why the atmospheric correction result is bigger than input apprant reflectance?

[bodhi fan]

Hi, is there anyone check the corrected result of 6s? It makes me confused.

Here is a sample result:

*******************************************************************************
*                        atmospheric correction result                        *
*                        -----------------------------                        *
*       input apparent reflectance            :    0.100                      *
*       measured radiance [w/m2/sr/mic]       :   42.653                      *
*       atmospherically corrected reflectance                                 *
*       Lambertian case :      0.32791                                        *
*       BRDF       case :      0.32791                                        *
*       coefficients xa xb xc                 :  0.00974  0.08017  0.06864    *
*       y=xa*(measured radiance)-xb;  acr=y/(1.+xc*y)                         *
*******************************************************************************

so, when pixel radiance is 42.653, the corrected result will be 0.327, calculated as below:

========================================

xa=0.00974  

xb=0.08017  

xc=0.06864

# 0.00280  0.03639  0.05079

def getCorrected(radiance):

    y=xa*(radiance)-xb

    # print(y)

    acr=y/(1.+xc*y)

    return acr

getCorrected(42.653) # it return 0.3277282273777346

=======================================

This doesn't make sense. Because the input apparent reflectance also contains the contribution from Rayleigh and aerosol. Can someone explain this, please? Thank you very much.