Adding variables from Noah, NoahMP and surface layer to history file by weiwangncar · Pull Request #399 · wrf-model/WRF

weiwangncar · 2018-03-07T23:43:40Z

TYPE: enhancement (no impact)

KEYWORDS: canopy resistance, LAI, wind speed, mosaic variables

SOURCE: Patrick Campbell, EPA, internal

DESCRIPTION OF CHANGES:

Passing Noah’s canopy resistance (RC2) and LAI (XLAIDYN: “dynamic LAI”) from Noah for both the dominant land use and mosaic tiling option to the grid.
Passing the lowest model layer WSPD variable from the surface driver to the grid now as a state variable, which could then be used (with already present USTAR) in calculating a simple and physically consistent aerodynamic resistance parameterization (RA = WSPD/USTAR^2) used in deposition calculations for coupled air quality models (e.g., WRF-CMAQ in our applications).
Passing a newly calculated canopy conductance (inverse of canopy resistance, RC: uses sunlit and shaded stomatal resistances based on Bonan et al., 2011) in Noah-MP and passing out to the grid (also as RS for WRF-CMAQ consistency) to similarly extend the Noah-MP potential (as in Noah and Noah-Mosaic in 1 and 2) for physically consistent atmospheric deposition and air quality applications.
Adding LAI_mosaic and RS_mosaic to collect mosaic values.
Adding sf_surface_mosaic to global attribute in history file.

LIST OF MODIFIED FILES: list of changed files:
M Registry/Registry.EM_COMMON
M Registry/Registry.NMM
M dyn_em/module_first_rk_step_part1.F
M dyn_em/module_first_rk_step_part2.F
M dyn_em/solve_em.F
M phys/module_sf_noahdrv.F
M phys/module_sf_noahmpdrv.F
M phys/module_sf_noahmplsm.F
M phys/module_surface_driver.F
M share/output_wrf.F

TESTS CONDUCTED:
Before and after tests gave identical results using Jan 2000 data.
reg test mostly successful. Will run one more time when Cheyenne returns.

mgduda · 2018-03-08T00:06:13Z

phys/module_sf_noahdrv.F

    LOGICAL,    INTENT(IN   )    ::     myj,frpcpn

+     REAL, INTENT(OUT) :: RC2(ims:ime,jms:jme)                      !Two-way WRF-CMAQ
+     REAL, INTENT(OUT) :: XLAI2(ims:ime,jms:jme)                    !Two-way WRF-CMAQ


For consistency with other variable declarations, should we change these to

REAL, DIMENSION(ims:ime,jms:jme), INTENT(OUT) :: RC2 ! Two-way WRF-CMAQ REAL, DIMENSION(ims:ime,jms:jme), INTENT(OUT) :: XLAI2 ! Two-way WRF-CMAQ

? Essentially, to use the DIMENSION attribute?

mgduda · 2018-03-08T00:07:41Z

phys/module_sf_noahdrv.F


+
+    REAL, INTENT(OUT) :: RC2(ims:ime,jms:jme)                      !Two-way WRF-CMAQ
+    REAL, INTENT(OUT) :: XLAI2(ims:ime,jms:jme)                    !Two-way WRF-CMAQ


Using the DIMENSION attribute here, too?

mgduda · 2018-03-08T00:10:33Z

phys/module_sf_noahdrv.F

-
+          XLAI2(i,j)= LAI_mosaic_avg(i,j)
+          if(RC_mosaic_avg(i,j) .Gt. 0.0) Then
+          rc2(i,j) = 1.0/(RC_mosaic_avg(i,j))


Why not use indentation here? (Also, why capitalize Then on the previous line?)

mgduda · 2018-03-08T00:13:33Z

phys/module_sf_noahmpdrv.F

+             RS       (I,J)                = 1.0/RS(I,J) !Resistance
+              if(RSSUNXY  (I,J) .le. 0.0.and.RSSHAXY(I,J).le.0.0) Then
+             RS       (I,J)                = 0.0
+              endif


Out-denting seems odd here. Could this logic for RS be changed to if-else-endif with indentation? Also, we could clean up the inconsistent whitespace in the if-test.

mgduda · 2018-03-08T00:14:45Z

phys/module_surface_driver.F

+           DO i = i_start(ij),i_end(ij)
+             wspdsfc(i,j) = wspd(i,j)
+           enddo
+           enddo


Should we capitalize end do here for consistency?

barlage · 2018-03-08T17:10:50Z

phys/module_sf_noahmplsm.F

                     SNEQVO ,SNEQV  ,SH2O   ,SMC    ,SNICE  ,SNLIQ  , & !inout
                     ALBOLD ,CM     ,CH     ,DX     ,DZ8W   ,Q2     , &   !inout
-                     TAUSS  ,                                         & !inout
+                     TAUSS  ,LAISUN ,LAISHA ,RB ,                     & !inout !LAISUN,LAISHA,RB Used in RS calc WRF-CMAQ PCC


comment not necessary

barlage · 2018-03-08T17:15:01Z

phys/module_sf_noahmplsm.F

                 SNEQVO ,SNEQV  ,SH2O   ,SMC    ,SNICE  ,SNLIQ  , & !inout
                 ALBOLD ,CM     ,CH     ,DX     ,DZ8W   ,Q2     , & !inout
-                 TAUSS  ,                                         & !inout
+                 TAUSS  ,LAISUN ,LAISHA ,RB ,                     & !inout  !LAISUN,LAISHA,RB Used in RS calc WRF-CMAQ PCC


comment not necessary

barlage · 2018-03-08T17:15:03Z

phys/module_sf_noahmplsm.F

 		   GHB     , IRG     , IRC     , IRB     , TR      , EVC     , & ! OUT :
 		   CHLEAF  , CHUC    , CHV2    , CHB2    , FPICE   , PAHV    , &
-                   PAHG    , PAHB    , PAH                                     &
+                   PAHG    , PAHB    , PAH     , LAISUN  , LAISHA  , RB        & !OUT   ! LAISUN, LAISHA, and RB Used for RS in WRF-CMAQ - PCC


comment not necessary

barlage · 2018-03-08T17:15:29Z

phys/module_sf_noahmpdrv.F

+    REAL,    DIMENSION( ims:ime,          jms:jme) ::  RBXY        ! leaf boundary layer resistance (s/m)  !Used in RS for WRF-CMAQ PCC
+    REAL,    DIMENSION( ims:ime,          jms:jme) ::  LAISUNXY    ! sunlit leaf area index (m2/m2)  !Used in RS for WRF-CMAQ PCC
+    REAL,    DIMENSION( ims:ime,          jms:jme) ::  LAISHAXY    ! shaded leaf area index (m2/m2)  !Used in RS for WRF-CMAQ PCC
+


Are these 2D variables necessary?

barlage · 2018-03-08T17:17:50Z

phys/module_sf_noahmpdrv.F

+             RS       (I,J)                = ((1.0/(RSSUNXY(I,J)+RBXY(I,J)))*LAISUNXY(I,J)) + ((1.0/(RSSHAXY(I,J)+RBXY(I,J)))*LAISHAXY(I,J))   
+             RS       (I,J)                = 1.0/RS(I,J) !Resistance
+             IF(RSSUNXY  (I,J) .le. 0.0.and.RSSHAXY(I,J).le.0.0) THEN
+                RS    (I,J)                = 0.0


I think this could be rewritten using local variables.

barlage · 2018-03-08T17:18:45Z

phys/module_sf_noahmpdrv.F

 		 SHGXY,    SHCXY,    SHBXY,     EVGXY,    EVBXY,     GHVXY, & ! OUT Noah MP only
 		 GHBXY,    IRGXY,    IRCXY,     IRBXY,     TRXY,     EVCXY, & ! OUT Noah MP only
-              CHLEAFXY,   CHUCXY,   CHV2XY,    CHB2XY,                      & ! OUT Noah MP only
+              CHLEAFXY,   CHUCXY,   CHV2XY,    CHB2XY, RS,                  & ! OUT Noah MP only !RS for WRF-CMAQ PCC


I think these types of comments are unnecessarily specific

barlage · 2018-03-08T17:19:39Z

phys/module_sf_noahmplsm.F

-  REAL                                              :: LAISUN !sunlit leaf area index (m2/m2)
-  REAL                                              :: LAISHA !shaded leaf area index (m2/m2)
+!  REAL                                              :: LAISUN !sunlit leaf area index (m2/m2)
+!  REAL                                              :: LAISHA !shaded leaf area index (m2/m2)


these three commented lines can be removed

weiwangncar · 2018-03-08T18:24:36Z

Code changed according to Mike's suggestions.
Mike, Do you want all CMAQ references removed from the code? There are still many.

mgduda · 2018-03-08T18:31:42Z

@weiwangncar @barlage I'm not sure whether there are general guidelines on this, but it looks like this PR will add about a dozen new variables to the default history files. Unless these fields are likely to be of interest to the general WRF user, should we consider not adding them by default, since they can be easily added with the run-time I/O capability by those who are interested?

barlage · 2018-03-08T18:47:46Z

@weiwangncar I only looked at the Noah-MP related code. Personally, I suggest removing all of these types of comments in mplsm and mpdrv: "!Used in WRF-CMAQ PCC" The variables are much more general than this.

barlage

@weiwangncar I only looked at the Noah-MP changes, but these comments seem overly specific, they are not only "used in rs for wrf-cmaq"

I've commented on the other comments to remove and rewrote the if block for clarity after removing the new variables

barlage · 2018-03-08T19:49:43Z

phys/module_sf_noahmpdrv.F

    REAL,    DIMENSION( ims:ime,          jms:jme ), INTENT(INOUT) ::  Z0        ! combined z0 sent to coupled model
    REAL,    DIMENSION( ims:ime,          jms:jme ), INTENT(INOUT) ::  ZNT       ! combined z0 sent to coupled model
-
+    REAL,    DIMENSION( ims:ime,          jms:jme ), INTENT(INOUT) ::  RS        ! Total stomatal resistance (s/m)  !Used in WRF-CMAQ PCC


remove comment

barlage · 2018-03-08T19:49:59Z

phys/module_sf_noahmpdrv.F

    REAL                                :: RSSHA        ! shaded leaf stomatal resistance (s/m)
+    REAL                                :: RB           ! leaf boundary layer resistance (s/m)  !Used in RS for WRF-CMAQ PCC
+    REAL                                :: LAISUN       ! sunlit leaf area index (m2/m2)  !Used in RS for WRF-CMAQ PCC
+    REAL                                :: LAISHA       ! shaded leaf area index (m2/m2)  !Used in RS for WRF-CMAQ PCC


remove three comments

barlage · 2018-03-08T19:50:11Z

phys/module_sf_noahmpdrv.F

+    REAL                                ::  RBXY        ! leaf boundary layer resistance (s/m)
+    REAL                                ::  LAISUNXY    ! sunlit leaf area index (m2/m2)
+    REAL                                ::  LAISHAXY    ! shaded leaf area index (m2/m2) 
+


remove these four lines

barlage · 2018-03-08T19:50:35Z

phys/module_sf_noahmpdrv.F

+         RB     = undefined_value  !For RS in WRF-CMAQ PCC
+         LAISUN = undefined_value  !For RS in WRF-CMAQ PCC
+         LAISHA = undefined_value  !For RS in WRF-CMAQ PCC
+         RS(I,J)= undefined_value  !For RS in WRF-CMAQ PCC


remove these comments

barlage · 2018-03-08T19:50:45Z

phys/module_sf_noahmpdrv.F

 	    GHB     , IRG     , IRC     , IRB     , TR      , EVC     , & ! OUT :
 	    CHLEAF  , CHUC    , CHV2    , CHB2    , FPICE   , PAHV    , & 
-            PAHG    , PAHB    , PAH                                     &
+            PAHG    , PAHB    , PAH     , LAISUN  , LAISHA  ,RB         & ! LAISUN, LAISHA, and RB Used for RS in WRF-CMAQ - PCC


remove comment

barlage · 2018-03-08T19:50:58Z

phys/module_sf_noahmplsm.F

  REAL                           , INTENT(OUT)   :: PONDING2!surface ponding [mm]
+  REAL                           , INTENT(OUT)   :: RB        ! leaf boundary layer resistance (s/m)  !Used in WRF-CMAQ PCC
+  REAL                           , INTENT(OUT)   :: LAISUN    ! sunlit leaf area index (m2/m2)  !Used in WRF-CMAQ PCC
+  REAL                           , INTENT(OUT)   :: LAISHA    ! shaded leaf area index (m2/m2)  !Used in WRF-CMAQ PCC


remove comments

barlage · 2018-03-08T19:51:52Z

phys/module_sf_noahmpdrv.F

             RSSHAXY  (I,J)                = RSSHA
+             LAISUNXY                      = MAX(LAISUN, 0.0) !New variables needed for RS
+             LAISHAXY                      = MAX(LAISHA, 0.0) !New variables needed for RS
+             RBXY                          = MAX(RB, 0.0)     !New variables needed for RS


remove the XY on the left of these three lines, since these are not output, OK to change via max command

remove comments

barlage · 2018-03-08T19:55:22Z

phys/module_sf_noahmpdrv.F

+             ELSE
+                RS    (I,J)                = ((1.0/(RSSUNXY(I,J)+RBXY)*LAISUNXY) + ((1.0/(RSSHAXY(I,J)+RBXY))*LAISHAXY))   
+                RS    (I,J)                = 1.0/RS(I,J) !Resistance
+             ENDIF


seems easier to understand and probably more efficient if:
change rssunxy(i,j) to rssun
change rsshaxy(i,j) to rssha
change rbxy, laisunxy, laishaxy to rb, laisun, laisha

weiwangncar · 2018-03-08T20:46:18Z

@mgduda It is also my suggestion to the code contributors. But they plead to have them in. Here is part of their reply: "I understand the desire to reduce output in the files, however, many of these changes were driven by some new CMAQ developments that would benefit the community (not just our applications) strongly from having the new small list of variables we have added to be output by default history. In fact, we are going forward making changes to the meteorological preprocessor code (MCIP) for CMAQ with the assumption that these variables may be included as history output now by default. Its not a showstopper for the small changes in #1-3, but just more streamlined for future WRF/Noah-CMAQ applications I suppose." I mentioned the the committee will decide.

mgduda · 2018-03-08T21:12:06Z

@weiwangncar @davegill Is it possible to use "packages" in WRF only to control which fields are written? For example, is there any mechanism that would allow adding a new namelist record, say, &history_fields with a new namelist option like cmaq_fields, which would activate the writing of these new fields? In general, can a package be used only for output control, and not for field allocation?

kkeene44

approved (Jimy)

…history, and use WSPD instead of WSPDSFC

weiwangncar · 2018-03-13T18:24:19Z

This PR is updated to remove all additional variables from history file. The contributor has agreed to use run-time option to add additional fields for CMAQ applications. Making existing i1 variable WSPD to a state variable, and replace the original added variable WSPDSFC. To help with CMAQ application, and to make global attribute more complete, sf_surface_mosaic is added.

… LAI is zero

KEYWORDS: Noah-MP SOURCE: Michael Barlage (NCAR) DESCRIPTION OF CHANGES: Bug fix introduced in wrf-model#399 where divide by zero can occur if LAI is zero. Added check in IF statement. This calculation is only for diagnostics so shouldn't change answers, except to prevent model crash. LIST OF MODIFIED FILES: M phys/module_sf_noahmpdrv.F TESTS CONDUCTED: Summer and winter 24-hr case

#646) TYPE: bug fix KEYWORDS: Noah-MP SOURCE: Michael Barlage (NCAR) DESCRIPTION OF CHANGES: Bug fix introduced in #399 where divide by zero can occur if LAI is zero. Added check in IF statement. This calculation is only for diagnostics so shouldn't change answers, except to prevent model crash. Mods to release-v4.0.1, not develop. Replaces PR #633 LIST OF MODIFIED FILES: M phys/module_sf_noahmpdrv.F TESTS CONDUCTED: Summer and winter 24-hr case

Add epa changes for CMAQ app. More output from history

4019af3

mgduda reviewed Mar 8, 2018

View reviewed changes

address Michaels comments

869e671

barlage reviewed Mar 8, 2018

View reviewed changes

clean up based on review

a43351f

barlage reviewed Mar 8, 2018

View reviewed changes

removing CMAQ specific comments, clean up code in noahmpdrv

f913d96

kkeene44 approved these changes Mar 9, 2018

View reviewed changes

weiwangncar added 5 commits March 9, 2018 17:27

fix compile for NMM

dfc0531

package xlaidyn array with Noah and remove more comments

df950c3

removing RS, RA, LANDUSEF, XLAIDYN and mosaic variables from default …

a940f59

…history, and use WSPD instead of WSPDSFC

add sf_surface_mosaic to global attribute in history file

6186c2f

still need to package two new variables

6de610a

make xlaidyn local var, and add missing rs to the call

699dc6a

clean up to remove CMAQ ref

16a3055

weiwangncar merged commit f80778c into wrf-model:master Mar 20, 2018

barlage pushed a commit to barlage/WRF that referenced this pull request Sep 25, 2018

Bug fix introduced in wrf-model#399 where divide by zero can occur if…

ac98022

… LAI is zero

barlage mentioned this pull request Sep 25, 2018

bugfix: fix potential divide by zero in stomatal resistance diagnostic #633

Closed

davegill mentioned this pull request Oct 1, 2018

bugfix: fix potential divide by zero in stomatal resistance diagnostic #646

Merged



		REAL, INTENT(OUT) :: RC2(ims:ime,jms:jme) !Two-way WRF-CMAQ
		REAL, INTENT(OUT) :: XLAI2(ims:ime,jms:jme) !Two-way WRF-CMAQ

Conversation

weiwangncar commented Mar 7, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

weiwangncar commented Mar 8, 2018

Uh oh!

mgduda commented Mar 8, 2018

Uh oh!

barlage commented Mar 8, 2018

Uh oh!

barlage left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

weiwangncar commented Mar 8, 2018

Uh oh!

mgduda commented Mar 8, 2018

Uh oh!

kkeene44 left a comment

Choose a reason for hiding this comment

Uh oh!

weiwangncar commented Mar 13, 2018

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

weiwangncar commented Mar 7, 2018 •

edited

Loading