search

NOAA-SWPC / GSMWAM-IPE

Coupled NEMS app for WAM-IPE
5 stars 11 forks source link

validation on kp derived Bz algorithm #65

Open ZhuxiaoLi66 opened 2 years ago

ZhuxiaoLi66 commented 2 years ago

The current issue will include some further validation on the Kp derived solar wind parameters methodology which has been described in #61.

ZhuxiaoLi66 commented 2 years ago

The following plots show the density changes at 400km between the newBz run and the operational run for 2015 St. Patrick storm (historical free run) and Nov.2-4, 2021 run. DEN_diff_newBz-ops_400km_WAM_2015_Mar16 DEN_diff_newBz-ops_400km_WAM_2015_Mar17 DEN_diff_newBz-ops_400km_WAM_2015_Mar18 DEN_diff_newBz-ops_400km_WAM_2015_Mar19

ZhuxiaoLi66 commented 2 years ago

The '7hrs' and '19hrs' on the title of the following plots means that the data are the output of the 7th or 19th hrs in each circle.

den_diff_400km_7hrs_4P_newBZ_Nov4_2021 den_diff_400km_19hrs_4P_newBZ_Nov4_2021

ZhuxiaoLi66 commented 2 years ago

Hi All, just send an email to you. to clarify the methodology, I would like to classify the data use here for 4 types,

a) observations, a) observations, e.g. GloTEC TEC, Grace neutral density for 2015 storm. b) the simulations with the observed Kp & F107, and other solar drivers. c) the simulations with the forecasted & observed (2015) Kp & F107, and the solar wind drivers derived by the old althorithm (current ops or 'timederive' way). d) the simulations with the forecasted & observed (2015) Kp & F107, and the solar wind drivers derived by the new althorithm ( say 'newBz') The results about 2015 storm is the comparison between d (newBz) and b (timeobs). The results about Nov 2021 runs, is the comparison between c and d. here after, I plan to focus on the comparison between c and d against a & b for the historical 2015 storm runs.

ZhuxiaoLi66 commented 1 year ago

The output of the WAM_para3_newBz run for the 20150316-19 storm has been interpolated and included in the orbit validation plot as attached. The orange curve (WAM_new_newBz) indicates the performance of the new SW algorithm. we can see it makes big progress than the current SW algorithm (WAM_new_ori_derive), although still kind of far from the satellite data and the run with the observational driver input. Since the magnitude of the difference between the new algorithm and the operational output shown in github issue 65 is about 2 orders less than the global mean neutral density (E-13.vs.E-11) and all other validation, I think the results are all good for the merge of the newBz branch into develop.

Den_20150316_storm_para3_newBz_ori_msis2_GRACE_orbit_mean300_6lines

timfullerrowell commented 1 year ago

Much better. Thanks Zhuxiao. We'll have to discuss if we should scale, given the uncertainty there will be with forecasts of Kp. I think we need more examples, but we can continue with quantifying differences in variability and structure (RMSE, SD, etc).

On Fri, Jan 20, 2023 at 12:34 PM ZhuxiaoLi @.***> wrote:

The output of the WAM_para3_newBz run for the 20150316-19 storm has been interpolated and included in the orbit validation plot as attached. The orange curve (WAM_new_newBz) indicates the performance of the new SW algorithm. we can see it makes big progress than the current SW algorithm (WAM_new_ori_derive), although still kind of far from the satellite data and the run with the observational driver input. Since the magnitude of the difference between the new algorithm and the operational output shown in github issue 65 is about 2 orders less than the global mean neutral density (E-13.vs.E-11) and all other validation, I think the results are all good for the merge of the newBz branch into develop.

[image: Den_20150316_storm_para3_newBz_ori_msis2_GRACE_orbit_mean300_6lines] https://user-images.githubusercontent.com/22546571/213790163-434d8390-837f-4447-8356-e17b924a745c.png

— Reply to this email directly, view it on GitHub https://github.com/NOAA-SWPC/GSMWAM-IPE/issues/65#issuecomment-1398842492, or unsubscribe https://github.com/notifications/unsubscribe-auth/AH5BFODUIMCAIB5XVQDGVRLWTLSELANCNFSM5QWNEVCQ . You are receiving this because you were assigned.Message ID: @.***>