XPost: alt.astronomy   
      
   We've looked at several datastreams coming from the TESS telescope   
   that's in an orbit between the earth and moon.   
      
   If objects are coming and going in the region we assume some might   
   block some of the 5000 stars being monitored by the telescope for   
   planetary transits and if several stars in the same region of the sky   
   seem to dim (or brighten) together we may have detected an object   
   moving between the telescope and the relevant target stars.   
      
   We produced movies showing the light variations seen by TESS using the   
   central pixel for each of its target stars, a processed version of   
   that data supposedly ready for use to detect planetary transits, and   
   also the "background" around the stars being targeted.   
      
   It turned out the background regions had more information than the   
   others. But since any supposed object moving around maybe did not   
   cross directly in front of a bright star -- one of those on the TESS   
   target list -- it was likely to be easier to see illuminated or bright   
   objects than dark ones. And that seemed to be borne out by the movies.   
   (Bright flashes outnumbers dark ones in the "background" movie but   
   were more equal in the "central pixel" movies).   
      
   But the next link in a long chain, one actually established before the   
   movie making process was started, is to show which sections of the sky   
   correlate with what kind of UFO activity, if any, reported in the   
   skies over N Am.   
      
   After updating the local copy of the UFO sightings database from the   
   latest UFO reports from the NUFORC to the end of August 2022 we now   
   have 2 different kinds of UFO reports that can be checked against the   
   light variation in each section of the telescope's sky -- the overall   
   UFO sighting numbers from NUFORC, and the "MADAR" sightings reported   
   in the NUFORC database but apparently related to a project originally   
   from MUFON that established a network of magnetic anomaly detectors   
   across the US that supposedly detect certain types of unusual activity   
   that has been informally linked with parallel sightings of unusual   
   lights and other objects in the sky.   
      
   For each 10x10deg "tile" of the sky that was produced for the various   
   movies previously discussed we can then go tile-by-tile and discover   
   how statistically close the various UFO sightings types are to the   
   sometimes erratic increasing and decreasing of brightness of multiple   
   stars allocated to the tile.   
      
   It turns out around 1/2 the sky seen between the earth and moon -- out   
   of around 650 tiles in all -- is very strongly correlated with UFO   
   activity over N Am.  By "highly correlated" we mean a time series   
   regression that allows for various effects related to serial   
   correlations in the data shows that 2 key statistical tests pass at   
   better than 95% confidence.  Each square must show a "dose   
   relationship" between the brightness of that tile and daily count of   
   UFO sightings of the relevant type.  And the ordering of data by   
   brightness must be the essentially the same as the ordering of the   
   data by UFO activity.  Together these tests should push the   
   possibility of a "false positive" down to 1 chance in 1000 or even   
   less. IOW, of the 650 tiles we monitor only about 1 tile might be   
   shown as a false positive in each time step -- the majority should be   
   "correct". (There is a quibble that some of them will be "false   
   negatives" but we conservatively assume anything that can not be shown   
   to correlate just does not correlate).   
      
   We've produced some "maps" to parallel the movies previously   
   discussed.  3 density plots show the "R2" of the time series   
   regression between each tile's background brightness and different   
   types of UFO activity.  The 3 types we've plotted to date are "all   
   types of UFO activity, including possible mistaken sightings", "MADAR   
   reports of UFO activity", and a 3rd map that takes the points of   
   highest R2 from each of the first 2 maps to show a "either MADAR or   
   overall activity, whichever is more likely".   
      
   These 3 maps are presently uploaded at   
   ,   
   ,   
   and .   
      
   In each plot the colour key is on the right. Tiles in the sky with no   
   correlation with the relevant activity are shown as yellow ("R2=0").   
   As the relevant R2 increases to its maximum the colour goes through   
   green, blue and violet.  The violet points are hard to see because   
   they are relatively small and few in number. But they must be there or   
   the scale would be truncated by the plotting package (an ancient   
   version of GNUPLOT).   
      
   Perhaps more interestingly we can make similar maps based on the   
   "lags" introduced to maximise the various R2's.  While the 2 types of   
   data -- brightness curve from some tile, and UFO activity of whatever   
   type -- are known to significantly correlate we can make the R2 a   
   little larger, meaning the 2 curves appear even more similar, if we   
   time-shift the light curve by some number of days. The shift can be   
   left (earlier) or right (later in time). A -ve shift means the UFO   
   activity changed first and the tile changed brightness some number of   
   days later.  A +ve shift means the brightness changed first and the   
   UFO activity changed a number of days later. We can interpret -ve   
   shifts as tiles where unknown objects have gone after being seen on   
   earth, while +ve shifts are the possible origin of objects that were   
   seen later.   
      
   As before we can plot 3 different types of data -- "all UFO activity",   
   "UFO activity related to magnetic anomaly detection" (aka MADAR), and   
   both types of activity together, selecting the most likely (higher R2)   
   for each tile.   
      
   Those 3 maps are presently uploaded at   
   ,   
   ,   
   and .   
      
   While the AI's already have some overall ideas of how to generalise   
   what the various plots show in a simile way -- they are programmed to   
   find the "simplest hypothesis" that explains the greatest part of the   
   data they have available at the time -- I'll leave it as an exercise   
   for a little while for readers to try to interpret what the plots show.   
      
   If any reader wants to try to reproduce the same results, or their own   
   results, the TESS output is also available at   
      
   as a gzip-ed tarball.   
   (Sorry, folks, I'm UNIX-oriented and too old to change).   
   Each file in the tarball is the avg brightness for a tile of the sky.  The   
   filename gives the central RA and Dec of the tile. The timesteps are days   
   from the date origin of the TESS project. We can assume here that is   
   01 Jan 2015.   
      
   --   
   "Nothing in life is to be feared, it is only to be understood.   
   Now is the time to understand more, so that we may fear less."   
   - Marie Curie   
      
   US Navy Admits It Has More UFO Videos, But Don't Expect to See Them Anytime   
   Soon   
   ScienceAlert, 11 Sep 2022 03:11Z   
   The US Navy holds unseen videos of unidentified flying objects (UFOs) - or   
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)