GIRAFFE Exposure Time Calculator
Note: These tools are only provided for the technical assessment of feasibility of the observations. Variations of the atmospheric conditions can strongly affect the required observation time. Calculated exposure times do not take into account instrument and telescope overheads. Users are advised to exert caution in the interpretation of the results and kindly requested to report any result which may appear inconsistent.
See also: Frequently Asked Questions
The GIRAFFE ETC is an exposure time calculator for the ESO optical multiple-target single-order echelle spectrograph. The HTML/Java based interface allows to set the simulation parameters and examine interactively the model generated graphs. The ETC programs allow easy comparison of the different options relevant to an observing program, including target information, instrument configuration, variable atmospheric conditions and observing parameters. As the ETCs are maintained on the ESO Web servers, they provide the most up-to-date information reflecting the known performance of ESO instruments.
These programs provide an HTML/Java based interface and consist of two pages. The observation parameters page presents the entry fields and widgets for the target information, expected atmospheric conditions, instrument configuration, observation parameters such as exposure time or signal-to-noise, and output selection. An "Apply" button submits the parameters to the model executed on the ESO Web server. The results page presents the computed results, including number of counts for the object and the sky, signal-to-noise ratios, instrument efficiencies, PSF size etc.. The optional graphs are displayed within Java applets allowing interactive manipulation. The results are also provided in ASCII and GIF formats for further analysis and printing. Finally a summary of the input parameters is appended to the result page.
The model includes an input spectrum (e.g. a template star spectrum), atmospheric parameters , optical instrument path and an observation criterium. The model generates output graphs describing the spectral illumination of the CCD, the instrument efficiency or the signal to noise as a function of the exposure time or seeing.
The target model is a spectral distribution constant with the wavelength.
The target model is a blackbody defined by its temperature and monochromatic apparent magnitude at a given wavelength. Temperature is expected in Kelvin and wavelengt in one of the band filters U, B, V, R or I.
The target model can be defined by a template spectrum . As with the blackbody it will be scaled to the provided magnitude and band filter U, B, V, R or I.
You must select the filter and filter magnitude for proper scaling of the template spectrum. Available filters are V, J, H, K, L and M. For extended sources, the magnitude must be given per square arc second. The effective central wavelengths are respectively B=440nm, V=550nm, R=640nm, I=790nm. The spectrum is scaled after integration in the corresponding photometric filter using the formula log10(F) = -0.4*M-Z, with M the magnitude in the band, F the monochromatic flux in W/m2/micron, and Z the zero point. The photometric zero points and effective wavelengths for (B,V,R,I) filters are taken from the reference: Bessel, 1979, PASP, 91, 589. The values used are Z(B)=7.180, Z(V)=7.442, Z(R)=7.64, Z(I)=7.91.
Point Source are sources whose spatial extent on the sky is much less than the seeing diameter. The signal to noise is computed over an area of diameter twice the seeing.
The signal to noise for extended sources is given per pixel on the detector. The magnitude is given per square arcsecond.
The sky brightness depends on the phase of the moon. The following table provides the observed sky brightnesses associated to the days from full Moon, from average values at Parnal. Note: Values for the Z band correspond to an extrapolation and will be updated when measurements are available.
The airmass can be computed with the Skycalc page.
The seeing is given in arc seconds.
To help observers in the choise of configuration, a table of computed S/N ratios obtained by the GIRAFFE ETC is provided. The source is a V=14.5 star of spectral type B1V, F0V or K2V. The table is covering the 468 combinations of the following four parameters:
The fiber mode is selected from a list with an option menu. The MEDUSA mode uses 132 fibers for observations, plus 5 fibers for calibrations. In the IFU mode, there are 300 fibers for observations, plus five for calibration. Sky sampling is the projected size of a fiber microlens onto the sky.
Two different gratings exist for the GIRAFFE echelle spectrograph. The links below will show tables of the spectral resolution for the two modes, and their dependence on fiber mode (due to size of microlens) and central wavelength.High Resolution:
The wavelength for the central pixel, in nanometers.
A curve of signal-to-noise as a function of exposure time is generated. The centre value and a range to both sides of the centre value are provided.
A curve of signal-to-noise as a function of exposure time is generated. The centre value and a range to both sides of the centre value are provided (in seconds).
The result graphs are Java based applications. A static version of the graphs is also provided in GIF and ASCII format.
Use fine Spectral Sampling With this option, the ETC will use a 16x higher sampling (2048 bins instead of the default 128). This may be needed to resolve highly peaked spectra. Execution time will be longer.
Spectroscopy results such as efficiency, signal, signal-to-noise estimates are dependent on the wavelength and given over the wavelength range in graphics form. A summary of results is provided in text form for the central pixel of the range (also corresponding to the central wavelength).
Central Wavelength: The wavelength of the central pixel, in nanometers.
Wavelength Range: The respective wavelength associated to the first and last pixel of the detector for the given configuration and dispersion, in nanometers.
Dispersion: The dispersion of the spectrum, in nanometers per pixel.
Plate scale: The plate scale of the system, in arcsecs per pixel.
Probability of realisation of seeing: The probability for the seeing to be better or equal than the specified value.
FWHM of the fiber spatial profile: The full-width at half-maximum of the fiber spatial profile. This value is the fiber size divided by the plate scale.
Efficiency at central wavelength: Total efficiency of the system at central wavelength, telescope transmission, optics and detector efficiency, in percent.
Fiber injection loss: The percentage of light that is lost at fiber entrance. A higher loss will occur at poorer seeing, since only the central part of the PSF will enter into the fiber.
Object-fiber displacement: In the MEDUSA fiber mode only, the fiber injection loss is partially caused by the possible displacement between fiber and object. In the computed results the entry "Loss due to object-fiber displacement" will be present. This is the fraction of the total fiber injection loss due to decentering, as specified by the input parameter "Object-fiber displacement". In modes other than MEDUSA, the value of this input parameter is irrelevant and has no effect on the computed results.
Object signal at central pixel: The total flux contribution from the object, integrated over all fibres within the extend of the PSF, and expressed in electrons per pixel along the dispersion direction. In Medusa the ETC gives directly the SNR of the extracted spectrum, while in IFU or Argus mode to obtain the SNR of the central extracted spectrum, the SNR given by ETC should be divided by sqrt(Number of Fibers covering the source). The value is given at the central wavelength and corresponds to "object_signal" in the signal-to-noise formula.
Sky background level at central pixel: The flux contribution from the sky for one row along the dispersion direction, in electrons per pixel along the dispersion direction. The value is given at the central wavelength and corresponds to "sky_signal" in the signal to noise formula.
Max. intensity at central wavelength (object+sky): This value is the sum of the sky background level and the fraction of the object signal falling on one pixel at the center of the slit profile. If there are more than 1 fiber on the source, this value refers to the sum of fibers and must be carefully interpreted (see VIMOS ETC User Manual).
Detector saturation: The detector saturation level. A message will be displayed if the maximum intensity is greater than this limit. Please note that the actual saturation level may depend on the CCD readout-mode, and that the saturation is here tested only for the central wavelength.
Detector read-out noise level: CCD read-out noise in electrons/pixel. This value corresponds to CCDnoise in the signal-to-noise formula.
Detector dark current: CCD dark current in e-/pixel/hour. This value corresponds to DarkCurrent in the signal-to-noise formula.
Number of fibers:The number of fibers covering the source over a circular area with radius=1.5*seeingFWHM, the radius inside which a gaussian profile is at least 1/16 of its central value. If the source is extended, results are given pr. fiber, ie. only one fiber is used in the calculations. MEDUSA mode is always single fiber.
Fiber diameter: The number of detector pixels the fiber projects onto. It is computed for the GIRAFFE dispersion physical model and takes into account the effect of inclined projection. It is approximately twice the FWHM of the fiber spatial profile.
Signal to Noise at central wavelength: The signal to noise is calculated over 1 pixel along the dispersion and summing the sky signal over (Number of fibers * fiber diameter in pixels) in the spatial direction.
PSF extension: number of pixels over which the signal-to-noise is estimated. This value is computed as twice the seeing divided by the plate scale. This value corresponds to "Npsf" in the signal-to-noise formula.
Signal to Noise at central pixel: Signal to noise at central wavelength, computed using the formula: S/N = object_signal / sqrt (object_signal + sky_signal*npsf + Npsf*DarkCurrent*ExpTime + Npsf*CCDnoise**2).
Signal to Noise near maximum transmission: Calculated over pixels where the transmission is above 50% of the maximum transmission.
The total integrated counts contribution from the object, in e-/pixel. The integration is done along the slit. The counts are expressed in electrons per pixel along the dispersion direction.
The sky contribution on each row of the detector, in e-/pixel. This value is not integrated along the slit.
The input flux distribution for the selected target is diplayed in units of 1e-20 ergs/cm**2/s/A
Toggling this option will display a curve showing the evolution of Signal to Noise Ratio against wavelength.
This option will display a curve showing the total efficiency of the system, and a second graph showing the dispersion relation.
The setting HR20B is now supported and the throughput of setting HR19B has improved. These are the only changes due to an update of the efficiency curve for the corresponding order sorting filter (order7) with an extended wavelength coverage towards the red.
New CCD "Carreras" enabled.
Wavelength coverage in mode HR19B corrected.
New layout of plots.
Added an option to have finer spectral sampling.
Corrected an index error. Results referring to specific wavelengths may differ with respect to previous versions.
Additional HR modes.
New HR grating.
ARGUS modes enabled.