Logged in as guest
Viewing Misc/74447 Full headers
Private message: yes no
Notes: Notification:
Date: Wed, 2 Jul 2014 14:48:38 GMT From: Alida.Wink@gmail.com To: xmmhelp@sciops.esa.int CC: Alida.Wink@gmail.com Subject: How to convert XMM-Newton time stamps to absolute BJD (TDB)?
Full_Name: Alida Odendaal Submission from: (NULL) (105.226.4.45) To whom it may concern I am doing timing with XMM-Newton archival data, and I want to assign absolute time stamps in BJD (with the TDB time standard) to the events (for event lists), and also the photometric measurements from OM images. I just want to make very sure that I am approaching this in the right manner, and will appreciate your answers / corrections / comments: XMM-Newton events are recorded in seconds (Mission Elapsed Time) after "1998-01-01 00:00:00.000" in the TT time standard (I assume with the geocentre as reference point, as XMM was not yet launched at that time?). For example, consider an event with TIME tag 300152342.110919 in the calibrated event file before barycentric corrections (this is the second event in the 1387_0500860301_EPN_S003_ImagingEvts.ds file of observation 0500860301 of CAL83). As TIMESYS='TT' and TIMEREF='LOCAL', I assume that the TIME tag is recorded with the satellite position as reference point? By performing 'barycen' with correct parameters, the TIME tag is updated to 300152361.707985, which is 19.59706598520279 larger than the uncorrected value (and I assume obtained by correcting for the Roemer delay according to the satellite position, the clock correction, and perhaps also the Shapiro and Einstein delays?). For the purpose of Fourier analysis of one or several XMM-Newton datasets this MET in the TDB system with the solar system barycentre (SSB) as the reference point is adequate. However, I want to go a step further to have the TIME stamps in absolute BJD (TDB) values, therefore I must add the MET (TDB) to the reference of "1998-01-01 00:00:00.000" (TT), equivalent to JD 2450814.5 (TT). Before I can do this, I must convert the reference point itself to a value in BJD (TDB). In other words, as I understand it, I have to convert the time-tag of a hypothetical event from my source recorded at the geocenter at "1998-01-01 00:00:00.000" in the TT time standard, to a time-tag recorded at the SSB in the TDB time standard. So, I first converted "1998-01-01 00:00:00.000" (TT) to "1997-12-31 23:58:56.816" (UTC) (equivalent to JD 2450814.4992687036 (UTC)) using astropy. Then I used the online tool at http://astroutils.astronomy.ohio-state.edu/time/utc2bjd.html with the following as input: the abovementioned UTC, the RA (05 43 33.5) and DEC (-68 22 23) of my object, and the geocentre as reference point. This yielded BJD 2450814.499788368 (TDB). Therefore I take the latter as the equivalent of the mission reference of "1998-01-01 00:00:00.000" (TT) for my specific source. I use astropy to add the barycentrically corrected MET mentioned previously to obtain the BJD (TDB) time stamp for my event: 2454288.485456285. Is this the correct way to do it? I apologize for the length of my question, and also if this information is already available somewhere - however, I could not find my answer in the manuals. I will greatly appreciate if you can comment on any shortcomings in my reasoning above. Thank you in advance for your help. Kind regards Alida Odendaal
From: Anna Lia Longinotti <xmmhelp@sciops.esa.int> To: Alida.Wink@gmail.com Subject: Re: How to convert XMM-Newton time stamps to absolute BJD (TDB)? (PR#74447) Date: Thu Jul 3 09:09:42 2014
Dear Alida, Thanks for your mail. It looks to us that the way to proceed may be simpler and that there is no need to go through UTC: in facts Terrestrial Time (TT) and Barycentric Dynamical Time (TDB) are running at the same speed, different than UTC. The only difference between TT and TDB is the location of the "observer". The problem here is to calculate the Barycentric Julian Date (BJD) corresponding to XMM time stamps from the event lists. As you rightly say, XMM time stamps correspond to Mission Elapsed Time (MET) in seconds from the "zero point" 1998-01-01 00:00:00.000 in Local time reference. The task "barycen" converts these MET_TT to MET_TDB. Therefore after applying the barycentric correction to the event lists, you have the barycentric corrected time stamps in seconds. To express the MET_TDB in BJD you need to know the BJD corresponding to the "zero point", and for that you first need the corresponding Julian Date in Terrestrial Time for the "zero point", let's call it JD_TT[0]. JD_TT[0]=2450814.5 (JD_UTC 2450814.4992687) Considering that: BJD_TT= JD_TT + B_corr and that for this particular case BJD_TT= JD_TT[0] + MET_TT + B_corr and that MET_TDB= MET_TT + B_corr is the output of the "barycen" task then BJD_TT= JD_TT[0] + MET_TDB I hope this is helpful, please contact us in case of further doubts/questions. Kind Regards Anna Lia > > > To whom it may concern > > I am doing timing with XMM-Newton archival data, and I want to assign absolute > time stamps in BJD (with the TDB time standard) to the events (for event lists), > and also the photometric measurements from OM images. I just want to make very > sure that I am approaching this in the right manner, and will appreciate your > answers / corrections / comments: > > XMM-Newton events are recorded in seconds (Mission Elapsed Time) after > "1998-01-01 00:00:00.000" in the TT time standard (I assume with the geocentre > as reference point, as XMM was not yet launched at that time?). > > For example, consider an event with TIME tag 300152342.110919 in the calibrated > event file before barycentric corrections (this is the second event in the > 1387_0500860301_EPN_S003_ImagingEvts.ds file of observation 0500860301 of > CAL83). As TIMESYS='TT' and TIMEREF='LOCAL', I assume that the TIME tag is > recorded with the satellite position as reference point? By performing > 'barycen' with correct parameters, the TIME tag is updated to 300152361.707985, > which is 19.59706598520279 larger than the uncorrected value (and I assume > obtained by correcting for the Roemer delay according to the satellite position, > the clock correction, and perhaps also the Shapiro and Einstein delays?). For > the purpose of Fourier analysis of one or several XMM-Newton datasets this MET > in the TDB system with the solar system barycentre (SSB) as the reference point > is adequate. > > However, I want to go a step further to have the TIME stamps in absolute BJD > (TDB) values, therefore I must add the MET (TDB) to the reference of "1998-01-01 > 00:00:00.000" (TT), equivalent to JD 2450814.5 (TT). Before I can do this, I > must convert the reference point itself to a value in BJD (TDB). In other > words, as I understand it, I have to convert the time-tag of a hypothetical > event from my source recorded at the geocenter at "1998-01-01 00:00:00.000" in > the TT time standard, to a time-tag recorded at the SSB in the TDB time > standard. > > So, I first converted "1998-01-01 00:00:00.000" (TT) to "1997-12-31 > 23:58:56.816" (UTC) (equivalent to JD 2450814.4992687036 (UTC)) using astropy. > Then I used the online tool at > http://astroutils.astronomy.ohio-state.edu/time/utc2bjd.html > with the following as input: the abovementioned UTC, the RA (05 43 33.5) and > DEC (-68 22 23) of my object, and the geocentre as reference point. This > yielded BJD 2450814.499788368 (TDB). Therefore I take the latter as the > equivalent of the mission reference of "1998-01-01 00:00:00.000" (TT) for my > specific source. I use astropy to add the barycentrically corrected MET > mentioned previously to obtain the BJD (TDB) time stamp for my event: > 2454288.485456285. > > Is this the correct way to do it? I apologize for the length of my question, > and also if this information is already available somewhere - however, I could > not find my answer in the manuals. I will greatly appreciate if you can comment > on any shortcomings in my reasoning above. > > Thank you in advance for your help. > > Kind regards > Alida Odendaal > > --------------------------------------- Anna Lia Longinotti XMM-Newton Science Operations Centre ESAC ---------------------------------------
Date: Mon, 7 Jul 2014 14:36:28 +0200 Subject: Re: How to convert XMM-Newton time stamps to absolute BJD (TDB)? (PR#74447) From: Alida Odendaal <alida.wink@gmail.com> To: Anna Lia Longinotti <xmmhelp@sciops.esa.int>
--001a113a2a2e577e1c04fd99b9cf Content-Type: text/plain; charset=UTF-8 Dear Anna Lia Thank you very much for your prompt answer. It seems that I have been overcomplicating matters, but I understand what to do now. Kind regards Alida On 3 July 2014 11:09, Anna Lia Longinotti <xmmhelp@sciops.esa.int> wrote: > Dear Alida, > > Thanks for your mail. > > It looks to us that the way to proceed may be simpler and that there is no > need > to go through UTC: in facts Terrestrial Time (TT) and Barycentric > Dynamical Time > (TDB) are running at the same speed, different than UTC. The only > difference > between TT and TDB is the location of the "observer". > > The problem here is to calculate the Barycentric Julian Date (BJD) > corresponding > to XMM time stamps from the event lists. > > As you rightly say, XMM time stamps correspond to Mission Elapsed Time > (MET) in > seconds from the "zero point" 1998-01-01 00:00:00.000 in Local time > reference. > The task "barycen" converts these MET_TT to MET_TDB. > Therefore after applying the barycentric correction to the event lists, > you have > the barycentric corrected time stamps in seconds. > > To express the MET_TDB in BJD you need to know the BJD corresponding to the > "zero point", and for that you first need the corresponding Julian Date in > Terrestrial Time for the "zero point", let's call it JD_TT[0]. > > JD_TT[0]=2450814.5 (JD_UTC 2450814.4992687) > > Considering that: > > BJD_TT= JD_TT + B_corr > > and that for this particular case > > BJD_TT= JD_TT[0] + MET_TT + B_corr > > and that MET_TDB= MET_TT + B_corr is the output of the "barycen" task > > then > BJD_TT= JD_TT[0] + MET_TDB > > > I hope this is helpful, please contact us in case of further > doubts/questions. > > Kind Regards > Anna Lia > > > > > > > > > > To whom it may concern > > > > I am doing timing with XMM-Newton archival data, and I want to assign > absolute > > time stamps in BJD (with the TDB time standard) to the events (for event > lists), > > and also the photometric measurements from OM images. I just want to > make > very > > sure that I am approaching this in the right manner, and will appreciate > your > > answers / corrections / comments: > > > > XMM-Newton events are recorded in seconds (Mission Elapsed Time) after > > "1998-01-01 00:00:00.000" in the TT time standard (I assume with the > geocentre > > as reference point, as XMM was not yet launched at that time?). > > > > For example, consider an event with TIME tag 300152342.110919 in the > calibrated > > event file before barycentric corrections (this is the second event in > the > > 1387_0500860301_EPN_S003_ImagingEvts.ds file of observation 0500860301 of > > CAL83). As TIMESYS='TT' and TIMEREF='LOCAL', I assume that the TIME tag > is > > recorded with the satellite position as reference point? By performing > > 'barycen' with correct parameters, the TIME tag is updated to > 300152361.707985, > > which is 19.59706598520279 larger than the uncorrected value (and I > assume > > obtained by correcting for the Roemer delay according to the satellite > position, > > the clock correction, and perhaps also the Shapiro and Einstein delays?). > For > > the purpose of Fourier analysis of one or several XMM-Newton datasets > this > MET > > in the TDB system with the solar system barycentre (SSB) as the reference > point > > is adequate. > > > > However, I want to go a step further to have the TIME stamps in absolute > BJD > > (TDB) values, therefore I must add the MET (TDB) to the reference of > "1998-01-01 > > 00:00:00.000" (TT), equivalent to JD 2450814.5 (TT). Before I can do > this, I > > must convert the reference point itself to a value in BJD (TDB). In > other > > words, as I understand it, I have to convert the time-tag of a > hypothetical > > event from my source recorded at the geocenter at "1998-01-01 > 00:00:00.000" > in > > the TT time standard, to a time-tag recorded at the SSB in the TDB time > > standard. > > > > So, I first converted "1998-01-01 00:00:00.000" (TT) to "1997-12-31 > > 23:58:56.816" (UTC) (equivalent to JD 2450814.4992687036 (UTC)) using > astropy. > > Then I used the online tool at > > http://astroutils.astronomy.ohio-state.edu/time/utc2bjd.html > > with the following as input: the abovementioned UTC, the RA (05 43 > 33.5) and > > DEC (-68 22 23) of my object, and the geocentre as reference point. This > > yielded BJD 2450814.499788368 (TDB). Therefore I take the latter as the > > equivalent of the m