Exposure timing is part of the total time required for a JWST proposal. This page explains the model used by JWST's Astronomer’s Proposal Tool (APT) to estimate and account for exposure time.
The primary component of the timing model is the time it takes to execute the exposures being specified. APT models the exposures that will be executed by the onboard scripts. Each exposure specification in APT may represent one or more actual exposures, due to number of exposures, dither points, mosaicking, and other exposure expansion rules. The following applies to the model APT has for the timing of an exposure modeled in APT.
An exposure consists of an exposure setup period in which the scripts are compiled onboard and variables are initialized. Because the detector is reading out constantly, there is a need to synchronize with the detector which on average can be a half a frame read. Then a series of integrations are performed, each one consisting of a group of detector frames, separated by a gap. Before each integration, there is a frame reset to clear the detector. Finally, there is a cleanup period.
The exposure timing model is shown in Figure 1.
Science Duration is the time spent collecting data. It is the sum of all integration durations. Components of Science Duration are described in the Science Duration section on this page.
Exposing Duration models the amount of time that the shutter will be open on the whole, and therefore includes Science Duration plus frame resets and synchronization. Exposing Duration is equivalent to "Total Exposure Time" in APT template forms, except that synchronization time is not included in Total Exposure Time.
Exposure Duration is the total time needed to execute the exposure, and therefore includes Exposing Duration plus all the setup and cleanup activities. "Exposure Duration" is equivalent to the term "Scheduling Duration" as seen in the visit forms of APT templates.
- Setup and cleanup times for each instrument are provided in the Instrument Specific Overheads articles.
- For each exposure, the Frame synchronization time is averaged to be ½ the FrameReadtime.
- Frame reset time is accumulated per integration and is usually equal to the FrameReadTime (see Frame Reset duration for details).
Note that for MIRI MRS exposures, the larger value of the long and short integrations is used to determine the FrameResets.
Science Duration ("PhotonCollect" in the APT times report) is calculated according to the following formulas. Components of each formula are described in Table 1.
PhotonCollect (sec) = FrameReadTime * ((Groups * NFrame) + ((Groups - 1) * GrpGap)) * Ints
Total Science Duration:
TotalPhoton = PhotonCollect * Diths * NumExp
Table 1. Components of Science Duration
|Name in Times report||Description||Source|
|FrameReadTime||Frame Read Time||PRD subarrays tables (PRD=Project Reference Database)|
|Groups||Number of Groups per Integration||Template|
|NFrame||Number of Frames per Group||PRD datamodes table|
|GrpGap||Gap between groups||PRD datamodes table|
|Ints||Number of Integrations per Exposure||Template|
|Diths||Number of Dithers||Template|
|NumExp||Number of Exposures||Template|
Frame Reset duration
FrameResetTime = FrameReadTime * number of resets
where number of resets is either NRESETS1 (prior to first integration) or NRESETS2 (between each integration) from the PRD datamodes table.
Overhead for fast full-frame reset
For each fast full-frame reset, overhead is calculated as:
number of background row resets * background row reset duration
Table 2. Components of Fast Full-Fame Reset duration
|Number of background row resets||NumResetRows from PRD subarrays tables|
|Background row reset duration||background_row_reset_duration from PRD timing_model_parameters table|
Version 4 (April 26, 2022)
- PROPINSTJWST-91509 Update diagram and update definitions.
Version 3 (February 4, 2022)
- PROPINSTJWST-91256 Added clarification for coordinated parallels.
Version 2 (December 4, 2018)
- editorial; removed Table 1 which was out of date and superseded by the Instrument Specific Overheads articles.
Version 1 (June 6, 2017)
- Initial version.