ScaRaB A2 Output Product

LMD - November , 1995

Detailed Record Structure for A2 Output file

A2 contains data for one day ( zero to 24 UT ) and one satellite.
Each record contains data for 48 seconds or 8 scans or 408 pixel measurements.



Index          Quantity                  Unit    Bits  Scale f. Offset


1         Julian Date ( whole part )      day     32      1        0
2         Julian Date ( fractional part ) day     32 1000000000    0
3         Earth-Sun Distance               AU     32 1000000000    0
4-5       Satellite Position, X            m      32      1        0
6-7       Satellite Position, Y            m      32      1        0
8-9       Satellite Position, Z            m      32      1        0
10-11     Satellite Velocity, X            m/s    32      1        0
12-13     Satellite Velocity, X            m/s    32      1        0
14-15     Satellite Velocity, X            m/s    32      1        0
16-17     Satellite Nadir, Colatitude      deg    16      100      0
18-19     Satellite Nadir, Longitude       deg    16      100   -180
20        Sun Position, Colatitude         deg    16      100      0
21        Sun Position, Latitude           deg    16      100   -180
22        Orbit Number                    ----    16      1        0

23-430    Pixel, Colatitude                deg    408*16  100      0
431-838   Pixel, Longitude                 deg    408*16  100   -180
839-1246  Filtered Radiance, Total     W m-2 sr-1 408*16  10       0
1247-1654 Filtered Radiance, SW        W m-2 sr-1 408*16  10       0
1655-2062 Filtered Radiance, Vis       W m-2 sr-1 408*16  100      0
2063-2470 Filtered Radiance, IR        W m-2 sr-1 408*16  100      0
2471-2878 Viewing Zenith Angle             deg    408*16  100      0
2879-3286 Solar Zenith Angle               deg    408*16  100      0
3287-3694 Relative Azimuth                 deg    408*16  100    -180
3695-3696 A' Coefficient                   ----   2*16    1000     0
3697-3744 Quality Index                    ----   48*8    1        0
3745      Full Record Index                ----   1*8     1        0 
3746-3747 Calibration Reference            ----   1*8     1        0 
3748-3951 Spare                            ----   204*8   

3952-4359 Unfiltered Radiance, SW      W m-2 sr-1 408*16  10       0
4360-4767 Unfiltered Radiance, LW      W m-2 sr-1 408*16  10       0
4768-5175 Outgoing Flux, SW              W m-2    408*16  10       0
5176-5583 Outgoing Flux, LW              W m-2    408*16  10       0
5584-5991 Identified Scene Type           ----    408*8   10       0
5592-6002 Spare                            ----   11*8   


Total                                             11360*8 Bits

Definition of A2 Output Product Parameters

A2 contains data for one day ( zero to 24 UT ) and one satellite and is equivalent to ERBE S8, except the lack of Wide FOV and Medium FOV data.

Whole part of the Julian Date at the beginning of the record.

Fractional part of the Julian Date at the beginning of the record.

Comments on A2(1) and A2(2) : The same date as ERBE is used and provides a continuous time record useful in the processing of regional means. The Julian date is the universal reference to date astronomical events. Its origin is : January 1st 4717 B.C at noon (UT)

Examples :
A2(1)           A2(2)   Calendar date  UT 

2432 550        0.5    JAN  02  1948    00:00:00
2444 000        0.0    MAY  23  1968    12:00:00
2445 000        0.0    OCT  09  1995    12:00:00
2451 544        0.5    JAN  01  2000    00:00:00

in astronomical unit ( 0.98 < A2(3) < 1.02 )

Comments on A2(4....9) :

The unit is : meter. They are relative to the Earth coordinate system defined from:

origin :    earth centre
X,Y plane : equator plane
OX axe:     Greenwich longitude

These positions are used to compute the view angles of the target point A2(2471..3694).

Comments on A2(10...15) :

The unit is : meter/second . Same coordinate system.

These elements are only necessary to interpolate X,Y,Z inside the record period, when the view angles are computed with a better accuracy.

Comments on A2 (16..21) : all in decimal degrees and correspond to same coordinate system as the preceding values

Relative number. Its origin should be close to the first real satellite orbit. It increases by 1 at the ascending node. Although not directly used in the secondary processing, this element is a practical index used in any satellite data comparisons. It is also useful to show the spatial coverage of one orbit without any overlap from another orbit.

Comments : borrowed from the "remote sensing" terminology, pixel stands for the elementary target viewed from the satellite. These colatitude and longitude correspond to the centre of pixel, i-e the point where the sensor's optical axis intercepts the 30 km altitude earth envelop (top of the atmosphere for the ERB study ). They are computed at the level 1 data processing, with orbit bulletin updated every 15 days. The accuracy is assumed to be the half-pixel, i.e. 30 km or 0.3 degree at the equator. Both colatitude and longitude are in decimal degrees and corresponds to the same coordinate system as the preceding elements

in W m-2 sr -1 - Raw measurement of channel 3 after count conversion.

in W m-2 sr -1 - Raw measurement of channel 2 after count conversion.

in W m-2 sr -1 - Raw measurement of channel 1 after count conversion.

in W m-2 sr -1 - Raw measurement of channel 4 after count conversion.

Relative azimuth between the solar plane and the viewing plane. Zero in the forwards plane.

Comments : these 3 angles (in degree ) are relative to the local ( or pixel ) coordinate system defined as following :

 origin : pixel centre 
 oz : vertical
 ox :solar plane

They are computed at the centre of each pixels. This centre is located at an altitude of 30 km, above the Earth's surface defined by an ellipsoid of 6378.160 km and 6356.775 km, respectively for the equatorial and polar axis.

In the initial design, this coefficient had to be adjusted for calibration variations over short periods of time. For ScaRaB flight model 1, it is now assumed constant and equal to 0.845

Due to change in the SW calibration method, most flags are pointed to 'substitution' and have a very low significance. All data for radiances and angles may be used, except if they have default value.

equal to 'one' if the second half of the record is missing. No significance for the users since the concerned data are set to default value.

This number refers to the job used to generate the calibration sets. Low significance for the users

in W m-2 sr -1 Correction for underestimation at the shortest wavelengths, domain where the instrument response diminishes.

in W m-2 sr -1 Subtraction of the SW unfiltered radiance from the Total unfiltered radiance. SW unfiltered radiance is weighted by coefficient A'.

in W m-2 The SW unfiltered radiance is converted into flux, using the view and sun angles, the scene identification and the SW Erbe bi-directional function ( Suttles et al, 1988). A linear interpolation of BRDF between angles is used in order to remove the discrete nature of the angular model TOA : top of atmosphere (30km altitude as in Erbe)

in W m-2 . The LW unfiltered radiance is converted into flux, using the view angle and colatitude, the scene identification and the LW Erbe anisotropic function ( Suttles et al, 1988). A linear interpolation of the anisotropic function between view angle and colatitude is also used.

There are 13 possible values for the whole part

 o unknown scene 
 1 .. 12 combination of 5 geotypes and 4 cloud category
 127 default value.

As in Erbe S8 format, the decimal part gives the data input of the geotype ( from 0 to 4 )