[GRASS-user] How to deal with the warning concerning Esun during i.landsat.toar job?

[Kozlova Maria]

Dear Nikos,
Thank you very much for detailed answer and references! 
I know the documents from usgs web site, though I was afraid I could miss some important information...

Such a way, this warning is not meaningful in the case of Landsat 8 OLI sensor imagery, if so can this warning be removed in latest Grass versions not to frighten the beginner users?

Kind thanks again,

-- 
Best regards,

Maria V. Kozlova

  /------------------------------------\
  Information System Lab. (ISysLab)
  State Oceanographic Institute,
  Kropotkinsky per., 6,
  MOSCOW, 119034,
  RUSSIA 
  ------------------------------------
  tel.    +7 499 246 6448
  fax.    +7 499 246 7288
  mailto:  kclo at yandex.ru 
  \------------------------------------/


26.05.2015, 13:40, "Nikos Alexandris" <nik at nikosalexandris.net>:
> Kozlova Maria:
>>  Dear colleagues,
>
> Hi Maria,
>>  I'm working with Grass Gis 6.4.4. and  7.0 (GUI interface).  I calculate TOA-reflectance with i.landsat.toar module. But some time I launch the module, the following warning appear.
>>
>>  WARNING! ESUN EVALUATED FROM REFLECTANCE MAXIMUM BAND
>
> Please read the "On Landsat-8 metadata file" section in i.landsat.toar's
> manual. The manual explains how it compensates for the not-provided ESUN values
> for Landsat8 imagery. Hopefully this answers all of the following questions.
>>  Some of my colleagues are sure that I shouldn't pay any attention to this warning as the output TOA-reflectance data is always the same with the data obtained by entering ESUN value for many routine kinds of work. Although they are not sure that it is acceptable for any special image processing, such as DOS atmospheric correction etc. And the problem is that my research concerned with many experiments of various image processing to get acceptable results.
>>
>>  And so I need reliable TOA reflectance data.
>>
>>  My question are:
>>
>>  1. Can anyone tell if I can get appropriate TOA-reflectance data without entering Esun value? Can this parameter significantly influence the output data? In what cases?
>
> But where is the user required to enter it?  It is fetched/computed from the
> other metadata (from the corresponding metadata file of the acquisition). I
> think the ToA-Reflectances derived by the module are appropriate.
>
> (
> ESUN is the mean solar exoatmospheric irradiance for a given spectral band. It
> is part of the formula converting radiance to (planetary) reflectance. Its
> calculation involves the relative spectral response of a band and the solar
> irradiance over wavelength [0]. Thus, it significantly contributes to the
> output data.
>
> As far as I know, it is required for the conversions of data from most sensors
> (Landsat 4, 5 & 7 [1], QuickBird2, IKONOS, WorldView -- yet, not for Landsat8 [2]).
>
> )
>>  2. How can I enter Esun value in i.landsat.toar module?
>
> You can't.  Not sure for ETM+ data though, if you could do anything different via the `gain=` option.  Normally you should use the provided metadata.
>>  3. Why the warning appear not every time but after the same actions and it depend on certain image (problem occur mostly with Landsat OLI images)?
>
> The "official" OLI conversion formula does not require it. Please read more at
> [2]. Thus, they don't provide these values.
>
> However, the i.landsat.toar module requires it in order to apply well
> known conversion formulas, as well as the simple DOS atmospheric correction
> methods.  And, it gets these values by assuming that the provided maximum
> radiance corresponds to the provided maximum reflectance and some math.
>
> Nikos
>
> [0] For example, for IKONOS bands see https://apollomapping.com/wp-content/user_uploads/2011/09/IKONOS_Esun_Calculations.pdf
> [1] http://landsathandbook.gsfc.nasa.gov/pdfs/Landsat_Calibration_Summary_RSE.pdf
> [2] http://landsat.usgs.gov/Landsat8_Using_Product.php