analysisErrorOI_mod

link to source code

Dependency Diagrams:

Direct Dependency Diagram

Reverse Dependency Diagram

Description

MODULE analysisErrorOI_mod (prefix=’aer’ category=’1. High-level functionality’)

Purpose

Calculate the analysis-error standard deviation. The method used is Optimal Interpolation, where it is assumed that only a subset of the total number of observations influence the analysis at a given grid point.

Quick access

Types

struct_neighborhood

Routines

aer_analysiserror(), aer_computeanlerrorstd(), aer_dayssincelastobs(), aer_propagateanalysiserror(), aer_propagatedslo(), findobs()

Needed modules

• columndata_mod: MODULE columnData_mod (prefix=’col’ category=’6. High-level data objects’)

• gridstatevector_mod: MODULE gridStateVector_mod (prefix=’gsv’ category=’6. High-level data objects’)

• gridstatevectorfileio_mod: MODULE gridStateVectorFileIO_mod (prefix=’gio’ category=’4. Data Object transformations’)

• kdtree2_mod: MODULE kdTree2_mod (prefix=’kdtree2’ category=’8. Low-level utilities and constants’)

• midasmpi_mod: MODULE midasMpi_mod (prefix=’mmpi’ category=’8. Low-level utilities and constants’)

• obsspacedata_mod: MODULE obsSpaceData_mod (prefix=’obs’ category=’6. High-level data objects’)

• physicsfunctions_mod: MODULE physicsFunctions_mod (prefix=’phf’ category=’8. Low-level utilities and constants’)

• statetocolumn_mod: MODULE stateToColumn_mod (prefix=’s2c’ category=’4. Data Object transformations’)

• varnamelist_mod: MODULE varNameList_mod (prefix=’vnl’ category=’7. Low-level data objects’)

• utilities_mod: MODULE utilities_mod (prefix=’utl’ category=’8. Low-level utilities and constants’)

• horizontalcoord_mod: MODULE horizontalCoord_mod (prefix=’hco’ category=’7. Low-level data objects’)

• verticalcoord_mod: MODULE verticalCoord_mod (prefix=’vco’ category=’7. Low-level data objects’)

• obsoperators_mod: MODULE obsOperators_mod (prefix=’oop’ category=’5. Observation operators’)

• message_mod: MODULE message_mod (prefix=’msg’ category=’8. Low-level utilities and constants’)

• oceanmask_mod: MODULE oceanMask_mod (prefix=’ocm’ category=’7. Low-level data objects’)

• timecoord_mod: MODULE timeCoord_mod (prefix=’tim’ category=’7. Low-level data objects’)

Types

• type  analysiserroroi_mod/struct_neighborhood

  Type fields

  • % bodyindex (*) [integer ,pointer]

  • % headerindex (*) [integer ,pointer]

  • % numobs [integer ]

  • % procindex (*) [integer ,pointer]

Variables

Subroutines and functions

subroutine  analysiserroroi_mod/aer_analysiserror(obsspacedata, hco_ptr, vco_ptr)

Purpose

Calculate analysis-error variance.

Arguments

• obsspacedata [struct_obs ,inout] :: observation data structure

• hco_ptr [struct_hco ,in,pointer] :: horizontal grid definition

• vco_ptr [struct_vco ,in,pointer] :: vertical grid definition

Called from

midas_analysiserroroi

Call to

utl_isnamelistpresent(), msg(), utl_abort(), gsv_varnameslist(), gsv_allocate(), gio_readfromfile(), ocm_readmaskfromfile(), aer_propagateanalysiserror(), gsv_copymask(), gsv_modifydate(), tim_getdatestamp(), gio_writetofile(), col_setvco(), col_allocate(), obs_numheader(), s2c_tl(), gsv_zero(), gsv_deallocate(), findobs(), aer_computeanlerrorstd(), col_deallocate(), aer_dayssincelastobs()

subroutine  analysiserroroi_mod/findobs(obsspacedata, statevectortrlerrorstd, numobs, variablename, lcorr, influentobs)

Purpose

Find all observations used for the analysis.

Arguments

• obsspacedata [struct_obs ,in] :: observation data structure

• statevectortrlerrorstd [struct_gsv ,in] :: state containing background error stddev

• numobs (*,*) [integer ,out] :: number of observations found

• variablename [character ,in] :: ‘GL’ for seaice or ‘TM’ for SST

• lcorr (*,*) [real ,in] :: horizontal background-error correlation length scale

• influentobs (*,*) [struct_neighborhood ,inout,pointer] :: details about observations to use in update

Called from

aer_analysiserror()

Call to

utl_abort(), obs_numheader(), obs_numbody(), obs_bodyelem_i(), obs_headelem_i(), obs_headelem_r(), s2c_getfootprintradius(), kdtree2_3dposition(), kdtree2_create(), s2c_getweightsandgridpointindexes(), kdtree2_r_nearest()

subroutine  analysiserroroi_mod/aer_dayssincelastobs(obsspacedata, hco_ptr, vco_ptr, inputfilename, outputfilename, variablename, propagatedslo, hourssincelastanalysis)

Purpose

Update the field “days since last obs” with the newly assimilated obs.

Arguments

• obsspacedata [struct_obs ,inout] :: observation data structure

• hco_ptr [struct_hco ,in,pointer] :: horizontal grid definition

• vco_ptr [struct_vco ,in,pointer] :: vertical grid definition

• inputfilename [character ,in] :: input file name

• outputfilename [character ,in] :: output file name

• variablename [character ,in] :: name of variable being treated

• propagatedslo [logical ,in] :: propagate (increase) Days Since Last Obs in time

• hourssincelastanalysis [integer ,in] :: number of hours between analysis times

Called from

aer_analysiserror()

Call to

gsv_allocate(), obs_numheader(), aer_propagatedslo(), gio_readfromfile(), gsv_copymask(), col_setvco(), col_allocate(), s2c_tl(), gsv_getnumlev(), vnl_varlevelfromvarname(), obs_headelem_i(), obs_bodyelem_i(), s2c_getweightsandgridpointindexes(), gio_writetofile(), col_deallocate(), gsv_deallocate()

subroutine  analysiserroroi_mod/aer_propagateanalysiserror(statevectorerrorstd, oceanmask, variablename, analysisetiket, errorgrowth, hco_ptr, vco_ptr)

Purpose

read analysis error standard deviation field and propagate it forward in time

Arguments

• statevectorerrorstd [struct_gsv ,inout] :: Input: analysis std error; Output: background std error.

• oceanmask [struct_ocm ,in] :: ocean-land mask (1=water, 0=land)

• variablename [character ,in] :: variable name

• analysisetiket [character ,in] :: analysis etiket in the input std file

• errorgrowth [real ,in] :: seaice: fraction of ice per hour, SST: estimated growth

• hco_ptr [struct_hco ,in,pointer] :: horizontal coordinates structure, pointer

• vco_ptr [struct_vco ,in,pointer] :: vertical coordinates structure, pointer

Called from

aer_analysiserror()

Call to

msg(), gsv_allocate(), gio_readfromfile(), gsv_deallocate()

subroutine  analysiserroroi_mod/aer_propagatedslo(statevectorerrorstd, inputfilename, outputfilename, hourssincelastanalysis, hco_ptr)

Purpose

propagate the field “days since last obs” in time.

Arguments

• statevectorerrorstd [struct_gsv ,inout] :: read “analysis” into it and increase by hoursSinceLastAnalysis

• inputfilename [character ,in] :: input file name

• outputfilename [character ,in] :: output file name

• hourssincelastanalysis [integer ,in] :: hours since last analysis (namelist variable)

• hco_ptr [struct_hco ,in,pointer] :: horizontal grid definition

Called from

aer_dayssincelastobs()

Call to

gio_readfromfile(), gsv_modifydate(), tim_getdatestamp(), gio_writetofile()

subroutine  analysiserroroi_mod/aer_computeanlerrorstd(obsspacedata, statevectoranlerrorstd, statevectortrlerrorstd, analysisvariable, maxanalysiserrorstddev, influentobs, lcorr)

Purpose

compute analysis error standard deviation for one analysis variable (grid point) at a time

Arguments

• obsspacedata [struct_obs ,in] :: obsSpaceData structure

• statevectoranlerrorstd [struct_gsv ,inout] :: state vector for analysis error std deviation

• statevectortrlerrorstd [struct_gsv ,in] :: state vector for background error std deviation

• analysisvariable [character ,in] :: variable name (‘GL’ or ‘TM’)

• maxanalysiserrorstddev [real ,in] :: maximum limit imposed on analysis error stddev

• influentobs (*,*) [struct_neighborhood ,in,pointer] :: details about observations to use in update

• lcorr (*,*) [real ,in] :: horizontal background-error length scale

Called from

aer_analysiserror()

Call to

msg(), utl_tmg_start(), obs_numbody(), oop_icescaling(), obs_bodyelem_r(), gsv_getnumlev(), vnl_varlevelfromvarname(), s2c_getweightsandgridpointindexes(), utl_abort(), phf_calcdistance(), utl_matinverse(), utl_tmg_stop()