========== 20190201: 1. Checked convergence of Sarith's C180 spinup. 2. Worked on the slides for the LMWG-BGCWG meeting. 3. Jan report. 4. Met with Eunjee to talk about the issue in her AGCM CNNEE clim simulation. ========== 20190202: 1. Worked on the slides for the LMWG-BGCWG meeting. Made GPP comparison plots for more FLUXNET sites. ========== 20190203: 1. Worked on the slides for the LMWG-BGCWG meeting. 2. Analyzed the spinup output: clm4.5_DE720: 16 through 20 clm4.5_Princeton_DE720: 31 through 40 Archived these cycles after the analyses. 3. Called and cancelled hotel in Boulder. ========== 20190204: 1. Worked on the slides for the LMWG-BGCWG meeting and practiced. 2. Watched some lectures of the CTSM tutorial. 3. Nana's dental appointment. ========== 20190205: 1. GMAO seminar. 2. Made GPP comparison plots for more FLUXNET sites. 3. Asked Qing the correct reference for the precipitation data we used in Catchment-CN. 4. Read some reference papers for precipitation and burned area. 5. Processed GMAO GEOS-5 February forecast. 6. Practiced my LMWG-BGCWG meeting in Randy's office. 7. Emailed my slides to the other co-authors for their comments. ========== 20190206: 1. Watched a few lectures of the CTSM tutorial. 2. Revised the presentations slides to incorporate Ben's suggestions. 3. Met with Sarith to talk about the pLAI spinup. 4. Read Sarith's Tech Note about land cover dataset used in Catchment-CN. ========== 20190207: 1. Watched a lecture of the CTSM tutorial. 2. Met with Eunjee to talk about what we should prepare for the IDS meeting next week. She also updated me on the issue in her AGCM CNNEE clim simulation. 3. Asked Robin if I can contact Fang Li about the fire model. ========== 20190208: 1. Met with Eunjee and Lesley about the analyses Eunjee made from the 2014 and 2016 simulations I did using bias corrected Jan hindcast data for the IDS meeting next week, and about the issue in Eunjee's AGCM CNNEE clim simulation. Emailed Lesley the path to the simulations. 2. Filled out, signed and submitted the Conflicts of Interest Form. 3. Completed SSAI Organizational Conflicts of Interest Training on Coggno. 4. Analyzed spinup output: clm4.5_DE720: 21 clm4.5_Princeton_DE720: 41 5. Checked compute_rc.F90 and process_cn.F90 to see how to write out potential GPP. Is the diagnostic psn in process_cn.F90 what we need? It's written out to the output. Checked CN_DriverMod.F90. CNAllocationMod.F90 178: real(r8), pointer :: psnsun(:) ! sunlit leaf-level photosynthesis (umol CO2 /m**2/ s) 271: real(r8), pointer :: psnsun_to_cpool(:) 404: psnsun => pcf%psnsun 490: psnsun_to_cpool => pcf%psnsun_to_cpool 576: ! The input psn (psnsun and psnsha) are expressed per unit LAI 580: psnsun_to_cpool(p) = psnsun(p) * laisun(p) * 12.011e-6_r8 593: gpp(p) = psnsun_to_cpool(p) + psnshade_to_cpool(p) 1380: psnsun_to_cpool(p) = psnsun_to_cpool(p)*(1._r8 - downreg(p)) Checked CNAllocationMod.F90. ! get the time step total gross photosynthesis ! this is coming from the canopy fluxes code, and is the ! gpp that is used to control stomatal conductance. ! For the nitrogen downregulation code, this is assumed ! to be the potential gpp, and the actual gpp will be ! reduced due to N limitation. ! Convert psn from umol/m2/s -> gC/m2/s ! The input psn (psnsun and psnsha) are expressed per unit LAI ! in the sunlit and shaded canopy, respectively. These need to be ! scaled by laisun and laisha to get the total gpp for allocation psnsun_to_cpool(p) = psnsun(p) * laisun(p) * 12.011e-6_r8 psnshade_to_cpool(p) = psnsha(p) * laisha(p) * 12.011e-6_r8 gpp(p) = psnsun_to_cpool(p) + psnshade_to_cpool(p) ! reduce gpp fluxes due to N limitation if (gpp(p) > 0.0_r8) then downreg(p) = excess_cflux(p)/gpp(p) psnsun_to_cpool(p) = psnsun_to_cpool(p)*(1._r8 - downreg(p)) psnshade_to_cpool(p) = psnshade_to_cpool(p)*(1._r8 - downreg(p)) end if Why don't I see gpp updated after psnsun_to_cpool and psnshade_to_cpool are reduced due to N limitation? ========== 20190210: 1. Analyzed spinup output: clm4.5_DE720: 22,23,24 clm4.5_Princeton_DE720: 42,43,44,45,46 2. Prepared slides for meeting with Randy on Monday. ========== 20190211: 1. Watched LMWG-BGCWG meeting presentations on webcast. 2. Revised the monthly reports following a particular format. 3. Prepared slides for meeting with Randy this afternoon. 4. Met with Randy and Eunjee: updated them the low CLM4.5 GPP debugging work I did during the shutdown. ========== 20190212: 1. Worked on upgrading CLM4 to CLM4.5. See notes in /discover/nobackup/fzeng/clm4-to-clm4.5/notes/notes_daily_2019. ========== 20190213: 1. Gave my presentation to the LMWG-BGCWG meeting. 2. Worked on upgrading CLM4 to CLM4.5. See notes in /discover/nobackup/fzeng/clm4-to-clm4.5/notes/notes_daily_2019. 3. IDS meeting. 4. Met with Randy and Eunjee ========== 20190214: 1. Worked on upgrading CLM4 to CLM4.5. See notes in /discover/nobackup/fzeng/clm4-to-clm4.5/notes/notes_daily_2019. ========== 20190219: 1. Dealt with emails after taking 1.5 days off last week. 2. Read Fang Li's 2014 paper. 3. Replied to Fang Li. 4. Worked on upgrading CLM4 to CLM4.5. See notes in /discover/nobackup/fzeng/clm4-to-clm4.5/notes/notes_daily_2019. ========== 20190220: 1. Created ~/python/extract_MERRA2_var_daily.py and ~/python/extract_MERRA2_corr_precip_daily.py to extract daily MERRA2 data for Qingyuan. ~/python > python extract_MERRA2_var_daily.py Traceback (most recent call last): File "extract_MERRA2_var_daily.py", line 7, in from netCDF4 import Dataset ImportError: No module named netCDF4 Loaded these modules below to resolve the issue: module load other/comp/gcc-6.3 module load lib/mkl-18.0.5.274 module load other/SIVO-PyD/spd_1.25.1_gcc-6.3_mkl-17.0.4.196 To run extract_MERRA2_corr_precip_daily.py: ~/python > qsub run_extract_MERRA2_corr_precip.scr 2. Convert compare_ts.m to compare_ts.py. ========== 20190221: 1. Prepared slides for meeting with Randy this afternoon. 2. Met with Eunjee about the IDS work. Emailed Abheera for their bias-corrected hindcast data. 3. Met with Randy and Eunjee to update my work 4. Processed output: clm4.5_DE720: 28 clm4.5_Princeton_DE720: 50,51,52,53,54,55,56 5. Extended the MERRA2 data from May 2018 through Dec 2018 for Randy using ~/merra2/extract_daily.gs. Randy will help me extract the data for the howard forest site for Qingyuan. ========== 20190222: 1. Emailed Randy about the extended MERRA2 data, and about extracting data for Qingyuan. Checked the precipitation extracted by . It seems to be the MERRA2 precipitation data before and after correction (but not the precip_corr_CPCUGPCP22clim_MERRA2_BMTXS). 2. Moved some emails from NASA mail to GMAO mail. 3. Checked ~/geos5/FORECASTS_BCSD/Fanwei/download/Download_Monthly_MERRA2.scr and found that PRECTOT is extracted instead of PRECTOTCORR. Emailed Abheera to check with her. 4. Checked /gpfsm/dnb04/projects/p71/aogcm/g5fcst/forecast/production/geos-s2s/runx/2015/dec12/ens1/geosgcm_vis2d/dec12.geosgcm_vis2d.monthly.201601.nc4 and don't see PRECLS there. Need to ask Zhao Li. 5. ncdump -h /gpfsm/dnb04/projects/p71/aogcm/g5fcst/forecast/production/geos-s2s/runx/2015/dec12/ens1/geosgcm_surf/dec12.geosgcm_surf.monthly.201601.nc4 float ANPRCP(time, lat, lon) ; ANPRCP:long_name = "anvil_precipitation" ; ANPRCP:units = "kg m-2 s-1" ; ANPRCP:_FillValue = 1.e+15f ; ANPRCP:missing_value = 1.e+15f ; ANPRCP:fmissing_value = 1.e+15f ; ANPRCP:vmin = -1.e+15f ; ANPRCP:vmax = 1.e+15f ; float CNPRCP(time, lat, lon) ; CNPRCP:long_name = "convective_precipitation" ; CNPRCP:units = "kg m-2 s-1" ; CNPRCP:_FillValue = 1.e+15f ; CNPRCP:missing_value = 1.e+15f ; CNPRCP:fmissing_value = 1.e+15f ; CNPRCP:vmin = -1.e+15f ; CNPRCP:vmax = 1.e+15f ; float LSPRCP(time, lat, lon) ; LSPRCP:long_name = "nonanvil_large_scale_precipitation" ; LSPRCP:units = "kg m-2 s-1" ; LSPRCP:_FillValue = 1.e+15f ; LSPRCP:missing_value = 1.e+15f ; LSPRCP:fmissing_value = 1.e+15f ; LSPRCP:vmin = -1.e+15f ; LSPRCP:vmax = 1.e+15f ; float PCU(time, lat, lon) ; PCU:long_name = "convective_rainfall" ; PCU:units = "kg m-2 s-1" ; PCU:_FillValue = 1.e+15f ; PCU:missing_value = 1.e+15f ; PCU:fmissing_value = 1.e+15f ; PCU:vmin = -1.e+15f ; PCU:vmax = 1.e+15f ; float PLS(time, lat, lon) ; PLS:long_name = "large_scale_rainfall" ; PLS:units = "kg m-2 s-1" ; PLS:_FillValue = 1.e+15f ; PLS:missing_value = 1.e+15f ; PLS:fmissing_value = 1.e+15f ; PLS:vmin = -1.e+15f ; PLS:vmax = 1.e+15f ; float SNO(time, lat, lon) ; SNO:long_name = "snowfall" ; SNO:units = "kg m-2 s-1" ; SNO:_FillValue = 1.e+15f ; SNO:missing_value = 1.e+15f ; SNO:fmissing_value = 1.e+15f ; SNO:vmin = -1.e+15f ; SNO:vmax = 1.e+15f ; 6. Compared the final 6-hourly corrected GEOS5 data between mine and Abheera's in Africa. Found some differences. The difference in T2M can be up to 3K in Sahara. Is it soly due to the half hour shift in the MERRA2 6-hourly climatology calculation between mine and Abheera's? ========== 20190225: 1. Emailed Abheera asking about the time window of averaging for the final 6-hourly corrected GEOS5 data. 2. Talked to Zhao Li. PRECTOT = PRECCON + PRECLS PRECCON and PRECLS contain snow (PRECSNO). Can follow GEOS_SurfaceGridComp.F90 to compute large-scale rainfall from PRECTOT, PRECCON and PRECSNO. 3. Followed GEOS_SurfaceGridComp.F90 to compute total rainfall and convective rainfall from PRECTOT, PRECCON and PRECSNO in LDAS_Forcing.F90. Compile the code: > cd /discover/nobackup/fzeng/offline_code/GEOSldas_m4-17_6/src/GEOSldas_GridComp/GEOSmetforce_GridComp > setenv ESMADIR /discover/nobackup/fzeng/offline_code/GEOSldas_m4-17_6/ > source $ESMADIR/src/g5_modules > gmake install > cd ../../Applications/LDAS_App/ > gmake install > cd ../../.. > /bin/cp -pr Linux/ exec/BChindcast/. Re-do all the 8 simulations. > cd /discover/nobackup/fzeng/Catchment > mv BChindcast BChindcast_old > mkdir BChindcast > /bin/cp -pr BChindcast_old/run/ BChindcast/. > cd BChindcast Set up the 2016Jan01_ens1 run: > cd run > cp ../../BChindcast_old/2016Jan01_ens1/run/BChindcast.exec . > cd /discover/nobackup/fzeng/offline_code/GEOSldas_m4-17_6/exec/BChindcast/Linux/bin > setenv ESMADIR /discover/nobackup/fzeng/offline_code/GEOSldas_m4-17_6/ > source $ESMADIR/src/g5_modules > ./ldas_setup setup --runmodel /discover/nobackup/fzeng/Catchment/BChindcast/ /discover/nobackup/fzeng/Catchment/BChindcast/run/BChindcast.exec /discover/nobackup/fzeng/Catchment/BChindcast/run/BChindcast.bat 475330 /discover/nobackup/fzeng/Catchment/M2n5P/GEOSldas_CF90/output/CF0090x6C/rc_out/GEOSldas_CF90.ldas_domain.txt 475330 restart: 1 creating dir structure creating restart and bc Correct the tile file if it is an old EASE tile format... cmd: ./preprocess_ldas.x correctease /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/output/CF0090x6C/rc_out/CF0090x6C_DE0360xPE0180-Pfafstetter.til /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/output/CF0090x6C/rc_out/MAPL_CF0090x6C_DE0360xPE0180-Pfafstetter.til nothing nothing nothing Creating f2g.txt.... cmd: ./preprocess_ldas.x c_f2g /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/output/CF0090x6C/rc_out/CF0090x6C_DE0360xPE0180-Pfafstetter.til LDAS_domain_def.nml /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/output/CF0090x6C /discover/nobackup/ltakacs/bcs/Icarus-NL/Icarus-NL_Reynolds/CF0090x6C_DE0360xPE0180/clsm/catchment.def 2016Jan01_ens1 Creating domain..., reading white and black lists if there have ones... Finish domain setup... linking bcs... Creating and lining restart... cmd: ./process_rst.csh g0620 2016Jan01_ens1 /discover/nobackup/fzeng/Catchment/BChindcast /discover/nobackup/ltakacs/bcs/Icarus-NL/Icarus-NL_Reynolds/CF0090x6C_DE0360xPE0180/ CF0090x6C_DE0360xPE0180-Pfafstetter.til 2 1 20160101 GEOSldas_CF90 CF0090x6C /discover/nobackup/fzeng/Catchment/M2n5P/ Please hold on for a while until the restart file is created ..... Updating restart path... creating RC Files Optimizing... decomposition of processes.... cmd: ./preprocess_ldas.x optimize /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/input/tile.data 120 nothing nothing nothing PE90x540-CF creating gcm style batch Run scripts lenkf.j Experiment directory: /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1 > cd /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/run Check restart file: > cat cap_restart 20160101 000000 > ls -l ../input/restart/ lrwxrwxrwx 1 fzeng g0620 148 2019-02-25 12:24 catchcn_internal_rst -> /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/output/CF0090x6C/rs/ens0000/Y2016/M01/2016Jan01_ens1.catchcn_internal_rst.20160101_0000 lrwxrwxrwx 1 fzeng g0620 123 2019-02-25 12:24 vegdyn_internal_rst -> /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/output/CF0090x6C/rs/ens0000/2016Jan01_ens1.vegdyn_internal_rst Check executable: > ls -l ../build/Linux/bin/GEOSldas.x -rwxr-xr-x 1 fzeng g0620 50663221 2019-02-25 12:33 ../build/Linux/bin/GEOSldas.x* Check CAP.rc about the END_DATE, JOB_SGMT and NUM_SGMT. HISTORY.rc: tavg1_2D_mf-st.resolution: 360 181, tavg1_2D_lnd_Nx.resolution: 360 181, Added to L236: 'CNCO2' , 'CATCHCN' , LDAS.rc: added the 3 lines below to the end (L38-40). IMPOSE_CT_CO2: 1 INIT_MON: jan01 ENS_NUM: 1 Did an interactive run: > interactive.py -A sp3 -n 120 -a g0620 -X --debug [Need to get interactive nodes every time for a new experiment!] > cd /discover/nobackup/fzeng/Catchment/BChindcast/2016Jan01_ens1/run > cp lenkf.j lenkf.j.orig > nedit lenkf.j & In lenkf.j: add "exit" in L181, i.e. after "mpirun -map-by core --mca btl ^vader -np $numprocs $GEOSBIN/GEOSldas.x" (so even if you delete lines nothing would happen) > ./lenkf.j It's running. Stopped it at 2016-01-01T05:00:00. > mv lenkf.j.orig lenkf.j > qsub lenkf.j Checked and made sure all the simulations are reading the correct ensemble data. 4. Compared the scripts I modified with the original ones from Abheera to see if I did everything correctly. Files checked: ~/geos5/FORECASTS_BCSD/Fanwei/download/Download_Monthly_MERRA2.scr ~/geos5/FORECASTS_BCSD/Fanwei/download/Download_GEOS5V2_all.H.sh ~/geos5/FORECASTS_BCSD/Fanwei/download/DNload_code_library/process_monthly_forecasts.scr ~/geos5/FORECASTS_BCSD/Fanwei/download/DNload_code_library/run_process_daily_forecasts.scr ~/geos5/FORECASTS_BCSD/Fanwei/download/DNload_code_library/process_daily_forecasts.scr ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/PART1_CLIM.sh Mine: FCST_SYR=1982 FCST_EYR=2010 CLIM_SYR=1982 CLIM_EYR=2010 Abheera's: FCST_SYR=1982 FCST_EYR=2016 CLIM_SYR=1982 CLIM_EYR=2010 This is ok because FCST_SYR and FCST_EYR are actually not used. CLIM_SYR and CLIM_EYR are important and they are identical between mine and Abheera's. ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/Calc_and_Write_observational_climatology.py [No modidification made.] ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/Calc_and_Write_forecast_climatology.py ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/PART2_BCSD-Calc.H.sh: Mine: Used MERRA2 mask Mask_merra2.nc Abheera's: Used CHIRPS mask CHIRPS_MASK.nc and CHIRPS_0.25_MASK.nc netcdf Mask_merra2 { dimensions: lon = 576 ; lat = 361 ; time = UNLIMITED ; // (1 currently) bnds = 2 ; lon = -180, -179.375, -178.75, -178.125, -177.5, -176.875, -176.25, -175.625, -175, -174.375, -173.75, -173.125, -172.5, -171.875, -171.25, -170.625, -170, -169.375, -168.75, -168.125, -167.5, -166.875, -166.25, -165.625, -165, -164.375, -163.75, -163.125, -162.5, -161.875, -161.25, -160.625, -160, -159.375, -158.75, -158.125, -157.5, -156.875, -156.25, -155.625, -155, -154.375, -153.75, -153.125, -152.5, -151.875, -151.25, -150.625, -150, -149.375, -148.75, -148.125, -147.5, -146.875, -146.25, lat = -90, -89.5, -89, -88.5, -88, -87.5, -87, -86.5, -86, -85.5, -85, -84.5, -84, -83.5, -83, -82.5, -82, -81.5, -81, -80.5, -80, -79.5, -79, -78.5, -78, -77.5, -77, -76.5, -76, -75.5, -75, -74.5, -74, -73.5, -73, -72.5, -72, -71.5, -71, -70.5, -70, -69.5, -69, -68.5, -68, -67.5, -67, -66.5, -66, -65.5, -65, -64.5, -64, -63.5, -63, -62.5, -62, -61.5, -61, netcdf CHIRPS_MASK { dimensions: lon = 576 ; lat = 361 ; time = UNLIMITED ; // (1 currently) bnds = 2 ; float mask(time, lat, lon) ; mask:standard_name = "convective precipitation rate" ; mask:long_name = "Climate Hazards group InfraRed Precipitation with Stations" ; mask:units = "kg m-2 s-1" ; mask:_FillValue = -9999.f ; mask:missing_value = -9999.f ; mask:time_step = "day" ; mask:geostatial_lat_min = -50.f ; mask:geostatial_lat_max = 50.f ; mask:geostatial_lon_min = -180.f ; mask:geostatial_lon_max = 180.f ; lon = -180, -179.375, -178.75, -178.125, -177.5, -176.875, -176.25, -175.625, -175, -174.375, -173.75, -173.125, -172.5, -171.875, -171.25, -170.625, -170, -169.375, -168.75, -168.125, -167.5, -166.875, -166.25, -165.625, -165, -164.375, -163.75, -163.125, -162.5, -161.875, -161.25, -160.625, -160, -159.375, -158.75, -158.125, -157.5, -156.875, -156.25, lat = -90, -89.5, -89, -88.5, -88, -87.5, -87, -86.5, -86, -85.5, -85, -84.5, -84, -83.5, -83, -82.5, -82, -81.5, -81, -80.5, -80, -79.5, -79, -78.5, -78, -77.5, -77, -76.5, -76, -75.5, -75, -74.5, -74, -73.5, -73, -72.5, -72, -71.5, -71, -70.5, -70, -69.5, -69, -68.5, -68, -67.5, -67, -66.5, -66, -65.5, -65, -64.5, -64, -63.5, -63, -62.5, -62, -61.5, -61, Since CHIRPS_MASK.nc and Mask_merra2.nc have the same dimension and same lat and lon values, using Mask_merra2.nc should not have caused the difference in T2M. I need to use Mask_merra2.nc to cover the entire globe. ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/run_BCSD_calc.scr ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/Bias_correction_module.py ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/PART3_TmpDisagg.H.sh ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/run_Temporal_disagg.scr Again, Mine: Used MERRA2 mask Mask_merra2.nc Abheera's: Used CHIRPS mask CHIRPS_MASK.nc and CHIRPS_0.25_MASK.nc ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/Daily_Temporal_disaggregation_module.py ~/geos5/FORECASTS_BCSD/Fanwei/BCSD/FAME_Dec_V2/code_library/Sub_Daily_Temporal_disaggregation_module.py ========== 20190226: 1. Investigated whether the difference we see in the final 6-hourly bias corrected GEOS5 data could be due to the difference in the MERRA2 6-hourly climatology. > cd /discover/nobackup/projects/gmao/geos_carb/fzeng/FORECASTS/GEOS5/CLIM/GEOS5v2/BCSD_Final/6-hourly/2016/jan01/ens1 > module load other/nco-4.6.8-gcc-5.3-sp3 > xdiff GEOS5.201601.nc4 /gpfsm/dnb02/projects/p63/FORECASTS/GEOS5/BCSD_Test/FAME_Jan_data/GEOS5v2/BCSD_Final/6-hourly/2016/jan01/ens1/GEOS5.201601.nc4 201601.diff.nc4 > ncview 201601.diff.nc4 & 2. Met with Eunjee and Lesley. 3. Prepared a couple of slides on how the bias correction is done for the IDS meeting tomorrow. ========== 20190227: 1. Continued working on the slides for the IDS meeting this afternoon. 2. IDS meeting. 3. Prepared slides for meeting with Randy tomorrow. 4. Created ~/python/IDS/extract_MERRA2_var_hourly.py and used it to plot hourly time series of MERRA2 T2M for 2 grid cells, one in Sahara and the other in China. The difference in T2M between 2 consecutive hours can be up to 6 degree in Sahara, and 2 degree for China. These differences seem to be able to explain the difference in the MERRA2 6-hourly climatology between mine and Code 617's. There is a half-hour shift between my MERRA2 6-hourly climatology and theirs. ========== 20190228: 1. Continued working on thte slides for meeting with Randy this afternoon. 2. Continued investigating whether the difference we see in the final 6-hourly bias corrected GEOS5 data could be due to the difference in the MERRA2 6-hourly climatology. > cd /discover/nobackup/projects/gmao/geos_carb/fzeng/FORECASTS/GEOS5/CLIM/GEOS5v2/BCSD_Final/6-hourly/2016/jan01/ens1 > ncview 201601.diff.nc4 & Also plotted differences in the final 6-hourly corrected data using ~/geos5/FORECASTS_BCSD/Fanwei/verify_output.m. 3. Added PARDR, PARDF and PS to ~/merra2/extract_daily.gs and ran it to re-extract all the 17 fields from 19800101 through 20181231. Used plot_merra2.gs under the same directory to check the fields for 19800101 and they look correct. 4. Met with Randy, Lesley and Eunjee. 5. Feb report.