CALBLOCK-PV-012: NISP Dark Current and Snowballs¶
Observation configuration file¶
Scheduling¶
Once, during PV phase.
In case of failure, repeat at next possible opportunity.
Time-critical¶
No
Objective¶
Dark current: Measure the thermal dark current for the photo- and spectro-MACC modes. During gound-testing, in-flight conditions cannot be established and hence this must be done from space. The dark current in the NISP detectors is very low (~0.02 e-/s) and thus a measurement with high S/N is time-consuming and will be carried out just once, during PV phase.
The number of dark exposures necessary to meet the CalCD-B requirement depends on the amplitude of the dark current. We must accuractely measure in particular the pixels with high dark current, since these appear as hot pixels if not corrected for, and might have to be masked if their Poisson noise violates the photometric error requirement.
For a dark current of \(d = 0.07\) e-/s, a photo-MACC readout noise \(RN=7\) e- (median 6 e-, 7e- is 95 percentile; R. Barbier), and wide-survey observation modes, 480 dark exposures are necessary to meet the requirement. The number depends weakly on the dark current itself, i.e. for the typical value of 0.02 e-/s we’d need 440 exposures, for 0.2 e-/s we need 580. It can be estimated with \(750*d + 8.8*RN^2\).
The non-Gaussian distribution of the dark current determines how many pixels with high dark current are still problematic with 500 dark exposures. The combined budget for hot and dead pixels is 5%.
Note that in-flight the dark frames are contaminated by persistence from the zodiacal background and astrophysical sources, unless very long waiting times are obeyed prior to the measurement to allow for persistence decay.
Slew darks are described separately.
Snowballs: These are highly energetic events that originate probably from cosmics, and / or radioactive decay in the detector substrate, detector coatings, or optical coatings. Usually, they appear radially symmetric and extended in NISP images and may saturate the detector. They might be accompanied by showers or increased number of hot pixels or secondary hits in their immediate surrounding.
Observations¶
With the FWA closed, obtain 500 dark exposures for the photo-MACC mode, and 100 for the spectro-MACC mode. Of most interest is the hot pixel tail of the distribution, and the identification / check of bad pixels.
The observations for each MACC mode must be consecutive and without interruption. While the photo- and spectro-sequences can in principle be scheduled independently, they are best executed after each other to save the extra wait time for persistence decay.
Fulfilling¶
CALPRODUCT-NI-006: NISP Dark Current
Targets and pointing constraints¶
Pointing: Straylight suppression as per MOCDA-SYS-003.
Instrument configuration¶
Instrument: NI-FWA closed, NI-GWA RSG000
Instrument Configuration Submode = 8 (SPECTRO DARK), to recalibrate the chi2-threshold for the onboard flagging of MACC(4,6,4) images.
Methodology / Uplink procedure¶
Entity: NISP
Commanding Method: NISP: Observation
Dither Sequence: None
Inter-operability constraints¶
VS-CU off
In-orbit duration¶
In-orbit Duration: 40.3 h (including 9h waiting time)
The dark sequence has the following two parts, which can be observed independently at different times:
Mode |
Time [s] |
|---|---|
photo |
|
spectro |
|
Note that the total time listed above contains a waiting period of two times 4.5 hours. The requirement (R-CAL-B-NP-1476) is that after 4.5 hours the persistence flux of any illumination must have dropped below 0.023 e-/s/pixel. This requirement is not fully met by all pixels, hence we conservatively chose a waiting time twice as long.
The waiting time can be dropped if the dark observations are scheduled at a time after which NISP has been inactive, or not exposed to any light. For example, this could be done after orbital maintenance.
Data volume¶
500*NP + 100*NS = 63500 MB
The data volume of NP is increased by about 60% from 59 MB to 94 MB due to the download of the chi2 values.
Processing input¶
Ground Processing Input Data: BPM, persistence masks
Data Description: 600 NISP science images
Processing method¶
Snowballs:
Subtract the master dark from each individual dark. In the latter, identify very bright and extended detections, meaningful thresholds thereof will only be known from the observations themselves.
Processing output¶
Darks:
Updated chi2 threshold for MACC(4,16,4) to account for radiation events in-flight
Combined master darks for the photo- and spectro-MACC modes
BPM update
Histograms of dark current values for each detector
Notable changes with respect to ground-characterization data not explained by temperature variation
Snowballs:
Histograms of fluxes and FWHM of detected snowball events
Search for different flavors, that is snowballs with and without secondary showers.
Required accuracy, constraints¶
NISP detector baseline stability: All exposures used in the production of the master dark shall have a reference pixel median value stable within TBD % (to be resolved between NISP-S instrument scientist and NISP detector scientist).
NISP detector thermal stability: The NISP NI-DS temperature variation shall remain stable within 20 mK/h (TBC) during the exposures used to produce the master dark.
Expected dark current levels within 0.02 e-/s
Spatial variations within TBD %
Spatial appearance as expected from ground-based measurements (possible dependence on detector temperature)
Absence of light leaks
Number of hot / bad pixels within expectations
Processing tools¶
L1 dark pipeline, trending statistics, histogram
Validation¶
Pure thermal dark conditions as for these characterizations are not encountered during routine operations. The dark current will always have contributions from persistence. For pure thermal darks, the prior illumination history must be respected and a waiting period calculated such that contamination from persistence is negligible. This period will be on the order of 10 h, and potentially longer. For example, the measurement could be done after orbital maintenance or other operations that prohibit NISP on-sky observations.
Contamination by persistence hampers the validation of the master dark because we are lacking a ground-based reference taken at the same detector temperature. A way forward would be a comparison of pixels with high dark current in the master dark with those in the slew darks. However, at PV phase, a sufficient number of slew darks for a meaningful comparison will not (yet) be available, and hence the master dark must got without initial validation.
Recovery¶
If the validation of the master dark fails, then one must first exclude other causes that could change the high flux tail of hot pixels. This can be changes in e.g. the RTN and persistence. Radiation damage toards the end of the mission is expected to chnage the number of hot / dead pixels, and therefore at some point a repetition of the thermal dark current could be required.
Quality control parameters¶
DataModel: https://gitlab.euclid-sgs.uk/
Contact person¶
NISP-P calibration scientist