Please contact Prof. Robert Holzworth, Director of WWLLN if you have any questions.
During this last year (2007 through Feb 29, 2008) we have been generating and distributing WWLLN data which is represented by the data on the left (a two week plot from January 2007). There is a haze stands out strongly in a plot like this (but does not strongly influence a density plot). We have now fixed the 'haze' which was caused by a programming problem in the old algorithm code. We now have a completely new code, written in C++ by Dr. James Brundell, rather than in Matlab, which has been used to generate the plot on the right for the same time period. All WWLLN distribution data, and all web presentation data begining March 1, 2008 are using this new code (on the right).
The new distribution is much cleaner. The entire WWLLN data set is being reprocessed with the new algorithm code. Only the WWLLN data distributed in 2007 until Feb 29, 2008 had this haze problem. Anyone who has purchased a 2007 DVD can request a replacement with the new code processed data. Data from 2006 and before did not have the haze, but the new code finds nearly double the good data points from the earlier years. (You will note that the new code produces fewer 2007 total points, than the old code by about 30% (as in the plots below), most of which can be attributed to the haze and other old code problems visible in the left hand plot. This 30% reduction applies only to the 2007 data, while the older data from 2006 and before, will show a great increase in good data points using the new algorithm.)