Yes, @bishtgautam has uploaded a calibrated ELM surface dataset to the input data server for the test case. @bishtgautam Can you confirm it?

use_modified_infil = .true. doesn't need calibrated parameter, and it works with any ELM surface dataset as default parameter values will be used.

I changed the urls to citation in the log. Please let me know if additional modifications are needed.

Thanks!

peterdschwartz · 2023-10-16T20:49:23Z

I changed the urls to citation in the log. Please let me know if additional modifications are needed.

Thanks!

@donghuix I should have been more precise. When I merge your PR, I'll copy/paste your PR description at the top of this page as the merge commit message. So, I just need this researchgate url replaced with the citation

In this PR, we introduced a modified infiltration scheme to constrain the infiltration in the surface water storage. The infiltration rate in the surface water storage is smaller than the infiltration rate in the non-water surface. More details about this new scheme and corresponding evaluations can be found in the preprint: https://www.researchsquare.com/article/rs-2733749/v1. The manuscript is still under review.

Thanks

rljacob · 2023-10-16T21:00:30Z

You can use the DOI https://doi.org/10.21203/rs.3.rs-2733749/v1

peterdschwartz · 2023-11-07T19:41:50Z

@peterdschwartz Yes, @bishtgautam has uploaded a calibrated ELM surface dataset to the input data server for the test case. @bishtgautam Can you confirm it?

Tried to run the new test on chrysalis and the input file is not available for download..... @donghuix @bishtgautam

bishtgautam · 2023-11-07T20:09:37Z

The surfdata_nldas2_simyr2000_c181207_surfacewater.nc is now available on https://web.lcrc.anl.gov/public/e3sm/inputdata/lnd/clm2/surfdata_map/

peterdschwartz · 2023-11-08T00:01:29Z

Tests show DIFFs in these tests:

 SMS_Ln9.ne4_oQU240.F2010.chrysalis_intel.eam-outfrq9s (Overall: DIFF) details:
  SMS_Ly2_P1x1.1x1_smallvilleIA.IELMCNCROP.chrysalis_intel.elm-force_netcdf_pio (Overall: DIFF) details:
  SMS_Ly2_P1x1.1x1_smallvilleIA.IELMCNCROP.chrysalis_intel.elm-per_crop (Overall: DIFF) details:
  SMS_Ly2_P1x1_D.1x1_smallvilleIA.IELMCNCROP.chrysalis_intel.elm-lulcc_sville (Overall: DIFF) details

Appear to be only due to the new history field added. Could put an if statement around the new history field addition with the new namelist flag ?

donghuix · 2023-11-08T05:11:57Z

Tests show DIFFs in these tests:
 SMS_Ln9.ne4_oQU240.F2010.chrysalis_intel.eam-outfrq9s (Overall: DIFF) details:
  SMS_Ly2_P1x1.1x1_smallvilleIA.IELMCNCROP.chrysalis_intel.elm-force_netcdf_pio (Overall: DIFF) details:
  SMS_Ly2_P1x1.1x1_smallvilleIA.IELMCNCROP.chrysalis_intel.elm-per_crop (Overall: DIFF) details:
  SMS_Ly2_P1x1_D.1x1_smallvilleIA.IELMCNCROP.chrysalis_intel.elm-lulcc_sville (Overall: DIFF) details
Appear to be only due to the new history field added. Could put an if statement around the new history field addition with the new namelist flag ?

I removed the new variable in the history field, as it is not needed. Let me know if it passes the test now.

peterdschwartz · 2023-11-08T18:52:43Z

Yes it looks like the DIFFs are resolved.

…n-scheme' into next (PR #5974) In default, the infiltration rate is the same for the non-water surface and surface water storage within the grid cell. This assumption results in a significant underestimation of surface water because the infiltration in the surface water storage is not constrained. In this PR, we introduced a modified infiltration scheme to constrain the infiltration in the surface water storage. The infiltration rate in the surface water storage is smaller than the infiltration rate in the non-water surface. More details about this new scheme and corresponding evaluations can be found in the preprint: https://doi.org/10.21203/rs.3.rs-2733749/v1 The manuscript is still under review. The new infiltration scheme can be turned by adding use_modified_infil = .true. in user_nl_elm. We passed all the tests in e3sm_land_developer test suite. We also created a test for the simulation of surface water dynamics with the new infiltration scheme and calibrated parameter for NLDAS domain: surface_water_dynamics. This test was added into e3sm_land_developer test suite. [BFB] [STEALTH]

peterdschwartz · 2023-11-14T17:24:56Z

merged to next

peterdschwartz · 2023-11-16T19:08:44Z

merged to master. Will wait till tomorrow to bless pm test

donghuix added 8 commits September 8, 2023 13:12

initial add code for modified infiltration scheme for surface water s…

cd518ae

…torage

debug

a0d0419

add mosart code back

401a417

fix typo

bdc29d3

fix typo

3e5226b

add a test for the new infiltration scheme

d231bea

debug fover, need to remove _r8 to keep it consistent with master

6fe63a4

add tests in elm for surface water dynamics with calibrated parameter…

8057ebc

…s for NLDAS domain

rljacob assigned peterdschwartz Oct 9, 2023

rljacob added the Land label Oct 9, 2023

bishtgautam added the BFB PR leaves answers BFB label Oct 9, 2023

bishtgautam approved these changes Oct 10, 2023

View reviewed changes

address comments

f85ea21

rljacob added the Stealth PR has feature which, if turned on, could change climate. fka FCC label Oct 16, 2023

remove frac_h2osfc_act from output

0b7cdf4

peterdschwartz merged commit a3a53ba into E3SM-Project:master Nov 16, 2023
2 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add a modified infiltration scheme in ELM to improve surface water dynamics #5974

Add a modified infiltration scheme in ELM to improve surface water dynamics #5974

donghuix commented Oct 9, 2023 •

edited by peterdschwartz

Loading

bishtgautam left a comment

bishtgautam Oct 10, 2023

donghuix Oct 10, 2023

bishtgautam Oct 16, 2023

peterdschwartz commented Oct 11, 2023

donghuix commented Oct 16, 2023

peterdschwartz commented Oct 16, 2023

rljacob commented Oct 16, 2023

peterdschwartz commented Nov 7, 2023

bishtgautam commented Nov 7, 2023

peterdschwartz commented Nov 8, 2023

donghuix commented Nov 8, 2023

peterdschwartz commented Nov 8, 2023

peterdschwartz commented Nov 14, 2023

peterdschwartz commented Nov 16, 2023

Add a modified infiltration scheme in ELM to improve surface water dynamics #5974

Add a modified infiltration scheme in ELM to improve surface water dynamics #5974

Conversation

donghuix commented Oct 9, 2023 • edited by peterdschwartz Loading

bishtgautam left a comment

Choose a reason for hiding this comment

bishtgautam Oct 10, 2023

Choose a reason for hiding this comment

donghuix Oct 10, 2023

Choose a reason for hiding this comment

bishtgautam Oct 16, 2023

Choose a reason for hiding this comment

peterdschwartz commented Oct 11, 2023

donghuix commented Oct 16, 2023

peterdschwartz commented Oct 16, 2023

rljacob commented Oct 16, 2023

peterdschwartz commented Nov 7, 2023

bishtgautam commented Nov 7, 2023

peterdschwartz commented Nov 8, 2023

donghuix commented Nov 8, 2023

peterdschwartz commented Nov 8, 2023

peterdschwartz commented Nov 14, 2023

peterdschwartz commented Nov 16, 2023

donghuix commented Oct 9, 2023 •

edited by peterdschwartz

Loading