EXECUTIVE SUMMARY:   
   - We use N2O as a proxy of historical UFO activity.   
   - Comparing atm N2O data for the past 2000 years obtained from ice   
     cores in Antarctica we see a variation in N2O that seems to   
     correspond with planetary positions, notably Uranus.   
   - We recall modern UFO sightings/activity also has a strong   
     correlation with certain planetary positions, incl Uranus.   
   - It may be modern UFO activity goes back a lot longer than 70y.   
   - N2O data now goes back 800k years and, briefly, the same kind of   
     patterns are also observed in that data although with less certainty   
     as to when it might have started.   
      
      
   In previous posts we've seen there seems to be a strong relationship   
   between UFO activity and ocean plankton. In one post I showed UFO   
   activity (we use the sightings database from NUFORC as a proxy of   
   "global UFO activity") is predicted by plankton abundance.  It seems   
   when plankton is blooming in the N Hem UFO activity over N Am is high;   
   when plankton is blooming in the S Hem UFO activity over N Am is   
   low. Even more detailed analysis suggests the approx 6 major blooms   
   around polar regions are "predictive" (aka "responsible") for   
   shuffling UFO activity to different parts of the world at different   
   times. Given the complex structure of the blooms -- each polar region   
   sees one major bloom and 2 minor blooms each year; with the seasons in   
   the Arctic and Antarctic offset by several months -- it seems almost   
   impossible a corresponding variation in UFO sightings can not be   
   causally related. While some of the variation is likely related to   
   weather across the US -- and therefore e.g. related to who often sit   
   out at night looking at the sky -- that can't explain all 6 bumps in   
   the relevant datasets.   
      
   And given plankton are responsible for some part of the variation in   
   atmospheric N2O we may have a way to decide whether UFO activity is a   
   recent thing or goes back 100s, 1000s or possible 100s of 1000s of   
   years.  We have data from ice cores that nowadays plot N2O over the   
   past 800ky.   
      
   We've looked at N2O data and UFO activity before. But, briefly, we can   
   establish that atm N2O is highly predictive of UFO sightings across N   
   America (and, we assume, "therefore" global UFO activity).  Using the   
   NUFORC data and modern N2O measurements we can perform the following   
   timeseries regressions:   
      
   N2O dataset	UFO dataset	Lag	Filt	Transf	R2   
   				(m)	(sd)   
   mn2o_brw	mvufo2020	12	1.5	-	0.49964528   
   mn2o_Global	mvufo2006	12	1.5	log	0.68380828   
      
   These are the models with highest R2s for the different N2O and UFO   
   dataseries.  The 2006 series is the NUFORC dataset as monthly counts --   
   with variance adjustment to allow for possible methodology change   
   between the 1st and 2nd halves, and missing values substituted by the   
   relevant means -- prior to 2006 when a web reporting form was implemented.   
   The change saw major changes in patterns of reporting that may make it   
   difficult to robustly combine the pre- and post-2006 parts of the   
   dataset. The "2020" series is the post 2006 data upto the end of 2020.   
      
   We see the 2006-2020 data is about 50% explained by atm N2O as   
   measured at Barrow AK. It turns out different locations where N2O are   
   measured differ, sometimes significantly, in their correlation with   
   UFO activity.   
      
   We also see the ~1940-2006 data is 68% explained by atm N2O averaged   
   over the whole globe (the average of available stations during each   
   year of the N2O data).   
      
   The best correlation was found in both cases with a 12m lag.   
   I.e. relating N2O data against UFO data in the same month of the NEXT   
   year.  Why there is a lag we might speculate relates to the seasonal   
   nature of plankton in different parts of the world (probably most   
   significantly in the Antarctic) vis a vis observation mostly occurring   
   in North America.   
      
   So the modern link between N2O and UFO activity seems clean.   
      
   In this post we'll then look at the last 2000y via the N2O data from   
   the "Law Dome" in Antarctica (a reading for almost every year in the   
   past 2000). In a subsequent post we'll look at the N2O data from the   
   EPICA Dome C that goes back 800ky (unf a density of only around 1   
   reading per 8000 years -- a bit of a challenge to handle for this kind   
   of work).   
      
   While the various relationships we'll discuss below are all determined   
   to be highly statistically significant via 2 tests we can't claim at   
   this point anything has been shown to any significant degree of   
   certainty. All these data studies are works in progress -- something   
   that characterises most applied data science in industry, commerce and   
   science. But we can claim they are highly suggestive and point at   
   areas we might investigate further with other tools.   
      
   The argument will run thus: UFO activity is predicted by seasonal   
   plankton blooms. Plankton blooms produce additional atm N2O.  Both UFO   
   activity and atm N2O are predicted by the movement of certain   
   planets. Looking at the historical N2O data from 1AD to the present   
   () we can see at least 2 features.   
   (a) There seems to be a periodic rise and fall in atm N2O before   
   e.g. 1000 AD.   
   (b) There is a very rapid up-tick in atm N2O starting around 1800.   
   Climate scientists ascribe (b) to human intervention and the   
   Industrial Revolution and modern industrial-scale agriculture and   
   widespread fertilizer use.   
      
   The (a) feature is more interesting. Given even in modern times there   
   is an annual variation underneath the huge trend in N2O increase, we   
   might suspect it could indicate something about UFO activity. Is the   
   waveform from e.g. 1AD to e.g. 1000 AD related to "something" or is it   
   just a random-walk of some kind?   
      
   It turns out -- no, it is not totally random.   
      
   Wouldn't it be interesting if the positions of e.g. the outer planets   
   highly correlate with atm N2O from centuries back?  Exactly how the   
   distance between the Earth and Saturn might influence how much   
   plankton grows in the ocean might be suspected ("farming"), but   
   regardless of suspects it would be interesting.   
      
   We can test each planet in turn to see how closely it's e.g. position   
   in its orbit or distance from Earth correlate with the atm N2O we can   
   see in the record.   
      
   I'll assume here a sinusoid of the relevant frequency is a good enough   
   proxy of planetary positions over centuries to draw conclusions from   
   the results. The alternative is to develop very high precision solar   
   system simulations that are valid over 1000s of years; something at   
   least this author doesn't have that much time to write/adapt and debug. :)   
      
   The periods of the relevant planets are as follows:   
   Jupiter 11.86y;   
   Saturn 29.46;   
   Uranus 84.01;   
   Neptune 164.8y.   
   Setting up the relevant synthetic positions, including a time-shift   
   adjustment to maximize the relevant matches, we find the R2's of   
   appropriate timeseries regressions across the 2000y N2O dataset are as   
   follows:   
      
   Period	Shift	R2   
           (y)   
   84.01	4	.9988   
   164.8	23	.1736   
   29.46	66	.0104   
   11.86	0	.0039   
      
   A 2nd code made slightly different calculations and found the   
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)