Causes of atmospheric blur: Comment on atmospheric scattering effect on spatial resolution of imaging systems Academic Article uri icon


  • A paper by Ben Dor [J. Opt. Soc. Am. A14, 1329 (1997)] concludes that the blur we measured in our experiments was not atmospherically scattered light and that our theoretical model is incorrect because it violates the rules of linearity. Their work is based in part on “lack of raw data” in one of our experimental papers [J. Opt. Soc. Am A12, 970 (1995)]. We present here the raw data measured in the experiments in question, which show clearly the measured atmospherically scattered light. Similar raw data has also been published elsewhere regarding other experiments. We also clarify some rules of linear systems that justify our conceptual approach, which is shown to be similar to that of turbulence modulation transfer function. A review of several dozen experiments and analyses by other investigators all over the world that directly contradict the Ben Dor et al. results and conclusions is presented. The well-known significance of aerosol blur in imaging through the atmosphere from satellites is discussed, and pictorial examples of satellite imagery are shown for different atmospheric optical depths. It is noted that atmospheric point-spread-function analyses in the remote-sensing literature generally neglect turbulence blur altogether and deal with aerosol blur only, which is often called the adjacency effect, and that such phenomena are well supported by many different types of experiments and many different Monte Carlo simulations for many different aerosol and instrumentation parameter situations. The Monte Carlo simulation results of Ben Dor et al. are shown also to contradict everyday reality such as the solar aureole. This wealth of literature by others strongly contradicts the results of Ben Dor et al. and confirms our conclusion that forward scatter of light by aerosols is indeed a significant source of blur in imaging through the atmosphere, especially if atmospheric optical depth is on the order of unity or more. This can be confirmed, too, by any observer looking through binoculars at the moon and surrounding moonlight even on a clear night. A broad system engineering approach involving both aerosol and turbulence blur is called for.

publication date

  • January 1, 1998