Comment on the Papers “On the influx of small comets into the Earth's upper atmosphere I. Observations and II. Interpretation”

Abstract

Frank et al. [1986a] report that exceptionally dark spots in the ultraviolet images of the sunlit Earth provided by the Spin-scan Auroral Imager (SAI) on the Dynamics Explorer I satellite occur far more frequently than would be expected from Poisson counting statistics, which ordinarily govern the statistical distribution of counts from a photomultiplier tube.Frank et al. [1986b] postulate that the spots, which are usually confined to a single pixel, could be due to obscuration of the photometer field of view by a population of screening objects.They conclude that these objects may be water vapor clot{ds from a previously undetected, 20 per minute influx of 100-ton icy comets into the Earth's upper atmosphere.We present here a statistical analysis of "dark pixel" events in a data base of 182 images kindly provided by Frank et al. Based on various instrumental and geophysical considerations,Chubb [1986] argued that these events are probably instrumental artifacts. Theoretical analysis provides a quantitative basis for Chubb's considerations regarding the overlap of adjacent pixels along scan lines in the image, and shows that pixels adjacent to the very dark ones should themselves be darkened by at least 9% on the average, assuming that screening objects cause the events. In our data base, however, the adjacent pixels are not darkened beyond statistical error bars, which are less than 1% for a data set containing •500 dark spots. Frank et al. [1986a] Because the fields of view for successive pixels overlap, the GSO screening of the field of view required to explain an anomalously dark pixel turns out to imply a partial screening of the fields of view for the pixels immediately preceeding and following it in the scan line. Consider, for example, screening by a long absorbing band of width W and optical transmissivity T, aligned perpendicular to the scan direction, but exactly centered on one pixel's field of view. Detailed calculations that include the exposure-time weighting effect show that if T -0 and W is chosen so as to yield a 50% reduction of the expected count in this pixel, then W turns out to be 0.14 ø and the expected average count reduction in both adjacent pixels is 5.9%. If T -0.5 and W _• 0.52 ø, then the average center-pixel count would again be reduced by 50%, but the two adjacent pixel counts would both be reduced on average by more than 25%. Typical anomalously dark pixels have count reductions significantly greater than the 50% value assumed above for illustration.We have also considered screening due to a more geophysically reasonable class of GSO's, circular or spherical objects of various sizes and zero transmissivity with random positions 573

Small‐comet “atmospheric holes” are instrument noise

“…It presents simple, direct, evidence that the dark pixel clusters in the Iowa catalog are the tail of an instrumental noise distribution-they are not produced by atmospheric holes. This same result was demonstrated by Cragin et al (1987) for the original DE-1 small-comet data, who showed that them was no change in the angular size of the purported atmospheric holes with altitude. Thus, 12 years ago, as today, it is shown that the "atmospheric holes" are insmLment noise.…”

The Small‐Comet Hypothesis

“…The sensitivity of the event rate to lifetime can be demonstrated with an example in which we assume the dark-spot lifetime is 60 ms and the darkness threshold such that 1 in 1000 pixels is dark. (This is approximately the actual ratio of dark to normal pixels seen by the DE imager [Cragin et al, 1987].) When Earth effectively fills the image frame, a DE Earth image consists of~I0 n pixels.…”

“…According to the instrument-artifact hypothesis, dark pixels appear in images of the geocorona simply because one in 800 pixels is dark; the observation becomes an instrumental, not a geophysical, effect. With one pixel in 800 randomly turning dark, a statistical fluctuation yielding two adjacent dark pixels is a reasonable expectation [Cragin et al, 1987]. However, a three-dark-pixel event is extremely rare, and four sequentially dark pixels is an almost impossible random coincidence.…”

“…[1987] and claim that their differences are due to "an incomplete theoretical treatment and an insufficiently large set of data" used by Cragin et al Although the statistical analysis contained in the Comment by Cragin et al [1987] undoubtedly makes for tedious reading, I found their theoretical treatment to be sound. An increase in multiple pixels with decreasing altitude, as called for by Chubb [1986] and by Cragin et al, is such a powerful first-order effect that it must be seen in just a few pictures taken at different altitudes, or the small-comet hypothesis fails in a fundamental way.…”

Atmospheric holes: Instrumental and geophysical effects

“…critics of the reality of atmospheric holes predict that the rates of atmospheric holes will be unaffected by the spacecraft altitude because they are instrument artifacts[Parks et al, 1997;Dessler, 1991;Cragin et al, 1987]. We refer in this paper to the Parks-Dessler-Cragin hypothesis that the atmospheric holes are instrument artifacts.…”