pyart
pyart copied to clipboard
ARM-DOE
About the paramters of ZPHI attenuation correction
what are the reference for the default parameters (a_coef/beta/c/d) of ZPHI atteuation ? I can not find those values in the reference of the ZPHI method (Gu et.al 2011)
@rcjackson Do you know the reference for the default coefficients that sysumeteo is referring to?
It looks like the default parameters are grabbed from a function:
def _param_attzphi_table(): """ Defines the parameters of Z-Phi attenuation estimation at each frequency band. Returns ------- param_att_dict : dict A dictionary with the coefficients at each band. """ param_att_dict = dict()
# S band:
param_att_dict.update({'S': (0.02, 0.64884, 0.15917, 1.0804)})
# C band:
param_att_dict.update({'C': (0.08, 0.64884, 0.3, 1.0804)})
# X band:
param_att_dict.update({'X': (0.31916, 0.64884, 0.15917, 1.0804)})
return param_att_dict
However, there is no reference for them. This was the new method added by MeteoSwiss. @meteoswiss-mdr , do you know where these values come from?
I do not recall exactly where they come from originally but they should be consistent with the article of Ryzhkov et al. 2014. They have been used internally at MeteoFrance and MeteoSwiss.
I suspect they used simulated DSDs from somewhere. I found a figure in Ryzhkov's paper showing their simulated values for alpha. The black dots represent 0 degrees C, the grey dots are 20 degrees C. It looks like the alpha values are in the ranges that would be expected. I don't really know of a way to verify the other values, which is troublesome for reproducibility.
Perhaps we need a constants section with this data in there and well referenced to various papers...
@rcjackson can you do two things so we can close this out:
- Add a notebooks showing application to X,C and S band in https://github.com/ARM-DOE/pyart/tree/master/notebooks (We will eventually binder this) .. Data don't have to be included but orderable (so XSapr, C-SAPR and NEXRAD)
- beef up docs
Sorry I cannot be more helpful here. For sure I used Ryzhkov paper for most of the coefficients (derived from disdrometer data). Others come from empirical measurements performed at MeteoFrance. They were collected in internal reports from MeteoFrance.
[http://chuvaproject.cptec.inpe.br/portal/pdf/relatorios/a2anexo3a7.pdf]
The coefficient values appear to be referenced from the linked site papers.
The title of the final thesis is as in the image. There are also differences in the coefficient values in the paper.
Seems the defaults are a reference to: Ryzhkov et al. Potential Utilization of Specific Attenuation for Rainfall Estimation, Mitigation of Partial Beam Blockage, and Radar Networking, JAOT, 2014, 31, 599-619.
With discussion with rcjackson the values change between midlatitude vs tropical which could explain the difference in the Chuva campaign and the paper mentioned in Py-ART's references
Are they hardset in the method? Can we make them keyword parameters? This is always going to be a challenge in Py-ART. Thank you for bringing this up Mo-Sun-Jin From: Zach Sherman ***@***.***>Date: Wednesday, February 2, 2022 at 1:40 PMTo: ARM-DOE/pyart ***@***.***>Cc: Scott ***@***.***>, Mention ***@***.***>Subject: Re: [ARM-DOE/pyart] About the paramters of ZPHI attenuation correction (#884)Seems the defaults are a reference to:Ryzhkov et al. Potential Utilization of Specific Attenuation for RainfallEstimation, Mitigation of Partial Beam Blockage, and Radar Networking,JAOT, 2014, 31, 599-619.With discussion with rcjackson the values change between midlatitude vs tropical which could explain the difference in the Chuva campaign and the paper mentioned in Py-ART's references—Reply to this email directly, view it on GitHub, or unsubscribe.Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you were mentioned.Message ID: ***@***.***>
@scollis Users can set the coefficients, those values are defaults that are used if not specified.
Cool. Just catching up2nd Bobby’s suggestion From: Zach Sherman ***@***.***>Date: Thursday, February 3, 2022 at 3:33 PMTo: ARM-DOE/pyart ***@***.***>Cc: Scott ***@***.***>, Mention ***@***.***>Subject: Re: [ARM-DOE/pyart] About the paramters of ZPHI attenuation correction ***@***.*** Users can set the coefficients, those values are defaults that are used if not specified.—Reply to this email directly, view it on GitHub, or unsubscribe.Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you were mentioned.Message ID: ***@***.***>