for help about STANMOD

Tina · Post by **Tina** » Fri Nov 28, 2008 9:47 am

Dear?¬
I??m using two-region model to estimate the parameters of Cd transportation. The experiment condition was one pulse input. I hope you can explain these items: can I use MASS=1? What??s the meaning of the item in Boundary Value Problem-Inverse Parameters: Application time? And what??s the difference between this application time and the Previous item: Application time of pulse input?

ntoride · Post by **ntoride** » Mon Dec 01, 2008 12:47 am

Tina:
CXTFIT has an option to include input solute mass in the parameter estimation in case of a Dirac (mass) , step (input concentration), or single pulse input (concentration or application time). This option came from the previous version. However, it is indeed important to detect all the input solutes in BTCs or solute profiles (input mass = observed mass). Since I have never succeeded to estimate parameters properly without having a good mass recovery, I would recommend not to use this option. Nobuo

Tina · Post by **Tina** » Sun Dec 07, 2008 7:14 am

Thank you very much. Can you explain how to estimate the mass recovery?

ntoride · Post by **ntoride** » Mon Dec 08, 2008 8:06 am

Tina: In case of the pulse input with a constant velocity, the input mass is equal to the inlet concentration c_0 multiplied with the duration time t_0. Observed mass of the BTC at a certain depth z is (integral from t=0 to t=infinity c(t, t) dt). If the mass recovery is good, the input mass is equal to the observed mass. Nobuo

Tina · Post by **Tina** » Mon Dec 08, 2008 8:32 am

Dear:
Can you explain these items: if I want to know the mass recovery, should I do the integral by myself? Can I know the mass recovery from the result of the estimation? What's the meaning of Sum(C1*dT) and Sum(C2*dT) in the output file??

ntoride · Post by **ntoride** » Mon Dec 08, 2008 12:40 pm

You probably misunderstood my comment. Mass recovery is for your measurement, not for the calculation. You need to evaluate the total amount of solute you detected in the observed BTC.

Sum(C1*dT) and Sum(C2*dT) in the output file are calculated amount of solutes in phase 1 and phase 2, respectively, as explained in p.79 of the CXTFIT manual.

Tina · Post by **Tina** » Tue Dec 09, 2008 4:15 am

Dear,
Thanks for the very prompt reply. I am still a bit confused: Is the mass recovery equal to the input mass divided by the observed mass? Is the observed mass equal to Sum(C1*dT)? Could you please give me some hints on this?

Thanks!
Regards,
Tina

ntoride · Post by **ntoride** » Tue Dec 09, 2008 3:10 pm

Tina: Sum(C1*dT) is the calculated result. When you estimate the@transport parameter using CXTFIT, the input mass is the condition you need to give. If your measurement is poor and does not detect all the solute at the BTC, CXTFIT cannot fit the BTC well and the parameter estimation would be also poor. Although I kept the option to estimate the input mass from the previous version, as stated before, I would recommend not use the option. Nobuo

Tina · Post by **Tina** » Sun Dec 28, 2008 10:36 am

Dear?¬
Can you send me an article about"Leij, F.J., Dane, J.H., 1992. Moment method applied to solute transport with binary and ternary exchange. Soil Sci. Soc. Am. J. 56 (3), 667¨C 674"? I can't contact with the author. Looking forward to your reply! Thank you very much!
Tina

Tina · Post by **Tina** » Sun Dec 28, 2008 10:52 am

Dear Nobuo,
My e-mail:sunjunna@126.com. Can I use the method of temporal moments to estimate the retardation factor in nonequlibrium deterministic model in order to reduce the number of estimated parameters?
Tina