he ERS -1 orbit 
■'nee (both at 11 
dons are due to 
igular difference 
■ytime (see Fig. 
es with angular 
ice temperature 
rences obtained 
iVHRR/ 2 . The 
ar / split-window 
are larger than 
^served is due to 
that the former 
;er temperature 
ow) effects. 
;y at nadir, ( 12 ) 
SR and in situ 
i(0)=0.962 and 
I °C and an rms 
it any variation 
and Tu{6 2 ) = 
and the trans 
in which biases 
rat the angular 
iffer- 
( 12 ). 
( 12 ) 
■ion. The actual 
face has a cover 
, for dual angle 
,ion is very pre- 
the algorithms. 
l-board calibra 
tion targets and is usually mounted onto am airborne platform. Full details of the instrument, 
it’s calibration and application may be found in Palluconi and Meeks (1985) and Kahle and 
Abbott (1986). 
The TIMS was flown over the field site at Hay, mounted aboard the CSIRO F27 aircraft. The 
aim of the mission was to investigate angular effects and the flights were arranged to coincide 
with an ATSR overpass. Nine flight lines were flown at three different heights and along paths 
that ensured a variety of zenith viewing angles would be obtained. The pixel sizes of the TIMS 
data varied between 4 m and 8 m. The total time taken to complete the flights was just over 
one hour (1040LT to 1152LT) and in that time the temperature of the surface changed by about 
5 °C. Figure 7 shows a plot of the average of the in situ data, the temperatures from one of the 
sites and the temperatures derived from calibrated TIMS channel 5 («10.8 /xm), as a function 
of local time (LT, in hours). The TIMS data follow closely the temporal variation of the in 
10.6 10.8 11.0 11.2 11.4 11.6 11.8 12.0 
Time (hours) 
Figure 6 : Temporal development of in situ and TIMS channel 6 («11.7 /xm) 
temperatures measured at the Hay site. W500 is the most westerly measure 
ment site in bare ground. N500 is the most northerly site and is situated in a 
vegetated area. 
situ data with the exception of the value obtained at around 1120LT. This temperature was 
obtained for a zenith angle of about 21 degrees and the depression in temperature is due largely 
to scan angle effects. These effects can be seen quite clearly in Figures 7 and 8 which show TIMS 
channel 6 («11.7 /xm) data for two flights with zenith angles of 5.9° and 20.8° respectively. 
The circles on these images show the locations of the in situ temperature devices and the dark 
patch near the central circle is a ‘cold pixel’ used as a reference point for navigation. This was 
obtained by laying out highly reflective (low infrared emissivity) aluminium foil over a large 
rectangular shaped area about the size of a TIMS pixel. There is considerable limb darkening 
noticeable on the image with the larger scan angle (Fig. 8 ). The image shown in Fig. 7 is also 
generally warmer than that in Fig. 8 and this is due to the fact that the image was obtained 
some twenty four minutes later. By subtracting out the temporal trend evident in the surface 
data («0.064°C min -1 ), the absolute temperatures near the centres of the two images are very 
close. 
The variation of temperature with zenith angle for the image shown in Fig. 8 has been 
plotted at 5 positions across the TIMS scan for which ground-truth data are available. The 
767