Chandra X-Ray Observatory
Skip to the navigation links
Last modified: 17 Jan 2017


Grids (rasters): RPS, Constraints and Scheduling

Proposer Threads (Cycle 19)


Thread Overview

This thread describes an observation which requires several adjacent ACIS pointings (i.e. a raster or a grid). It is based on the Cycle 5 SWIRE Survey. The survey consists of nine 70 ks pointings in a 3x3 grid covering 0.6 square degrees in the vicinity of the Lockman Hole. The four ACIS-I chips are used. The layout of the fields is shown schematically in Figure 1.

[A simple grid observation]
[Print media version: A simple grid observation]

Figure 1: Acis Grid Pointing

A schematic figure showing the proposed grid layout.

It is possible to specify a grid of pointings in RPS with 4 parameters as described below, provided all pointings use identical observing parameters. In this thread, we show how to specify such a grid in RPS.

The roll angles of observations in a grid will not necessarily "line up". We also show how to constrain the observations if your science were to require that there are no gaps in the survey.

RPS Grid Parameters

It is possible to specify a grid in one of two ways:

  • Enter each position as a separate target (time consuming but still viable for a small number of pointings).
  • Use the grid parameters. These are:
    • Coordinates of survey center
    • Number of pointings
    • Distance from the survey center to the farthest grid pointing
    • Total exposure time (sum of all pointings)

The use of the term "grid" is slightly misleading in that the survey region can be irregular; it does not need to be a rectangle.
For the SWIRE survey we use the following RPS inputs:

  • Target Coordinates: RA=10:46:00.00, Dec=+59:01:00.00
  • Set the flag "Is this observation (part of) a grid?" to: Yes
  • Number of Grid Pointings: 9
  • Total Observing Time: 630 ks
  • Distance from center to farthest grid pointing: 0.59 degrees.
    Here we take the farthest point to be the diagonal from the survey center to the aimpoint of any "corner" grid pointing.

CXC staff will contact PIs of successful grid proposals after the peer review at which time PIs must supply a list of target positions.

Slew Tax for Grids

A "slew tax" of 1.5 ks is routinely added to each observation approved at the Chandra peer review. The slew tax may be reduced for programs consisting of several closely spaced observations. For longer exposures the slew tax formula is equivalent to the default 1.5ks per pointing. The slew tax for this survey is therefore 9x1.5 = 13.5ks.

Details of how to calculate slew tax for exposures shorter than 44 ks is given in the Call for Proposals and the Slew Tax Worksheet.

Constraints and Scheduling

The observations in a grid are NOT automatically constrained. In practical terms, this means that that they may scheduled at any time during the cycle. Since the nominal roll of Chandra changes with time, the roll angle of each observation may be different. THE ROLL ANGLES WILL NOT LINE UP, as illustrated in the following figure. This will most often lead to gaps in the sky coverage.

[What a grid observation may look like without a roll constraint.]
[Print media version: What a grid observation may look like without a roll constraint.]

Figure 2: Unconstrained Grid

What a grid observation may look like without a roll constraint.

The "Constraints/Slewtax" tool, accessible from the top of the RPS form, can properly estimate the number and category of constraints that a grid observation will entail. Note that additional constraints or difficulties imposed in the Remarks section cannot be evaluated by this method. The tool will also assess the slew time associated with each pointing in the grid, and give an accurate count of the total time required for your observation.

Please note that it is often most efficient to place several pointings from a survey contiguously in the observing schedule (this minimizes slew time). The number of such pointings that can be accommodated before slewing to a different location on the sky is dependent on the pitch angle of the survey observations. The most likely outcome for a survey with many short pointings is for a group of them to be scheduled together, followed by a second group at a later date, followed by another group until the observations are complete. Observations within a group will have essentially the same roll, but different groups will have different rolls (illustrated above)

If it is essential to have uniform coverage, the proposer must restrict the roll angles of the observations. This can be done in two ways:

  • Impose a roll constraint (with associated tolerances) if the roll angle of the survey must be fixed.
  • A group constraint should be used if the absolute value of the group constraint forces all the observations to be done within a certain time interval, but does not specify the exact window (e.g. all done within 30 consecutive days, but the 30 day period can start at any time during the cycle).

Determine Roll Tolerance

It is important not to have gaps in the coverage of the SWIRE survey. The absolute value of the roll is not important. The maximum acceptable change in roll between two adjacent pointings is illustrated in the following figures.

[Two adjacent fields at zero roll angle.]
[Print media version: Two adjacent fields at zero roll angle.]

Figure 3: Zero Roll Angle

Two adjacent fields at zero roll angle. Note the overlap between the two chips.

[Two adjacent fields where the second field is rotated.]
[Print media version: Two adjacent fields where the second field is rotated.]

Figure 4: Unacceptably Large Roll

Two adjacent fields where the second field is rotated: this is NOT acceptable for our example because there is a gap.

[Two adjacent fields with maximum roll difference.]
[Print media version: Two adjacent fields with maximum roll difference.]

Figure 5: Maximum Roll Difference

Two adjacent fields with maximum roll difference. Further rotation will create a gap. This geometry is correct if the overlap between two adjacent chips is small compared to the width of the chips (in this case the overlap is 1 arcmin and the ACIS-I chips are 16 arcmin on a side).

The maximum roll (theta) is calculated from:

cos(45 - theta) = 8/10.6
45 - theta = 41
theta = 4 degrees

Evaluate Group Constraint

The roll angle may change by a maximum of +/-4 degrees between the first and last pointing. The corresponding time interval can be estimated using ProVis tool. Enter the coordinates of the SWIRE field center (RA=10:46:00.00, Dec=+59:01:00.00) in the box labeled "Target Coordinates" and hit the "Generate Plot" button. The resulting page is a plot of roll and pitch angles and target visibility as a function of time. Roll angles and date are read out via the cursor. Inspection of the plot in Cycle 19 shows that a reasonable estimate for the rate of change of roll with time is about 7 degrees per week. Therefore, +/-4 degrees tolerance translates to about 8 days. The RPS group parameters are set as follows:

  • Set Group Observation Flag to Y
  • Enter "Group Identification" as SWIRE (can be any string to identify targets)
  • Time Interval for Group=8 days

Number and Type of Constrained Observations

Constrained observations are classified as "Easy", "Average" or "Difficult" (see the Call for Proposals and Constrained Observations Worksheet). The dimensionless grading parameter for a group constraint is as follows:

                        Time Interval for the Group
                   Total Duration of Observations in the Group

The time interval for the group is 8 days. The total duration of the observations is 70x9 = 630 ks or 7.3 days. The grading parameter is 1.1. According to Table 5.1 of the Call for Proposals, this counts as 9 Difficult observations. Please note that if the exposure time for each observation was 44 ks or less, the number of time constrained observations would be charged at a reduced rate, as described in the Slew Tax Worksheet.


Using the SWIRE survey as as example, this thread shows how to plan a grid observation i.e. several adjacent or close pointings. Provided your targets have identical observing parameters, it is possible to specify a survey with 4 parameters -- total exposure time (sum of all pointings), coordinates of survey center, number of pointings, and distance of the furthest pointing from the survey center. For the purposes of filling in RPS forms, a "grid" can be an irregular region. If your pointings are not all identical, you will need to enter a separate target form for each pointing.

The observations in a grid will not necessarily "line up". This is because the nominal roll angle of Chandra changes with time. Observations done at different times will have different roll angles. If your science demands that the observations "line up", it is necessary to impose a roll or group constraint. A roll or group constraint will limit the time when the observations can be carried out. For the SWIRE survey we calculate the maximum acceptable roll angle change, and use ProVis to evaluate the corresponding group constraint. Finally, we use the grading scheme in the Call for Proposals to determine that the SWIRE survey counts as 9 difficult observations.

Last modified: 17 Jan 2017
Smithsonian Institute Smithsonian Institute

The Chandra X-Ray Center (CXC) is operated for NASA by the Smithsonian Astrophysical Observatory. 60 Garden Street, Cambridge, MA 02138 USA.   Email: Smithsonian Institution, Copyright © 1998-2016. All rights reserved.