LBT-SCIDAR Status Update

Robert A. Weiss and Stefan Hippler

September 2001

Robert has worked over the past three months on the automatic SCIDAR auto-correlation analysis, which is now working fine (i.e. in 95% of all cases)


Users can now choose between the Maximum Entropy Method (MEM), conjugate gradients, or CLEAN algorithms for auto-correlation data inversion


The previously slow MEM inversion has been fixed and runs now on the order of a few seconds per inversion. CLEAN takes much longer; probably due to IDL restrictions. The conjugate gradients method always takes about 30 seconds per inversion because the number of iterations is fixed (around 20000)


Results from MEM and conjugate gradients agree to within a few percent; CLEAN seems to have some problems with generalized SCIDAR modes


Robert is currently working on automatic SCIDAR cross-correlation analysis


The SCIDAR data reduction computer will be ordered in October 2001


Robert gave a talk about our combined SCIDAR/ALFA measurements in Toulouse (SPIE conference). The conference paper can be found here