Sensor saturation is the condition in which the radiance reaching a sensor band exceeds the range the band can record, so that its output no longer varies in step with the true signal.12
One source defines a maximum valid radiance as the level above which the linear relationship between the recorded digital number and calibrated radiance no longer holds. The same paper notes that a sensor’s dynamic range is often fixed, so a more sensitive band tends to saturate at a lower radiance, and design has to balance sensitivity against range.1
Bright surfaces are the usual cause. For Landsat Thematic Mapper and Enhanced Thematic Mapper Plus visible bands over snow, one study reports that a saturated pixel records a reflectance lower than the true value. It adds that a later Landsat instrument with a 12-bit data stream avoided saturation in that case, and the instrument page describes the later instruments as recording more bits per sample than the earlier ones.23
No general saturation threshold is established across the sources, and bit depth is not the same as saturation radiance, which depends on the gain and range chosen for each band. The evidence retained here comes from ocean-colour and Landsat examples, so it should not be read as a statement about every instrument.123
Footnotes
-
Hu et al., “Dynamic range and sensitivity requirements of satellite ocean color sensors: learning from the past”, Applied Optics, 2012. https://ntrs.nasa.gov/api/citations/20140006591/downloads/20140006591.pdf ↩ ↩2 ↩3
-
Frontiers in Remote Sensing, “Evaluation of VIIRS and MODIS Snow Cover Fraction in High-Mountain Asia Using Landsat 8 OLI”, 2021. https://www.frontiersin.org/journals/remote-sensing/articles/10.3389/frsen.2021.647154/full ↩ ↩2 ↩3
-
NASA Landsat Science, “Operational Land Imager (OLI)”. https://landsat.gsfc.nasa.gov/satellites/landsat-8/spacecraft-instruments/operational-land-imager/ ↩ ↩2