Recipe 19. SMART: General Steps for Extracting Staring Mode Spectra
19.1 About this recipe
The present cookbook explains how to extract IRS data for observations done in staring mode.
The cookbook is intended to be used with SMART v8.1.2 or later.
Check the online online documentation for any help on the SMART funcionalities described in this cookbook. For an screenshot example of a basic extraction using optimal please go to the next recipe, which provides a screen shot example of an extraction.
19.2 Data preparation
Create a new project and a new dataset. Remember to save the project after each step!
Import the data into the dataset by browsing for *_{bcd,func,bmask}.fits files in the system.
(optional): Select all the imported files and click on Image operations -> 3-plane data -> Make 3-plane data. This will new create *_bcd3p records which contain the uncertainty and BMASK information for each BCD image.
Cleaning: Select all the sets of {bcd,func,bmask} records (or all the bcd3p records if the data was combined into 3-planes), and click on Image operations -> Clean using IRSCLEAN -> Clean!. Leave the default values and click on Proceed. If the cleaning is not satisfactory, remove the cleaned files and re-perform the operation with a lower mask value but the user must be aware that a too-low maskvalue can result in line clipping.
Co-adding: Select all the cleaned records (cl_*{bcd,func,bmask} or cl_*bcd3p) and click on Image operations -> Automatically combine DCEs for a given ExpID?. This will combine the individual exposures.
19.3 Low-level rogue pixels removal
The data is now cleaned and the individual exposures are combined. However, bad pixels likely remain that were not cleaned. These are low-level rogues pixels that can be removed by subtracting an empty field image.
For this step, it is advised to use 3-plane data (see previous section). Three methods can be used:
Subtraction of a dedicated background observation (advised for high-resolution data): select all the records (including the background) and click on Image operations -> Image combinations & arithmetics -> Multiple subtraction. Then select the records that should be removed from the others.
Subtraction by nod (only for low-resolution data): for each module and order (SL1, SL2, LL1, LL2), select the 2 nods, and click on Image operations -> Image combinations & arithmetics -> Arithmetics. Then type the appropriate subtraction equation.
Subtraction by order (only for low-resolution data): for each module and nod, select the 2 orders, and click on Image operations -> Image combinations & arithmetics -> Arithmetics. Then type the appropriate subtraction equation.
19.4 Spectral extraction
During extraction, the calibration files will be chosen automatically by SMART.
19.4.1 Point-like sources
High-resolution data: select the records and click on Extract -> Full aperture (point-like source).
Low-resolution data:
Before attempting the extraction, it is advised to run the AdOpt algorithm on at least one exposure of each module in order to check the actual extent of the source, the possible presence of multiple sources, and/or the presence of a significant extended background emission. The AdOpt GUI works with sets of 3 records rather than 3-plane data, so if the data is 3-plane at this point, select them, click on Image operations -> 3-plane data -> Split 3-plane data, select the new records and click on Extract -> Optimal extraction (manual). Check the AdOpt help for more details on the use of the GUI.
Two methods are available for extracting the low-resolution data:
Optimal extraction. An extraction is performed that uses the profile of the instrumental point spread function to weighs the data. It usually provides a better signal-to-noise ratio, especially for faint sources. Optimal extraction can be performed with the automatic source finder or with the manual source finder. The automatic optimal extraction will perform with no diagnostics available. A warning message will appear only if the source was found significantly off the intended position. To perform the automatic optimal extraction, select the records and click on Extract -> Optimal extraction (automatic). The manual optimal extraction will open the AdOpt GUI. To perform the automatic optimal extraction, select the records and click on Extract -> Optimal extraction (manual).
Tapered column extraction. A column extraction is used that scales with wavelength to account for the varying FWHM of the point spread function.
19.4.2 Extended sources
High-resolution data and low-resolution data: select the records and click on Extract -> Full aperture (extended source).
19.5 Spectra visualization and analysis
Once the extraction is complete, the spectra will be displayed in IDEA automatically (except if manual optimal extraction was chosen, in which case the spectra are displayed in the AdOpt GUI. They can be exported to IDEA from there).
Displaying the spectra: only the first spectrum is shown by default. To display another spectrum, click on the corresponding checkbox in the Stored datasets window, click on Choose an apply a function, and click on Make prime dataset. To overplot all the spectra, simply click on Merge/overplot all datasets
Order edges: In IDEA, remove the order edges by clicking on Order edges -> Zap order edges. Important note: the new spectra is not yet stored, it can be by clicking on Store Prime.
Defringing: LL1 spectra can require defringing. This can be done by clicking on Defringe Tool.
Numerous functions are available in IDEA, such as line flux measurements, photometry measurements, blackbody fitting, spectra combination, etc... A wrapper for PAHFIT is also present. Refer to the IDEA help for more details.
Copy spectra to project manager: the spectra can be copied to the project manager by clicking on Copy to spectral manager.
Saving spectra: If desired the spectra can be saved in several formats (e.g. fits table, ascii table) in a given output directory. This can be done within the IDEA gui but it is easier performed once the spectra is copied into the project manager, and then clicking on Export/Conver. This step is not needed and the spectra remains in the project manager but the user must not forget to save the project before closing the session.
