Skip to content

added module for assimilating ML SD retrievals #54

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
drdunmire1417 wants to merge 1 commit into
base: archive/master-kul
Choose a base branch
from
Open

added module for assimilating ML SD retrievals #54

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
292 changes: 292 additions & 0 deletions lis/dataassim/obs/ML_SNWD/ML_SNWD_Mod.F90
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,292 @@
!-----------------------BEGIN NOTICE -- DO NOT EDIT-----------------------
! NASA Goddard Space Flight Center Land Information System (LIS) v7.2
!
! Copyright (c) 2015 United States Government as represented by the
! Administrator of the National Aeronautics and Space Administration.
! All Rights Reserved.
!-------------------------END NOTICE -- DO NOT EDIT-----------------------
!BOP
!
! !MODULE: ML_SNWD_Mod
!
! !DESCRIPTION:
! This module contains interfaces and subroutines to
! handle machine leaning ML snow depth retrievals.
!
! !REVISION HISTORY:
! 29 Jul 2024 Devon Dunmire; Initial Specification
!
! This reader can be used for netcdf data in the following format:
! filename: 'SD_yyyymmdd_.nc' with variables 'SD' [m],'lat' and 'lon'


module ML_SNWD_Mod
! !USES:
use ESMF
!EOP
implicit none

PRIVATE

!-----------------------------------------------------------------------------
! !PUBLIC MEMBER FUNCTIONS:
!-----------------------------------------------------------------------------
public :: ML_SNWD_setup
!-----------------------------------------------------------------------------
! !PUBLIC TYPES:
!-----------------------------------------------------------------------------
PUBLIC :: ML_SNWD_struc

type, public :: ML_SNWD_dec
logical :: startMode
real :: ssdev
integer :: snowfield
integer :: nc,nr
real, allocatable :: snwd(:,:)
real, allocatable :: snwdtime(:,:)
end type ML_SNWD_dec

type(ML_SNWD_dec), allocatable :: ML_SNWD_struc(:)

contains
!BOP
!
! !ROUTINE: ML_SNWD_setup
! \label{ML_SNWD_setup}
!
! !INTERFACE:
subroutine ML_SNWD_setup(k, OBS_State, OBS_Pert_State)
! !USES:

use LIS_coreMod
use LIS_timeMgrMod
use LIS_historyMod
use LIS_perturbMod
use LIS_DAobservationsMod
use LIS_logMod
use netcdf

implicit none

! !ARGUMENTS:
integer :: k
type(ESMF_State) :: OBS_State(LIS_rc%nnest)
type(ESMF_State) :: OBS_Pert_State(LIS_rc%nnest)
!
! !DESCRIPTION:
!
! This routine completes the runtime initializations and
! creation of data strctures required for ML snow depth data.
!
! The arguments are:
! \begin{description}
! \item[OBS\_State] observation state
! \item[OBS\_Pert\_State] observation perturbations state
! \end{description}
!EOP
integer :: n
integer :: ftn
integer :: i
integer :: status
type(ESMF_Field) :: obsField(LIS_rc%nnest)
type(ESMF_ArraySpec) :: intarrspec, realarrspec
type(ESMF_Field) :: pertField(LIS_rc%nnest)
type(ESMF_ArraySpec) :: pertArrSpec
character*100 :: MLsnowobsdir
character*100 :: temp
real, allocatable :: obsstd(:)
character*1 :: vid(2)
character*40, allocatable :: vname(:)
real, allocatable :: varmin(:)
real, allocatable :: varmax(:)
type(pert_dec_type) :: obs_pert
real, pointer :: obs_temp(:,:)
real, allocatable :: ssdev(:)

allocate(ML_SNWD_struc(LIS_rc%nnest))

call ESMF_ArraySpecSet(intarrspec,rank=1,typekind=ESMF_TYPEKIND_I4,&
rc=status)
call LIS_verify(status)

call ESMF_ArraySpecSet(realarrspec,rank=1,typekind=ESMF_TYPEKIND_R4,&
rc=status)
call LIS_verify(status)

call ESMF_ArraySpecSet(pertArrSpec,rank=2,typekind=ESMF_TYPEKIND_R4,&
rc=status)
call LIS_verify(status)

call ESMF_ConfigFindLabel(LIS_config,"ML snow depth data directory:",&
rc=status)
do n=1,LIS_rc%nnest
call ESMF_ConfigGetattribute(LIS_config,MLsnowobsdir,&
rc=status)
call LIS_verify(status,'ML snow depth data directory: not defined')
call ESMF_AttributeSet(OBS_State(n),"Data Directory",&
MLsnowobsdir, rc=status)
call LIS_verify(status)
enddo

do n=1,LIS_rc%nnest
call ESMF_AttributeSet(OBS_State(n),"Data Update Status",&
.false., rc=status)
call LIS_verify(status)

call ESMF_AttributeSet(OBS_State(n),"Data Update Time",&
-99.0, rc=status)
call LIS_verify(status)

call ESMF_AttributeSet(OBS_State(n),"Data Assimilate Status",&
.false., rc=status)
call LIS_verify(status)

call ESMF_AttributeSet(OBS_State(n),"Number Of Observations",&
LIS_rc%obs_ngrid(k),rc=status)
call LIS_verify(status)

enddo

write(LIS_logunit,*)'[INFO] read ML snow depth data specifications'

!----------------------------------------------------------------------------
! Create the array containers that will contain the observations and
! the perturbations. Since there is only one layer
! being assimilated, the array size is LIS_rc%ngrid(n).
!----------------------------------------------------------------------------

do n=1,LIS_rc%nnest

write(unit=temp,fmt='(i2.2)') 1
read(unit=temp,fmt='(2a1)') vid

obsField(n) = ESMF_FieldCreate(grid=LIS_obsvecGrid(n,k),&
arrayspec=realarrspec,&
name="Observation"//vid(1)//vid(2), rc=status)
call LIS_verify(status)

!Perturbations State
write(LIS_logunit,*) '[INFO] Opening attributes for observations ',&
trim(LIS_rc%obsattribfile(k))
ftn = LIS_getNextUnitNumber()
open(ftn,file=trim(LIS_rc%obsattribfile(k)),status='old')
read(ftn,*)
read(ftn,*) LIS_rc%nobtypes(k)
read(ftn,*)

allocate(vname(LIS_rc%nobtypes(k)))
allocate(varmax(LIS_rc%nobtypes(k)))
allocate(varmin(LIS_rc%nobtypes(k)))

do i=1,LIS_rc%nobtypes(k)
read(ftn,fmt='(a40)') vname(i)
read(ftn,*) varmin(i),varmax(i)
write(LIS_logunit,*) '[INFO] ',vname(i),varmin(i),varmax(i)
enddo
call LIS_releaseUnitNumber(ftn)

call ESMF_StateAdd(OBS_State(n),(/obsField(n)/),rc=status)
call LIS_verify(status)

allocate(ssdev(LIS_rc%obs_ngrid(k)))

if(trim(LIS_rc%perturb_obs(k)).ne."none") then

allocate(obs_pert%vname(1))
allocate(obs_pert%perttype(1))
allocate(obs_pert%ssdev(1))
allocate(obs_pert%stdmax(1))
allocate(obs_pert%zeromean(1))
allocate(obs_pert%tcorr(1))
allocate(obs_pert%xcorr(1))
allocate(obs_pert%ycorr(1))
allocate(obs_pert%ccorr(1,1))

call LIS_readPertAttributes(1,LIS_rc%obspertAttribfile(k),&
obs_pert)

ssdev = obs_pert%ssdev(1)
ML_SNWD_struc(n)%ssdev =obs_pert%ssdev(1)

pertField(n) = ESMF_FieldCreate(arrayspec=pertArrSpec,&
grid=LIS_obsEnsOnGrid(n,k),name="Observation"//vid(1)//vid(2),&
rc=status)
call LIS_verify(status)

! initializing the perturbations to be zero
call ESMF_FieldGet(pertField(n),localDE=0,farrayPtr=obs_temp,rc=status)
call LIS_verify(status)
obs_temp(:,:) = 0

call ESMF_AttributeSet(pertField(n),"Perturbation Type",&
obs_pert%perttype(1), rc=status)
call LIS_verify(status)

if(LIS_rc%obs_ngrid(k).gt.0) then
call ESMF_AttributeSet(pertField(n),"Standard Deviation",&
ssdev,itemCount=LIS_rc%obs_ngrid(k),rc=status)
call LIS_verify(status)
endif

call ESMF_AttributeSet(pertField(n),"Std Normal Max",&
obs_pert%stdmax(1), rc=status)
call LIS_verify(status)

call ESMF_AttributeSet(pertField(n),"Ensure Zero Mean",&
obs_pert%zeromean(1),rc=status)
call LIS_verify(status)

call ESMF_AttributeSet(pertField(n),"Temporal Correlation Scale",&
obs_pert%tcorr(1),rc=status)

call ESMF_AttributeSet(pertField(n),"X Correlation Scale",&
obs_pert%xcorr(1),rc=status)

call ESMF_AttributeSet(pertField(n),"Y Correlation Scale",&
obs_pert%ycorr(1),rc=status)

call ESMF_AttributeSet(pertField(n),"Cross Correlation Strength",&
obs_pert%ccorr(1,:),itemCount=1,rc=status)

call ESMF_StateAdd(OBS_Pert_State(n),(/pertField(n)/),rc=status)
call LIS_verify(status)
endif

deallocate(vname)
deallocate(varmax)
deallocate(varmin)
deallocate(ssdev)
enddo
!-------------------------------------------------------------
! set up the ML domain %and interpolation weights.
!-------------------------------------------------------------

do n=1,LIS_rc%nnest
call ESMF_ConfigFindLabel(LIS_config,&
"ML snow depth domain x-dimension size:",rc=status)
call ESMF_ConfigGetAttribute(LIS_config,ML_SNWD_struc(n)%nc,rc=status)
call LIS_verify(status,'[ERR] ML snow depth domain x-dimension size: not defined')

call ESMF_ConfigFindLabel(LIS_config,&
"ML snow depth domain y-dimension size:",rc=status)
call ESMF_ConfigGetAttribute(LIS_config,ML_SNWD_struc(n)%nr,rc=status)
call LIS_verify(status,'[ERR] ML snow depth domain y-dimension size: not defined')

allocate(ML_SNWD_struc(n)%snwd(LIS_rc%obs_lnc(k),LIS_rc%obs_lnr(k)))
allocate(ML_SNWD_struc(n)%snwdtime(&
LIS_rc%obs_lnc(k), LIS_rc%obs_lnr(k)))
ML_SNWD_struc(n)%snwd = LIS_rc%udef
ML_SNWD_struc(n)%snwdtime = -1

call LIS_registerAlarm("ML snow depth read alarm",&
86400.0, 86400.0)

ML_SNWD_struc(n)%startMode = .true.
enddo

write(LIS_logunit,*) '[INFO] Created ESMF States to hold ML observations data'

end subroutine ML_SNWD_setup

end module ML_SNWD_Mod

Loading