Add observation preparation job for aerosols DA to workflow (NOAA-EMC…

…#2624) Add a prepaeroobs job to prepare aerosol obs files for DA. This job does quality control of the VIIRS aerosol raw observations and convert them to ioda format. Resolves NOAA-EMC#2623 --------- Co-authored-by: ypwang19 <[email protected]> Co-authored-by: TerrenceMcGuinness-NOAA <[email protected]> Co-authored-by: Cory Martin <[email protected]> Co-authored-by: David Huber <[email protected]>
KateFriedman-NOAA · Jun 14, 2024 · 6c93b45 · 6c93b45
1 parent 5a5fc2b
commit 6c93b45
Show file tree

Hide file tree

Showing 20 changed files with 441 additions and 3 deletions.
diff --git a/env/HERA.env b/env/HERA.env
@@ -102,6 +102,13 @@ elif [[ "${step}" = "atmanlfv3inc" ]]; then
     [[ ${NTHREADS_ATMANLFV3INC} -gt ${nth_max} ]] && export NTHREADS_ATMANLFV3INC=${nth_max}
     export APRUN_ATMANLFV3INC="${launcher} -n ${npe_atmanlfv3inc} --cpus-per-task=${NTHREADS_ATMANLFV3INC}"
 
+elif [[ "${step}" = "prepobsaero" ]]; then
+
+    nth_max=$((npe_node_max / npe_node_prepobsaero))
+
+    export NTHREADS_PREPOBSAERO=${nth_prepobsaero:-1}
+    export APRUN_PREPOBSAERO="${launcher} -n ${npe_prepobsaero} --cpus-per-task=${NTHREADS_PREPOBSAERO}"
+
 elif [[ "${step}" = "snowanl" ]]; then
 
     nth_max=$((npe_node_max / npe_node_snowanl))

diff --git a/env/HERCULES.env b/env/HERCULES.env
@@ -99,6 +99,12 @@ case ${step} in
     [[ ${NTHREADS_AEROANL} -gt ${nth_max} ]] && export NTHREADS_AEROANL=${nth_max}
     export APRUN_AEROANL="${launcher} -n ${npe_aeroanlrun} --cpus-per-task=${NTHREADS_AEROANL}"
  ;;
+ "prepobsaero")
+    nth_max=$((npe_node_max / npe_node_prepobsaero))
+
+    export NTHREADS_PREPOBSAERO=${nth_prepobsaero:-1}
+    export APRUN_PREPOBSAERO="${launcher} -n ${npe_prepobsaero} --cpus-per-task=${NTHREADS_PREPOBSAERO}"
+;;
  "snowanl")
 
     nth_max=$((npe_node_max / npe_node_snowanl))

diff --git a/env/JET.env b/env/JET.env
@@ -82,6 +82,13 @@ elif [[ "${step}" = "aeroanlrun" ]]; then
     [[ ${NTHREADS_AEROANL} -gt ${nth_max} ]] && export NTHREADS_AEROANL=${nth_max}
     export APRUN_AEROANL="${launcher} -n ${npe_aeroanlrun}"
 
+elif [[ "${step}" = "prepobsaero" ]]; then
+
+    nth_max=$((npe_node_max / npe_node_prepobsaero))
+
+    export NTHREADS_PREPOBSAERO=${nth_prepobsaero:-1}
+    export APRUN_PREPOBSAERO="${launcher} -n ${npe_prepobsaero} --cpus-per-task=${NTHREADS_PREPOBSAERO}"
+
 elif [[ "${step}" = "snowanl" ]]; then
 
     nth_max=$((npe_node_max / npe_node_snowanl))

diff --git a/env/ORION.env b/env/ORION.env
@@ -90,6 +90,13 @@ elif [[ "${step}" = "aeroanlrun" ]]; then
     [[ ${NTHREADS_AEROANL} -gt ${nth_max} ]] && export NTHREADS_AEROANL=${nth_max}
     export APRUN_AEROANL="${launcher} -n ${npe_aeroanlrun} --cpus-per-task=${NTHREADS_AEROANL}"
 
+elif [[ "${step}" = "prepobsaero" ]]; then
+
+    nth_max=$((npe_node_max / npe_node_prepobsaero))
+
+    export NTHREADS_PREPOBSAERO=${nth_prepobsaero:-1}
+    export APRUN_PREPOBSAERO="${launcher} -n ${npe_prepobsaero} --cpus-per-task=${NTHREADS_PREPOBSAERO}"
+
 elif [[ "${step}" = "snowanl" ]]; then
 
     nth_max=$((npe_node_max / npe_node_snowanl))

diff --git a/env/S4.env b/env/S4.env
@@ -82,6 +82,13 @@ elif [[ "${step}" = "aeroanlrun" ]]; then
     [[ ${NTHREADS_AEROANL} -gt ${nth_max} ]] && export NTHREADS_AEROANL=${nth_max}
     export APRUN_AEROANL="${launcher} -n ${npe_aeroanlrun}"
 
+elif [[ "${step}" = "prepobsaero" ]]; then
+
+    nth_max=$((npe_node_max / npe_node_prepobsaero))
+
+    export NTHREADS_PREPOBSAERO=${nth_prepobsaero:-1}
+    export APRUN_PREPOBSAERO="${launcher} -n ${npe_prepobsaero} --cpus-per-task=${NTHREADS_PREPOBSAERO}"
+
 elif [[ "${step}" = "snowanl" ]]; then
 
     nth_max=$((npe_node_max / npe_node_snowanl))

diff --git a/env/WCOSS2.env b/env/WCOSS2.env
@@ -76,6 +76,13 @@ elif [[ "${step}" = "aeroanlrun" ]]; then
     [[ ${NTHREADS_AEROANL} -gt ${nth_max} ]] && export NTHREADS_AEROANL=${nth_max}
     export APRUN_AEROANL="${launcher} -n ${npe_aeroanlrun}"
 
+elif [[ "${step}" = "prepobsaero" ]]; then
+
+    nth_max=$((npe_node_max / npe_node_prepaeroobs))
+
+    export NTHREADS_PREPOBSAERO=${nth_prepobsaero:-1}
+    export APRUN_PREPOBSAERO="${launcher} -n ${npe_prepobsaero} --ppn ${npe_node_prepobsaero}--cpu-bind depth --depth=${NTHREADS_PREPOBSAERO}"
+
 elif [[ "${step}" = "snowanl" ]]; then
 
     nth_max=$((npe_node_max / npe_node_snowanl))

diff --git a/jobs/JGLOBAL_PREP_OBS_AERO b/jobs/JGLOBAL_PREP_OBS_AERO
@@ -0,0 +1,43 @@
+#! /usr/bin/env bash
+
+source "${HOMEgfs}/ush/preamble.sh"
+source "${HOMEgfs}/ush/jjob_header.sh" -e "prepobsaero" -c "base prepobsaero"
+
+##############################################
+# Set variables used in the script
+##############################################
+
+export COMIN_OBS="${DATA}"
+YMD=${PDY} HH=${cyc} declare_from_tmpl -rx COMOUT_OBS:COM_OBS_TMPL
+
+##############################################
+# Begin JOB SPECIFIC work
+##############################################
+
+###############################################################
+# Run relevant script
+
+EXSCRIPT=${GDASPREPAEROOBSPY:-${SCRgfs}/exglobal_prep_obs_aero.py}
+${EXSCRIPT}
+status=$?
+[[ ${status} -ne 0 ]] && exit "${status}"
+
+
+##############################################
+# End JOB SPECIFIC work
+##############################################
+
+##############################################
+# Final processing
+##############################################
+if [[ -e "${pgmout}" ]] ; then
+  cat "${pgmout}"
+fi
+
+##########################################
+# Remove the Temporary working directory
+##########################################
+cd "${DATAROOT}" || exit
+[[ "${KEEPDATA}" = "NO" ]] && rm -rf "${DATA}"
+
+exit 0
diff --git a/jobs/rocoto/prepobsaero.sh b/jobs/rocoto/prepobsaero.sh
@@ -0,0 +1,24 @@
+#! /usr/bin/env bash
+
+source "${HOMEgfs}/ush/preamble.sh"
+
+###############################################################
+# Source UFSDA workflow modules
+. "${HOMEgfs}/ush/load_ufsda_modules.sh"
+status=$?
+[[ ${status} -ne 0 ]] && exit "${status}"
+
+export job="prepobsaero"
+export jobid="${job}.$$"
+
+###############################################################
+# setup python path for workflow utilities and tasks
+wxflowPATH="${HOMEgfs}/ush/python:${HOMEgfs}/ush/python/wxflow/src"
+PYTHONPATH="${PYTHONPATH:+${PYTHONPATH}:}${wxflowPATH}"
+export PYTHONPATH
+
+###############################################################
+# Execute the JJOB
+"${HOMEgfs}/jobs/JGLOBAL_PREP_OBS_AERO"
+status=$?
+exit "${status}"
diff --git a/parm/archive/arcdir.yaml.j2 b/parm/archive/arcdir.yaml.j2
@@ -47,6 +47,11 @@ Deterministic: &deterministic
             - ["{{ COM_CHEM_ANALYSIS }}/{{ head }}aerostat", "{{ ARCDIR }}/aerostat.{{ RUN }}.{{ cycle_YMDH }}"]
             {% endif %}
 
+            {% if DO_PREP_OBS_AERO %}
+            - ["{{ COM_OBS }}/{{ head }}aeroobs", "{{ ARCDIR }}/aeroobs.{{ RUN }}.{{ cycle_YMDH }}"]
+            - ["{{ COM_OBS }}/{{ head }}aerorawobs", "{{ ARCDIR }}/aerorawobs.{{ RUN }}.{{ cycle_YMDH }}"]
+            {% endif %}
+
     {% endif %}
 
 {% if RUN == "gfs" %}

diff --git a/parm/archive/gdas.yaml.j2 b/parm/archive/gdas.yaml.j2
@@ -69,6 +69,10 @@ gdas:
         {% if AERO_ANL_CDUMP == "gdas" or AERO_ANL_CDUMP == "both" %}
         - "{{ COM_CHEM_ANALYSIS | relpath(ROTDIR) }}/{{ head }}aerostat"
         {% endif %}
+        {% if DO_PREP_OBS_AERO %}
+        - "{{ COM_OBS | relpath(ROTDIR) }}/{{ head }}aeroobs"
+        - "{{ COM_OBS | relpath(ROTDIR) }}/{{ head }}aerorawobs"
+        {% endif %}
         {% if DO_JEDISNOWDA %}
         - "{{ COM_SNOW_ANALYSIS | relpath(ROTDIR) }}/{{ head }}snowstat.tgz"
         {% endif %}

diff --git a/parm/archive/gfsa.yaml.j2 b/parm/archive/gfsa.yaml.j2
@@ -40,6 +40,10 @@ gfsa:
         {% if AERO_ANL_CDUMP == "gfs" or AERO_ANL_CDUMP == "both" %}
         - "{{ COM_CHEM_ANALYSIS | relpath(ROTDIR) }}/{{ head }}aerostat"
         {% endif %}
+        {% if DO_PREP_OBS_AERO %}
+        - "{{ COM_OBS | relpath(ROTDIR) }}/{{ head }}aeroobs"
+        - "{{ COM_OBS | relpath(ROTDIR) }}/{{ head }}aerorawobs"
+        {% endif %}
 
         # BUFR inputs
         - "{{ COM_OBS | relpath(ROTDIR) }}/{{ head }}nsstbufr"

diff --git a/parm/config/gfs/config.base b/parm/config/gfs/config.base
@@ -178,6 +178,7 @@ export DO_WAVE="NO"
 export DO_OCN="NO"
 export DO_ICE="NO"
 export DO_AERO="NO"
+export DO_PREP_OBS_AERO="NO"
 export AERO_FCST_CDUMP="" # When to run aerosol forecast: gdas, gfs, or both
 export AERO_ANL_CDUMP="" # When to run aerosol analysis: gdas, gfs, or both
 export WAVE_CDUMP="" # When to include wave suite: gdas, gfs, or both

diff --git a/parm/config/gfs/config.prepobsaero b/parm/config/gfs/config.prepobsaero
@@ -0,0 +1,17 @@
+#!/bin/bash -x
+
+########## config.prepobsaero ##########
+# Prepare and thin/superob aerosol observations
+
+echo "BEGIN: config.prepobsaero"
+
+# Get task specific resources
+source "${EXPDIR}/config.resources" prepobsaero
+
+export OBSPROCYAML="${PARMgfs}/gdas/aero/obs/lists/gdas_aero_obsproc.yaml.j2"
+export OBSPROCEXE="${EXECgfs}/gdas_obsprovider2ioda.x"
+export VIIRS_DATA_DIR="/scratch2/NCEPDEV/stmp3/Yaping.Wang/VIIRS/AWS/"
+export SENSORS="npp,n20"
+
+
+echo "END: config.prepaeroobs"
diff --git a/parm/config/gfs/config.resources b/parm/config/gfs/config.resources
@@ -13,7 +13,7 @@ if (( $# != 1 )); then
     echo "atmanlinit atmanlvar atmanlfv3inc atmanlfinal"
     echo "atmensanlinit atmensanlletkf atmensanlfv3inc atmensanlfinal"
     echo "snowanl"
-    echo "aeroanlinit aeroanlrun aeroanlfinal"
+    echo "prepobsaero aeroanlinit aeroanlrun aeroanlfinal"
     echo "anal sfcanl analcalc analdiag fcst echgres"
     echo "upp atmos_products"
     echo "tracker genesis genesis_fsu"
@@ -289,6 +289,14 @@ case ${step} in
     export npe_node_snowanl=$(( npe_node_max / nth_snowanl ))
     ;;
 
+  "prepobsaero")
+    export wtime_prepobsaero="00:30:00"
+    export npe_prepobsaero=1
+    export nth_prepobsaero=1
+    export npe_node_prepobsaero=1
+    export memory_prepobsaero="96GB"
+    ;;
+
   "aeroanlinit")
     # below lines are for creating JEDI YAML
     case ${CASE} in

diff --git a/scripts/exglobal_archive.py b/scripts/exglobal_archive.py
@@ -19,8 +19,8 @@ def main():
 
     # Pull out all the configuration keys needed to run the rest of archive steps
     keys = ['ATARDIR', 'current_cycle', 'FHMIN', 'FHMAX', 'FHOUT', 'RUN', 'PDY',
-            'DO_VERFRAD', 'DO_VMINMON', 'DO_VERFOZN', 'DO_ICE', 'DO_AERO', 'PARMgfs',
-            'DO_OCN', 'DO_WAVE', 'WRITE_DOPOST', 'PSLOT', 'HPSSARCH', 'DO_MOS',
+            'DO_VERFRAD', 'DO_VMINMON', 'DO_VERFOZN', 'DO_ICE', 'DO_AERO', 'DO_PREP_OBS_AERO',
+            'PARMgfs', 'DO_OCN', 'DO_WAVE', 'WRITE_DOPOST', 'PSLOT', 'HPSSARCH', 'DO_MOS',
             'DO_JEDISNOWDA', 'LOCALARCH', 'REALTIME', 'ROTDIR', 'ARCH_WARMICFREQ',
             'ARCH_FCSTICFREQ', 'ARCH_CYC', 'assim_freq', 'ARCDIR', 'SDATE',
             'FHMIN_GFS', 'FHMAX_GFS', 'FHOUT_GFS', 'ARCH_GAUSSIAN', 'MODE',

diff --git a/scripts/exglobal_prep_obs_aero.py b/scripts/exglobal_prep_obs_aero.py
@@ -0,0 +1,23 @@
+#!/usr/bin/env python3
+# exglobal_prep_obs_aero.py
+# This script collect available viirs
+# obs files, combine and preprocess
+# them.
+import os
+
+from wxflow import Logger, cast_strdict_as_dtypedict
+from pygfs.task.aero_prepobs import AerosolObsPrep
+
+# Initialize root logger
+logger = Logger(level='DEBUG', colored_log=True)
+
+
+if __name__ == '__main__':
+
+    # Take configuration from environment and cast it as python dictionary
+    config = cast_strdict_as_dtypedict(os.environ)
+
+    AeroObs = AerosolObsPrep(config)
+    AeroObs.initialize()
+    AeroObs.runConverter()
+    AeroObs.finalize()