There are 5 sets of data being looked at here:
The original set of 500 showers is taken from this list. The set starts at 03-01-04 UTC 3160973338 and ends at 05-15-04 UTC 3167536930. The total data set is taken from this list. The shower numbers corresponding to the orignal time interval are 12003 through 17296.
Hand-reconstructed data:
Auto-reconstructed data:
Older auto-reconstructed data that has been adjusted for temperature effects:
Data set of 16076 showers (auto-reconstructed, not calibrated):
Auto-reconstructed data vs. hand-reconstructed data:
Auto-reconstructed data (total set, scaled to 500) vs. hand-reconstructed data:
Auto-reconstructed set of 16k vs older set of 500:
The above difference reflects the move to analysiscut2 in selecting showers.
Doing this made me realize that I should be comparing a set time interval rather than a set number of showers. So I took all the showers found by analysiscut2 during the original time interval for the set of 500 showers.
Scaled data histogram (green is the set found by analysiscut2):
There is not much difference above 10^18 eV:
Auto-reconstructed data vs. hand-reconstructed data:
Auto-reconstructed data vs. calibrated (auto) data:
Auto-reconstructed, calibrated data vs. hand-reconstructed data:
