JWST slitless spectroscopy uses a grism to disperse light from all targets observed in the field. Planning is required to mitigate overlapping spectra.
Several of the JWST instruments offer a slitless spectroscopic mode. Slitless spectroscopy is particular in that every source in the field results in a dispersed spectra. This mode differs substantially from regular direct imaging observations and slit spectroscopy. Some of the similarities and differences between imaging and slitless observations that can impact observing strategies and planning are discussed below.
JWST slitless spectroscopic modes
There are two main categories of slitless observations: wide field slitless spectroscopy (WFSS) and single object slitless spectroscopy (SOSS).
WFSS mode disperses the light of any object that is within the field of view of the instrument. This often results in hundreds, if not thousands of spectra that often overlap in the final observation. This mode is similar to the HST NICMOS, ACS and WFC3 grism observations. NIRCAM and NIRISS both implement WFSS using two different grisms that disperse the spectra either horizontally or vertically onto the detector. This allows for the dispersed spectra to overlap in completely different manner without having to change the orientation of the whole JWST telescope.
In the following sections, some of the WFSS issues are discussed and these are also relevant to SOSS observations. SOSS observations however differ in the type and amount of expected spectral contamination.
The following table summarizes the WFSS modes of NIRCAM and NIRISS. Information about the MIRI slitless mode can be obtained from the MIRI instrument pages.
Similarities to direct imaging
Slitless spectroscopy consists of inserting a disperser (in most cases a grism) into the optical path of light that would otherwise result in a regular image. All of the JWST slitless modes use the same detector for both direct imaging and slitless spectroscopy. Slitless observations are subject to some of the same limitations as direct imaging observations. Just like imaging data, slitless data will be affected by the background level, bad pixels, and cosmic rays. The consequences of these are dealt with in a similar fashion as direct imaging: observations should be dithered and on-the-ramp-fitting used to mitigate the effects of cosmic rays.
Similarities to slit observations
See also: JWST Slit Spectroscopy
In the sense that both slit and slitless observations both disperse the light from an object, the two are very similar. The trace of the dispersed spectra is determined by the dispersing element in both cases. In both cases, if more than one object is present in the field or slit then multiple spectra will be generated. Both slit and slitless observations often have more than one dispersed orders and in both cases one can have an instrument that is designed to disperse spectra horizontally or vertically.
Peculiarities of slitless observations
Slitless spectroscopy however differs from direct imaging in some crucial ways. Since a disperser is inserted into the field of view, the light from every point on the sky is dispersed, resulting in a much larger background level than with slit spectroscopy. Light that would have been incident on a given detector pixel will be dispersed, or spread out as a function of wavelength.
This is illustrated in Figure 2 that shows how light that would normally be incident on a given pixel is instead dispersed onto the detector (defining what is referred to as the "spectral trace," shown by a solid blue line). In this particular example, a spectrum is dispersed in the horizontal direction, as would be the case when observing with the NIRCAM R grism. Some JWST grisms actually disperse spectra in the vertical direction (such as the NIRCAM C grism, the NIRISS GR150R grism, and the MIRI prisms). Also shown is the effect of the sensitivity variation, as a function of position along the trace (i.e., wavelength), using different shades of blue.
Spectral confusion and contamination
Because the light is dispersed, light from adjacent pixels can now overlap, smearing and lowering the effective spectral resolution of the slitless spectrum of an extended source. This is illustrated in Figure 3.
If the light varies across the sky (e.g., an object with different colors in different pixels), then the resulting dispersed spectra are combined mainly in the dispersion direction, as shown in Figure 4.
Mutiple spectral orders
The discussion above referred to first order spectra. Dispersers have been designed to produce very little light in higher order spectra. Still, one should include these higher orders in contamination estimates. The repressed second order of a very bright nearby source could negatively affect the first order spectum of a science target, for example. Determining the impact of second order spectra contamination requires a careful simulation of the observations. In the case fo NIRCAM, second order spectra only impact a small number of GRISM+filter combination and this is described in the NIRCAM documentation.
Nearby source contamination
Contamination can also be caused by objects that are visually near a source of interest. Since the light from every object is dispersed in the same direction, overlap can occur. Observing these sources using different position angles on the sky will result in different types of contamination.
Background and field of view limits
The background of WFSS slitless observations cannot be expected to be as spatially flat as that of direct imaging. This is because the field of view has a limited size and this results in a deficit of dispersed light on the edges of the slitless observations. If spectra are dispersed nearly horizontally, then the background will be lower on the left and right edges. Similarly, top and bottom edges will be affected when light is dispersed vertically. The strength of this effect is dependent on the size and throughput of the multiple orders of the dispersed spectra.
Additionally, the field of view of the dispersing element is larger than the focal plane array and light from the background as well as objects that are outside of the field of view of a direct image will be dispersed onto the detector.
Figure 6 shows the effect of the field of view limits with instruments performing WFSS (such as NIRCAM and NIRISS), with spectra being dispersed horizontally as well as vertically.
Main articles: NIRCam WFSS Recommended Strategies, NIRISS WFSS Recommended Strategies
The combination of the dispersed background and dispersed light from other sources can quickly become a limiting factor when observing faint targets. Contamination can be difficult to estimate, and realistic simulations of slitless data are the only definitive way to plan observations if contamination can be significant or needs to be avoided. This requires prior knowledge of all the sources in (and nearby) the field of view, as well as their approximate spectroscopic characteristics (e.g., continuum slope and intensity). Images obtained using broadband filters are a good way to determine the location as well as the color of each source, providing information that can be folded in when generating simulations.
In cases where contamination needs to be avoided, multiple orientations on the sky can be simulated and the best one can be selected. In the case of JWST, not all position angles are available to an observer at any given time, so careful planning and timing can be required. NIRCAM and NIRISS have two distinct dispersers, allowing for spectra to be dispersed either horizontally or vertically. Combining these observations allows for possible decontamination of the spectra.