NOAA-SWPC / WAM

Whole Atmosphere Model extension of the GFS
GNU General Public License v3.0
3 stars 7 forks source link

Verification weimer2005 (swden) .vs. old weimer in WAM against the satellite data #9

Closed ZhuxiaoLi66 closed 4 years ago

ZhuxiaoLi66 commented 5 years ago

The new Weimer (2005) has been plugged in WAM with the extra input driving parameter 'swden'. Recently, with Martin's great works, Tim, Martin and I are grouped to do the verification of the two version Weimer against the GOCE and GRACE satellite data. We are investigating two storm cases, the 2013 (GOCE) and 2015 (GRACE) St. Patrick storm. Based on the current results, we would like to say the new weimer makes the total number density at about 410km (Grace track height) during 2015 St. Patrick storm time (Mar.17-18) down to a much more reasonable amplitude although still has kind of higher density compared with the Grace data.

The details of the ongoing verification, further experiments and checking & adjustment will be posted here.

ZhuxiaoLi66 commented 5 years ago

Here is Martin's validation works on 2015 St. Patrick storm. del8_WAM.docx 2015_Weimer_WAM.docx old_weimer_2015_storm Weimer2005_2015_storm_grace

ZhuxiaoLi66 commented 5 years ago

Here is Martin's validation work on 2013 St. Patrick storm, 2013_st_patrick_storm_GOCE_weimer2005 2013_st_patrick_storm_GOCE_weimer2005 old_weimer_GOCE_2013_storm

ZhuxiaoLi66 commented 5 years ago

For the further investigation on the higher total number density on Mar.17-18 2015, we checked the input driving parameters for both 2015 and 2013 St. Patrick storm period. A run with swden=5 for 2015 storm case has conducted and the output are processing. we would like to check the dependency of number density on swden.

the following are some inputs for the two storm. SW_Den_two_storms_date_5days Kp_2_storms_5days Bz_2_storms_5days SW_Vel_2_storms_5days

ZhuxiaoLi66 commented 5 years ago

The following for the longer period Mar1-20 for both 2013 and 2015. F107_two_storms_date Kp_two_storms_date SW_Den_two_storms_date SW_Bt_two_storms_date SW_Vel_two_storms_date

ZhuxiaoLi66 commented 5 years ago

Hi Adam, would like to ask your help on the 'fixall' run issue in my e-mail yesterday as following.

I plan to do a driving parameters 'fixall' WAM standalone run. would like to double check with you the setting options for parameters in the user.config files,

if you are using fixderive or fixall, uncomment the following and set your fixed values

export FIX_F107=120.0 # F10.7 and F10.7d export FIX_KP=3.0 # Kp and 24hr average Kp export FIX_SWBT=1.0 # total solar wind export FIX_SWANG=0.0 # solar wind angle export FIX_SWVEL=400.0 # solar wind velocity export FIX_SWBZ=1.0 # z-component of solar wind export FIX_GWATTS=4.5 # gigawatts export FIX_HPI=2 # hemispheric power index are these enough? it seems have 4 variables NHemiPow, NHemiPowIdx, SHemiPow SHemiPowIdx about the 'hemisphere power' in your wam_input_f107_kp.txt file.

I submitted a test run with the 'fixall' option, and the above setting, I got some error message in /scratch3/NCEPDEV/swpc/scrub/Zhuxiao.Li/PTMP/swden_weimer2005_fixall_20130301_60s_1day/ fcst.806550

The run script is in: /scratch3/NCEPDEV/swpc/save/Zhuxiao.Li/git/WAM-IPE_swden_2/scripts/compsets/user.config_20130301_fix_swden_60s

akubaryk commented 5 years ago

FIX_GWATTS and FIX_HPI are used for NHemiPow/SHemiPow and NHemiPowIdx/SHemiPowIdx respectively -- they are simply duplicated values for each hemisphere at the moment. The script seems to not have FIX_SWBZ based on the fcst.806550 file, which I don't understand exactly because it is in the user.config file. Going to run a test now.

akubaryk commented 5 years ago

Oops -- it's the $FIX_SWDEN variable instead. While I added that functionality to the branch, I forgot to put it in the example user.config files. Simply add that value to the user.config you are using and it will work.

ZhuxiaoLi66 commented 5 years ago

Adam, it does works after adding FIX_SWDEN, and I checked the wam_input_f107_kp.txt, the parameters looks great, thanks!

ZhuxiaoLi66 commented 5 years ago

the WAM run with 'fixall' option in run script seems too warm.

fixall_run_20130301_15

ZhuxiaoLi66 commented 5 years ago

Martin's validation plots for the Apr 5-7 2010 storm run, we don't know the reason, while the difference seems pretty large for the comparison of the WAM output with each satellite (GRACE, CHAMP and GOCE).
Normalized_2010Results.docx

ZhuxiaoLi66 commented 5 years ago

GOCE_20100405_07

timfullerrowell commented 5 years ago

Zhuziao, Is this the annual run with 1.25 as the JH_factor? Still a bit hot?

Did the new April run with 1.25 go through? Tim

On Mon, Jun 24, 2019 at 12:15 PM ZhuxiaoLi notifications@github.com wrote:

[image: fixall_run_20130301_15] https://user-images.githubusercontent.com/22546571/60041944-0c838d00-96ac-11e9-8374-fbadace7f7ca.png

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/SWPC-IPE/WAM-IPE/issues/296?email_source=notifications&email_token=AH5BFOASPGULAPFUX5UZBL3P4EFMPA5CNFSM4HSLZR4KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYNYU6A#issuecomment-505121400, or mute the thread https://github.com/notifications/unsubscribe-auth/AH5BFOHWEJAIHFXDMH5LMCDP4EFMPANCNFSM4HSLZR4A .

ZhuxiaoLi66 commented 5 years ago

Here is the result for the new JH_factor of 1.25 + ... The result seems too cold and flat.

GOCE_20100405-07_JH