Raw image, false color composite

• melting lakes are quite obvious
• oblique illumination
• eastern regions are badly saturated
• westwards: altitude decreases steadily
• westwards: path radiance increases steadily

• because of the conditions,
• a consistent optical calibration may only be obtained at a very local scale

Raw image, false color composite of study site

Calibration A is not satisfactory

• WL[red    ]=661 nm mid-waveband: 2K=0.803
• WL[green]=561 nm mid-waveband: 2K=0.164

Calibration B is questionable

• WL[red    ]=631 nm is not even sufficient
  • and yields 2K=0.659
• WL[green]=561 nm mid-waveband: 2K=0.164

Calibration C

• WL[red    ]=661 nm mid-waveband: 2K=0.835
• WL[green]=578 nm.........................: 2K=0.252

• in spite of the fact that the shallow bottom ice (blue ice) is likely to exhibit spectral properties somewhat different from those of the alledgedly dirty greyish snow over "dryland"???

The conclusion is quite obvious

• BPL: with the shallow bottom exclusively comprised of ice, we can expect the BPL assumption to be applicable

• Solution A: WL at mid waveband is not sustainable for the green band

• Solution B is not sufficient, and physically unacceptable

• Solution C would appear to be acceptable

Computed depth

• Computed depth reaches ~18 m
  • NIR band was disabled
• Enabling the NIR band would allow to map a network of very shallow gullies
  • this would be a valuable contribution to the monitoring of melt waters using a time series of high resolution images

Average bottom reflectance

• Computed bottom brightness can be locally very high (blue)
  • but it is seen that most pixels exhibit are distinctly less bright (purple to yellow)
  • with some patches of pixels exhibiting just half the maximum brightness