The initial choice that you need to make is to decide which pointing mode to use. This is judged on the extent of the source or field to be observed. The radio button list at the top left of the main part of the screen (as shown in Figure 6.13) lists the two choices:

You should select "Single Pointing" for areas smaller than 2 arcmin in diameter and "Raster" for areas larger than a circle of 2 arcmin diameter. These are described in Chapter 4. On selection of "Raster" you will see the "Raster Map Parameters" panel appear as shown in Figure 6.14 (the use of these are described later in Section 6.4.6).

The next choice that you need to make is to decide which image sampling to use. This is judged on the spatial sampling that is required for the observation. The radio button list (as shown in Figure 6.13) lists the three choices:

You should select "Sparse" to measure the spectrum of a point or compact source well centred on the central detector of the Spectrometer or to provide a sparse map (2 beam spacing) of size 2 arcmin. Select "Intermediate spatial sampling" for an observation that has higher spatial sampling (1 beam spacing), without taking as long as a fully Nyquist sampled map. Or select "Full" for fully Nyquist sampled (1/2 beam spacing) imaging spectroscopy of a region of sky or extended source. These are described in Chapter 4.

The next choice that you need to make is to decide on the spectral resolution you require. The radio button list (as shown in Figure 6.13) lists the four choices:

As pointed out in Section 4.2.1, the choice of spectral resolution requires particular considerations because low resolution spectra can a posteriori be extracted from high resolution observations. You should select "High" for spectra at the highest resolution available with the SPIRE spectrometer, Δσ = 0.04 cm^-1, "Medium" for a resolution of Δσ = 0.25 cm^-1, or "Low" to make continuum measurements at a resolution of 1 cm^-1. The "High and Low" choice is to make both line, and higher sensitivity low resolution continuum spectra more efficiently (in a shorter time and in a single observation) than can be done by increasing the number of repetitions in "High" resolution observations. This is intended for sources where an observation in "High" resolution mode does not contain enough scan repetitions to achieve sufficient sensitivity in the low resolution continuum, and it is more efficient to add additional "Low" resolution scans (see the example in Section 6.7.1 for more details).

The duration of the observation is controlled by the repetition factor which is shown in Figure 6.15. You should put in the box the number of repetitions you wish to carry out, where one repetition is:

See Chapter 4 for more details.

Note that the actual time taken depends on the spectral resolution, the image sampling (and the size of the raster), as does the maximum value that can be entered. For Spectral Resolution "High and Low" two boxes appear allowing the number of scan pairs at each resolution to be controlled independently.

In Section 6.4.7 it is described how to see the actual duration using the "Observation Estimates" button.

Pressing the "Source Flux Estimates" button brings up a table in which you may enter details of your source. Note that it is optional to enter data here, you can enter data for zero to eight wavelengths, if you do enter information then you will be presented with an extra table in the Observation Estimates result panel with the signal to noise (S/N) information as calculated from the input values. The Source Flux Estimates input table depends on the Spectral Resolution selected, the High, Medium and "High and Low" Resolution table is shown in Figure 6.16 and the Low Resolution one is shown in Figure 6.17. In the High, Medium and "High and Low" Resolution table you can enter values of "Wavelength (μm)", "Estimated line flux (10^-17 W/m²)" and "Estimated Continuum". The units for the "Estimated Continuum" are selectable from the "Select Continuum Units" pulldown menu, there is a choice between Jy (the default) and 10^-17 W/m²/μm. In the Low Resolution table you can enter Wavelength and Estimated Continuum as well as Select the Continuum Units. Press "OK" once the appropriate values have been input. These values will be used for the noise estimates.

This option is only available when the Pointing Mode Raster is selected. The panel is displayed in Figure 6.18.

You should use this to enter the details of the area to be mapped. In Length you should enter the length dimension of the area to be mapped, this is the length of the raster rows in arcmin that the telescope will make the raster points along (note, the spacing of the raster points is 116 arcsec so to control the number of raster points well you need to put in decimals of arcmin to avoid getting too many extra raster pointings as the number of pointings is rounded up).

In Height you should enter the other dimension of your map in arcmin. From this the number of raster rows is calculated using the optimal spacing of 110.6 arcsec. Note that the number of raster rows needed will be rounded up to make sure that your dimension is fully covered so again be careful to control the number of pointings using decimals of arcmin.

If the target is not necessarily the centre of the map then map offsets can be used. These offsets in arcmin are given in detector Y, Z coordinates (see the example in Section 6.6.5 to understand the behaviour of the offsets).

As the orientation of the map on the sky rotates, unless you are near the ecliptic plane (see the Herschel Observers' Manual and Chapter 4), then the area you define rotates with time. To check this you should do a visualisation at the extremes of a visibility window. You can make sure that the area you want to map is covered by looking at the extent of the rotation and make your area bigger. As the time to make, in particular, observations with High Spectral Resolution and Full Image Sampling can be long you might like to consider to set the Map Orientation constraint, or to spilt up your observation into several observations, see the example in Section 6.7.3. You can set this constraint on the orientation of the raster rows of your map via the Map Orientation box and selecting from the pulldown menu "Array with Sky Constraint". This displays a warning similar to that shown in Figure 6.8 as by selecting "Array with Sky Constraint" you are setting a constraint on when your observation can be performed, so you should think about this carefully. Press "OK" to click away the warning and then the Angle From and Angle To boxes will be ungreyed allowing you to enter the relevant values of the angles that you wish the raster rows to lie between. To see the raster rows use Animate of the visualisation, see HSpot Users' Guide for details. If you want the raster rows to lie between the angles 30 - 150 degrees east of north then you should enter 30 in the Angle From box and 150 in the Angle To box. Note if you want the raster rows to lie in the direction to the north (say between 300 to 40 degrees) then you should enter 300 in the Angle From and 40 in Angle To box. It is very important that you visualise your observation at different dates to make sure that you observation is still possible as the visibility windows do not take into account your constraints (see the example in Section 6.7.3).

An accurate time estimate and associated noise is obtained by clicking the "Observation Est..." button to bottom left of the AOT window. The software calculates the time that the observation should take and presents that in a new panel. If the values are okay to you then click OK and DONE on the main AOT window to complete the request. The Time Estimation Summary as shown in Figure 6.19 gives the return information for the default Spectrometer observation (with the following values selected Single Pointing, Sparse and High and a Repetition factor of 2).

Figure 6.20 and Figure 6.21 show the results for the Spectral Resolution of "Low" and "High and Low" respectively. The 1-σ flux sensitivity (for all resolutions except "Low"), the 1-σ continuum sensitivity and the unapodised resolving power for each of eight standard wavelengths is returned in the top table followed by the time estimate. The time estimate shows the total time of the observation and also the various components that contribute to the time. If you entered Source Flux Estimates these will be reported in a second table as well as corresponding Signal to Noise information in the S/N column.

You are provided with more information on the observation by pressing the "Details" button in the bottom right of the Time Estimation Summary window.