I would like to create a program using MATLAB which automatically gives the salinity value to the user. I have gathered all the algorithms to calculate the dielectric constant of seawater and trying to iterate the value of TB measured (from SMOS) and TB simulated using bisection method until get the value of salinity.

My problem is the coding of iteration did not work correctly. Even I change the value of TB measured, the iteration still gives the same value of Salinity. I am wondering if the mistakes either because of my iteration coding or the iteration cannot be run using bisection method. I am kindly request if you can share some ideas for solving this problem.Thank you for your help.

Thank you for your suggestion....btw, i choose to apply the most preferred method which is Lavenberg Marquadt...This iteration will continuously occur to minimise the cost function (TBmeas-TBsimulated) until the value nearest zero can be accepted as the parameter's value required.

From my understanding, this method will tune the parameters of SSS, SST and U(wind speed). Is it these parameters to be tuned is from equation of:
i- Flat surface : SSS,SST (in the function of emissivity)
ii- Rough surface (semi-empirical method): U (wind speed)

I am confuse because the parameters to be defined is not straightforward... especially in the dielectric constant of Klein and Swift model which consist a bundle of equation as function of SSS and SST...Can you briefly explain?or the SSS and SST in this model was given based on other ancillary data?


Futhermore, in the Klein and Swift model,the bundle of equation for dielectric constant's consists the function of SSS and T..and the parameters to be tune is SSS,SST and U. May I know if the function of T in the Klein and Swift model and SST which is to be tuned is the same parameters to be defined or it is different? Really hope somebody can clarify this...thank you so much.