[Comm2011] Error bar for extinction / Fast anisotropy
Steffen Mieske
smieske at eso.org
Tue Aug 9 09:13:55 CEST 2011
Hi Ewout,
thanks for your fast reply.
Actually, for the IQ check in real-time -- and hence our OB
classifications -- we do need a precise value of the IQ over the FOV,
and as f(chip). This is in principle provided by the ESO pipeline in
almost real time. But I hope the anisotropy script will give a good
complement especially in borderline cases, where you want some graphical
impression how the IQ varies over the FOV and relative to the user
constraint for seeing. Furthermore, depending on how many short
exposures are taken, we may get a pipeline backlog, and then the
anisotropy script will be very useful for reasonably quick and precise
IQ assessment.
Cheers,
Steffen
On 08/09/11 02:54, Ewout M. Helmich wrote:
> Hi Steffen,
>
> - Error bars are probably difficult. I have to look closer if this is
> possible, but it is likely the data required to calculate errors is
> lost early on in the data structure.
> - I have partly implemented (in reverse) this notion. I.e. if you run
> astro/filerecipes/Monitoring.py directly there is a switch '-i' which
> will open the plot as you suggest. I can implement as you suggest, and
> for the OCAM_monitoring script, though.
>
> - About the PSF anisotropy, well it seems to me the plots are really
> only useful when you display the results for all CCDs. The script has
> proved very useful during commissioning, but afterwards I don't think
> it makes much sense. For real-time checks on the PSF it is probably
> easier to use Ctrl-P ("Pick an object") on the RTD (skycat).
>
> Ewout
>
> On 08/09/2011 07:19 AM, Steffen Mieske wrote:
>> Hi John and Ewout,
>>
>> I enjoy working with your 3 tools. I would have two minor requests
>> for the extinction tool
>>
>> - would it be possible to have error bars for the data points on the
>> extinction monitoring output plot
>> - it would be nice if the script would automatically open the plot.
>>
>> On a related note, Gerd and I made a small addition to the
>> OCAM_psf_anisotropy script:
>>
>> Lines 53-56 in OCAM_psf_anisotropy
>>
>> if extensions == ['f']:
>> self.extensions = ['1','20','32','13','25','12','21','8']
>> else:
>> self.extensions = extensions
>>
>>
>> So you would call it "OCAM_psf_anisotropy image.fits -e f" to get
>> some default sub-set of chips. 'f' for 'f'ast
>>
>> Cheers,
>> Steffen
>>
>>
>>
>>
>>
>>
>>
>> On 08/09/11 00:16, Steffen Mieske wrote:
>>> Hi John,
>>>
>>> the reason for the strange exposure times was a 1 day 'glitch' in
>>> the FIERA software, which occured presumably due to a non 100% clean
>>> installation of a new FIERA software version. In particular, it is
>>> due to a wrong sign in the time recorded for the Shutter
>>> Opening/Closing. As a matter of fact, you will notice that all
>>> images taken between roughly 20:00 UT on August 2 and 3 have the
>>> exposure times wrong by +-0.804 s, which is the time of the shutter
>>> motion.
>>>
>>> As of August 4 the situation is stable, due to a clean restart of
>>> the software.
>>>
>>> Regarding the lamp switch off: this would up to today have happened
>>> every time that the lamp is in operation for more than one hour.
>>> There is an automatic time-out of 1hr, which is only reset when the
>>> lamp is switched on for the first time, but not if you take a
>>> sequence of flat images. Having said this, in normal daytime
>>> calibrations it should not have occured anyway, since the automatic
>>> calibrations would do a 1.5hr dark before the gain. It may be that
>>> the calibration OB was edited by hand on that day, omiting the
>>> darks, and resulting in a very long sequence of dome flats + gain, >
>>> 1h.
>>>
>>> Andrea has increased the timeout to 3 hrs, so I think we are fine.
>>>
>>> Cheers,
>>> Steffen
>>>
>>>
>>>
>>> On 08/08/11 18:23, John P. McFarland wrote:
>>>> Hi Steffen,
>>>>
>>>> The keyword treatment can wait unless the ingest problems can't be
>>>> fixed (highly unlikely that they can't).
>>>>
>>>> The integration times for 3 out of the last 5 gain templates have
>>>> those exact time sequences. The exceptions are on the 3rd of
>>>> August as Ewout reported and the one for 1st July where the times
>>>> look good, just much longer (1-60 seconds).
>>>>
>>>> Cheers,
>>>>
>>>>
>>>> -=John
>>>>
>>>>
>>>> On Mon, 8 Aug 2011, Steffen Mieske wrote:
>>>>
>>>>> Hi John,
>>>>>
>>>>> Andrea implemented the default lamp keyword, as I had written to
>>>>> you. We can discuss again the issue of the individual lamp
>>>>> keywords with Andrea, at the time we agreed that this was low
>>>>> priority during OCAM3 since it is not lacking information as such.
>>>>> The permanent inclusion of this keywords with values either T/F is
>>>>> something I can also ask Paranal Software to implement after
>>>>> OCAM3, so no need for Andrea to do this now.
>>>>>
>>>>> The gain template issue is reaaaally strange, I will include it
>>>>> again in tomorrow's calibration.
>>>>>
>>>>> I first thought that the lamp switches off because there is some
>>>>> integration with EXPTIME=0 sec, some bias inbetween. But this
>>>>> should not the case according to the below content of the gain
>>>>> template that is used for the daily calibrations. Can you confirm
>>>>> that those are the integration times that you see in the normal
>>>>> case (the data from August 3 are not here anymore, so I cannot
>>>>> check)?
>>>>>
>>>>>
>>>>> # calib template OMEGACAM_img_cal_gain
>>>>> TPL.ID "OMEGACAM_img_cal_gain";
>>>>> TPL.NAME "Use Dome FF to check gain";
>>>>> TPL.MODE "FLAT";
>>>>> DET1.WIN1.UIT1 "0.2 4.3 4.0 0.3 0.7 3.5 2.7 0.1 1.4 2.1
>>>>> 2.1 1.4 0.1 2.7 3.5 0.7 0.3 4.0 4.3 0.2";
>>>>> SEQ.NEXPO "20";
>>>>> INS.FILT.NAME "z_SDSS";
>>>>> INS.LAMP.SET "Default";
>>>>>
>>>>> Cheers,
>>>>> Steffen
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 08/08/11 15:14, John P. McFarland wrote:
>>>>>> Hi Ewout,
>>>>>>
>>>>>> I have asked Steffen to discuss with Andrea the logic of how the
>>>>>> lamp keywords are included. If the logic is affecting the ingest
>>>>>> we need to either adapt or adjust the logic.
>>>>>>
>>>>>> Andrea and Steffen, have you come to any resolution concerning
>>>>>> the lamp keywords? It seems to me that removal of the keywords
>>>>>> is a poor way to indicate the lamp is off. I assume that is what
>>>>>> the state keyword (e.g., INS.LAMP1.ST) is meant to indicate.
>>>>>> Please tell us of your ideas regarding this. Thanks!
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>>
>>>>>> -=John
>>>>>>
>>>>>>
>>>>>> On Mon, 8 Aug 2011, Ewout M. Helmich wrote:
>>>>>>
>>>>>>> Hi Bernard,
>>>>>>>
>>>>>>> It looks like all the gain templates for OCAM3 taken up to this
>>>>>>> point
>>>>>>> are wrong. After about 10 minutes, in the middle of the gain
>>>>>>> template,
>>>>>>> the lamp appears to shut down. Lamp header items then disappear,
>>>>>>> which
>>>>>>> resulted in errors during ingest here (note to Danny).
>>>>>>>
>>>>>>> Also, the first gain template taken during OCAM3 has strange
>>>>>>> exposure
>>>>>>> times (see below). The exposure times for the 2nd and 3rd
>>>>>>> template are ok.
>>>>>>>
>>>>>>> Ewout
>>>>>>>
>>>>>>> ---------------------------------------------------------------------
>>>>>>>
>>>>>>>
>>>>>>> # Filename, template.start, template.type, EXPTIME
>>>>>>>
>>>>>>> OMEGACAM.2011-08-03T13:22:00.734.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN -0.704
>>>>>>> OMEGACAM.2011-08-03T13:30:39.989.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN -0.604
>>>>>>> OMEGACAM.2011-08-03T13:19:06.032.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN -0.504
>>>>>>> OMEGACAM.2011-08-03T13:27:45.437.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN -0.104
>>>>>>> OMEGACAM.2011-08-03T13:24:54.515.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 0.596
>>>>>>> OMEGACAM.2011-08-03T13:25:37.206.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 0.904
>>>>>>> OMEGACAM.2011-08-03T13:16:51.761.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 1.004
>>>>>>> OMEGACAM.2011-08-03T13:28:30.448.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 1.104
>>>>>>> OMEGACAM.2011-08-03T13:23:28.225.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 1.296
>>>>>>> OMEGACAM.2011-08-03T13:19:50.152.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 1.504
>>>>>>> OMEGACAM.2011-08-03T13:26:17.436.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 1.896
>>>>>>> OMEGACAM.2011-08-03T13:22:41.584.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 2.204
>>>>>>> OMEGACAM.2011-08-03T13:20:30.803.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 2.696
>>>>>>> OMEGACAM.2011-08-03T13:24:11.945.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 2.904
>>>>>>> OMEGACAM.2011-08-03T13:29:11.268.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 3.196
>>>>>>> OMEGACAM.2011-08-03T13:17:32.721.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 3.496
>>>>>>> OMEGACAM.2011-08-03T13:21:14.223.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 3.504
>>>>>>> OMEGACAM.2011-08-03T13:27:01.517.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 4.304
>>>>>>> OMEGACAM.2011-08-03T13:18:20.842.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 4.804
>>>>>>> OMEGACAM.2011-08-03T13:29:55.428.fits 2011-08-03 13:12:39
>>>>>>> FLAT,DOME,DETGAIN 5.104
>>>>>>>
>>>>>>>
>>>>>>> awe> q = (RawFitsData.template.start ==
>>>>>>> dateutil.parser.parse('2011-08-05 15:17:32'))
>>>>>>> awe> for f in q.order_by('DATE_OBS'): print f.filename,
>>>>>>> len(RawDomeFlatFrame.raw_fits_data == f)
>>>>>>> ...
>>>>>>> OMEGACAM.2011-08-05T15:19:26.683.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:20:10.644.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:20:55.594.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:21:41.954.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:22:41.965.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:23:23.495.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:24:07.366.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:24:50.316.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:25:34.247.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:26:15.407.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:27:00.868.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:27:42.958.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:28:26.899.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:29:10.329.fits 32
>>>>>>> OMEGACAM.2011-08-05T15:29:52.919.fits 0
>>>>>>> OMEGACAM.2011-08-05T15:30:40.200.fits 0
>>>>>>> OMEGACAM.2011-08-05T15:31:23.560.fits 0
>>>>>>> OMEGACAM.2011-08-05T15:32:03.781.fits 0
>>>>>>> OMEGACAM.2011-08-05T15:32:47.751.fits 0
>>>>>>> OMEGACAM.2011-08-05T15:33:34.692.fits 0
>>>>>>>
>>>>>>> awe> q = (RawFitsData.template.start ==
>>>>>>> dateutil.parser.parse('2011-08-05 13:38:18'))
>>>>>>> awe> len(q)
>>>>>>> 20
>>>>>>> awe> for f in q.order_by('DATE_OBS'): print f.filename,
>>>>>>> len(RawDomeFlatFrame.raw_fits_data == f)
>>>>>>> ...
>>>>>>> OMEGACAM.2011-08-05T13:40:12.904.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:40:56.774.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:41:41.684.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:42:25.385.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:43:05.425.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:43:45.766.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:44:28.766.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:45:11.576.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:45:51.587.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:46:36.537.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:47:18.658.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:48:03.278.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:48:44.649.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:49:24.999.fits 32
>>>>>>> OMEGACAM.2011-08-05T13:50:10.539.fits 0
>>>>>>> OMEGACAM.2011-08-05T13:50:56.480.fits 0
>>>>>>> OMEGACAM.2011-08-05T13:51:41.590.fits 0
>>>>>>> OMEGACAM.2011-08-05T13:52:25.761.fits 0
>>>>>>> OMEGACAM.2011-08-05T13:53:12.371.fits 0
>>>>>>> OMEGACAM.2011-08-05T13:53:59.232.fits 0
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Comm2011 mailing list
>>>>>>> Comm2011 at astro-wise.org
>>>>>>> http://listman.astro-wise.org/mailman/listinfo/comm2011
>>>>>>>
>>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Dr. Steffen Mieske
>>>>> European Southern Observatory
>>>>> Alonso de Cordova 3107
>>>>> Vitacura, Santiago
>>>>> CHILE
>>>>> Tel.: +56 2 463 3118
>>>>> Fax: +56 2 463 3001
>>>>>
>>>>>
>>>
>>>
>>
>>
--
Dr. Steffen Mieske
European Southern Observatory
Alonso de Cordova 3107
Vitacura, Santiago
CHILE
Tel.: +56 2 463 3118
Fax: +56 2 463 3001
More information about the Comm2011
mailing list