*********************************************************************
    Derivation of the SXT Total Solar Irradiance  (June-2008)
*********************************************************************
 
The radiation of the sun at X-ray wavelengths has been used since the dawn of 
the space age as a measure of solar activity. Solar X radiation is variable 
over all timescales down to a fraction of a second and over several factors of 
10 in intensity, depending on wavelength and level of activity.
Unfortunately, knowledge of solar spectral irradiance at X-ray wavelengths is
poor. With a few exceptions, solar X-rays have been monitored with broadband
sensors with oly coarse spectral discrimination, such as the ionization 
chambers on the Geostationary Operational Environmental Satellite (GOES) and 
earlier spacecraft. 

We derived the solar X-ray irradiance from the SXT images, most recently
calibrated in the YLA project, with the similar method used in our previous
study (Acton, et. al, 1999). The brief summary of the method is as follows.

1) Image selection and quality check, i.e., eliminating images suffering
   atmospheric absorpion (satellite dawn and dusk) or high energy 
   particles (radiation belt), during large pointing offset, 
   and other anomalies.

2) Find the pairs of SSC images taken in the Al.1 and AlMg analysis filters,
   then collecting the SXT signals within 1.3Rs for each pair. 
   Figures 3a and 3b shows the spectral and temperature response of the
   two filters used in this study (from Acton et. al, 1999).
      
3) Daily averaging the signals from each pair, and calculate the color 
   temperature (Te) and emission measure (EM) using filter-ratio method 
   Figure 3c shows the the ratios of the above SXT response fuctions as 
   the function of Te (also from Acton et. al, 1999). 

4) From calculated Te and EM, and atomic data (by Mewe et. al, 1985, 1986), 
   calculate total radiance, and then irradiance for the wavelength bands 
   including 0.1-0.8nm.

Figure 4 (solid line) shows the SXT irradiance from January, 1992 to December,
2001 drived in this study. We compared our results with the daily averaged 
GOES/XRS low band (0.1-0.8) flux. They generaly show good correlation, but
our values are 5-10 % lower in logarithmic scale. However, during the
period of solar minimum (roughly Apr-1995 to Aug-1997), when the solar activity 
drops below the detection limit of the GOES low band.

Since over the decades, GOES flux is widely used as a primary source of the 
solar X-ray irradiance, it would be useful if it is corrected for such period
of low solar activity. And since our mothod is also applicable for the low
activity phase, our resluts through the fitting procedure described in below 
could provide the correction to the GOES around the solar minimum. 

For simplicity, we first fit our light curve outside the solar minimum 
to the GOES curve of the same period (by a simple constant shift).
We thereby obtained the constant, 0.44 (in logarithmic scale) that gives the
best fit (Figure 5). Then we applied this constant to our light curve during 
the solar minimum to obtain the corrected GOES low band flux (Figure 6).

As the future work, we are considering the followings to improve the 
reliability of our results.

- proper treatment of the "no data" days.
- better way for fitting SXT with GOES. 
- use of more recent anomic data for the derivation of irradiance.  

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% references %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

"Deriving solar X-ray irradiance from Yohkoh observations",
 Acton, L. W., Weston, D. C., and Bruner, M. E.,
 Journal of Geophysics Research (1999) vol.104, 14827.

"Calculated X-radiation from optically thin plasmas, Paper V", 
 Mewe, R., Gronenshild, E. H. B. M., and van den Oord, G. H. J.,
 Astron. Astrophys. J. Suppl., Ser. (1985) vol.62, 197

"Calculated X-radiation from optically thin plasmas, Paper VI", 
 Mewe, R., Lemen, J. R., and van den Oord, G. H. J.,
 Astron. Astrophys. J. Suppl., Ser. (1986) vol.65, 511

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%