These parameters are discussed in the online help for FLAASH.  Is there specific additional information you need?  Since Quickbird data does not have an appropriate spectral band to perform an aerosol retrieval, you should set your aerosol retrieval to None.

1) We did set the aerosol retrival to none and used the aerosol model (rural and maritime). The reflectance value of 4 bands (QB) are zero. Any further advice? 2) In the QB radiance process, the output is for example "QB Calibrated Radiance (Band 1) (0.4850)" when we put "NO" for scale output to integers. This output can use to be input file to FLAASH process. And the output is  "QB Calibrated Radiance (Band 1)" when we put "Yes" for scale output to integers & Scale factor 1000. This out put cannot use to be input file to FLAASH process. Any advice?    

Setting aerosol retrieval to none would not cause the output to be blank.  Did you make any changes to the output scale factor in the Advanced parameters?  FLAASH always outputs scaled integers so if you set it to 0, you will get an all-0 image.  It is best to leave the output scale factor at 10,000 to allow for a reasonable dynamic range. The scaled integers from the QB radiance tool can be used in FLAASH.  You need to set a scale factor of 1000 when importing the file.  I would suggest using scaled integer over floating point values to reduce the file size. If you are getting specific errors, can you post them?

Thank you for your advice but the output still be blank.  I post the figure of Flaash setting and output follow this website. http://i255.photobucket.com/albums/hh134/nuttorn_nut/flaashsetting.jpg and http://i255.photobucket.com/albums/hh134/nuttorn_nut/flaashoutput.jpg. would you please suggest me again?Thank you   

What are the statistics on the input file?  Is the elevation really 75km?  You might need to contact technical support directly for further help with this.

Dear mminari Thank you for your advice. We have got reflectance value. But some more question, we use step for DN to Radiance (scale factor 1000) and radiance to reflectance (scale factor 1000). The reflectance value is thounsand. Is it correct?

Dear Friend, I encounter the same problem when using Flaash for atmospheric correction of MODIS, may I ask how to solve this problem ? Thanks.

Since you set the output scale factor by 1000, then a reflectance value of approximately 0-1 x 1000 would be in the thousands.  You can have some pixels which are below zero or above 1, but it is not uncommon to have a poor retrieval for some pixels.

MODIS is very different and requires some pre-processing prior to input to FLAASH.  It is important to make sure bands in the input file match what is used in the filter function.  You need to also be sure to not include the thermal bands from MODIS in a calculation.