| Fig. 
        4-1 | SCIAMACHY’s 
          scientific observation modes: 1 = nadir, 2 = limb, 3 = occultation. 
          (graphics: DLR-IMF) |   | 
      
      | Fig. 
        4-2 |  
          An orbit with planned limb/nadir matching on the dayside of the orbit. 
          The sequence of nadir and limb states in a timeline is arranged so that 
          limb ground pixels (blue), defined by the lineof- sight tangent point, 
          fall right into a nadir ground pixel (green). At the beginning and end 
          only limb or only nadir measurements are executed. (graphics: DLR-IMF) 
           |   | 
      
      | Fig. 
        4-3 |  
          ENVISAT’s yaw steering, the yaw steering correction of limb states and 
          the resulting SCIAMACHY yaw steering. Between ENVISAT and SCIAMACHY 
          yaw steering an orbital shift of 27° exists which reflects the observation 
          geometry when looking to the horizon in flight direction. (graphics: 
          DLR-IMF)  |   | 
      
      | Fig. 
        4-4 | SCIAMACHY’s 
          monthly lunar visibility occurs between 1 and 2 over the southern hemisphere 
          (lunar phase > 0.5). The hatched area illustrates the limb TCFoV of 
          88°. Visibility at smaller lunar phases over the northern hemisphere 
          between 3 and 4 is not used because it coincides with solar occultation. 
          (graphics: DLR-IMF)  |   | 
      
      | Fig. 
        4-5 |  
          The rising moon seen from a spacecraft in a low- Earth orbit. Differential 
          refraction distorts the lunar disk. (photo: NASA)   |   | 
      
      | Fig. 
        4-6 |  
          SCIAMACHY reference orbit with sun/moon fixed events along the orbit. 
          The events define orbital segments which are filled with timelines. 
          State duration is not to scale. (graphics: DLR-IMF)   |   | 
      
      | Fig. 
        4-7 |  
          The pattern of ground pixels in a nadir measurement for an integration 
          time of 1 sec (left) and 125 msec (right). Only the forward scans are 
          shown. This causes the along-track gaps between consecutive scans which 
          vary in width due to a projection effect. Across-track extent is defined 
          by the integration time while along track the size reflects the dimension 
          of the IFoV with only a small contribution of the integration time. 
          (graphics: DLR-IMF)   |   | 
      
      | Fig. 
        4-8 |  
          Calibration & monitoring scenarios from orbital to monthly timescales. 
          In the top row the individual measurements and their targets, e.g. sun, 
          moon, lamps, are listed. The states used in each calibration orbit, 
          referring to the definitions in table 4-2, are outlined below. All states 
          unrelated to the sun or the moon can be executed several times at any 
          position along the orbit. (graphics: DLR-IMF)  |   | 
      
      | Fig. 
        4-9 |  
          Information flow during timeline execution. Timelines are started by 
          macrocommand and end when the last state in the timeline has run to 
          completion. (graphics: DLRIMF)  |   | 
      
      | Fig. 
        4-10 | Example 
          of the seasonal temporal variability of orbital segments. The time interval 
          between end of SO&C window and start of eclipse varies only slightly 
          over a year (yellow). In the monthly moon visibility periods, the time 
          between end of MO&C window and start of eclipse shows a much higher 
          variation (red curves). The blue segments indicate lunar visibility 
          phases where moonrise occurs on the nightside, i.e. those which can 
          be used for occultations. (graphics: DLR-IMF)  |   |