Chapter 5 has described the design and evolution of the LT production PL and shows that the infrastructure used to construct the PL has been able to successfully adapt to new code and methods of reduction without having to be re-built. Chapters 2 and 3 have shown that the infrastructure design of the LT PL has also been implemented into the ING WFS PL and is due to be implemented into future SL software releases, indicating forward thinking used in the PL. This demonstrates the forward thinking design of the LT PL.
The run time of the LT PL is wholly dependent on the number of objects in each frame, the more objects the longer the run-time, due to the larger number of iterations (e.g., object detection and photometry). However the R, I, Z test data, all observed in a one night’s observation run, reduced in ~4 hours, with ~17 objects, reduced in both optimal and aperture photometry, in each of ~90 frames. This time includes running the online PL, which will not be necessary in practice as the frames will have already been preprocessed.
The production PL results match those of the proven prototype PL detailed in Chapter 4. The LT PL has therefore shown that automated data reduction is certainly a viable option to perform the initial reduction necessary, before any, more detailed and individual analysis, should it be necessary.
The mean photometric errors on the optimal photometry is 0.19mags, on the aperture the mean errors is 0.1mags, however he the mean difference between the optimal and aperture photometry is 0.32mags. This discrepancy between is concerning and warrants further investigation. As highlighted earlier this effect may be due to an incorrect profile correction applied to the optimal photometry - an technique that uses data from the brightest stars on a frame, whilst the objects of interest are amongst the faintest stars. Therefore the aperture photometry results may be more reliable for this data.
In Chapters 6 and 8 the PL is used to reduce Strömgren photometric Be star data.
Print
Email
News!