questions about srctool

[baotong]

Hello everyone,

I am extracting the spectra of some diffuse emission from a large area (e.g. 4 deg×2deg) and I am a bit confused about some parameters automatically derived by the *srctool* method.

1)I find that the photon counts given by the *srctool* are less than the value which I directly count. The question is how *srctool* calculate the GTI and where to find the GTI information?
B.T.W. It seems that *srctool* cannot correctly calculate the exposure? For my region, the EXPOSURE keyword in derived spectrum file is about 2000 s but the exposure map shows that the exposure in any point is about 100 s. I guess that it may relate to the calculation of exposure from GTI. The GTI generated by *srctool* may describe photons in the total region but actually these photons are not observed simultaneously.

2) Also I am confused by the BACKSCAL keyword. As I know, this keyword describes the relative area of spectrum-extracting region. But when I use a 5' circular source region and a 10'-15' annulus background region, in the source spectrum, the REGAREA is ~ 0.02 deg^2 which matches the absolute area. So the BACKSCAL is ~ 0.015 which is relative to the corresponding area of CCD size? But the REGAREA and BACKSCAL of the background spectrum is only 0.0037 and 0.0026. It seems that the description on the eRASS task website "intersection of the source region with the detector during the source-specific GTI" cannot fully explain this. Since the small BACKSCAL value will lead to a background-dominated source spectra, which seems not the case here.
Also I want to estimate the contribution from instrumental background. So I generate a spectrum file using the information provided by FWC data. Should I modify the BACKSCAL parameters here?

3) Will the *exttype* parameter in *srctool* affect the results? I guess that this parametes will control the weight of each point in my region to calculate the mean exposure and then ARF. Is there any other differences may bring by the change of *exttype*?

It seems that the description about those questions above may be too brief for me to understand it. I hope there can be a more detailed description about the calculation of parameters like BACKSCAL, GTI, EXPOSURE ... in the future.

Many thanks.

Ziqian Hua

[JeremySanders]

Hi Ziqian Hua

1)I find that the photon counts given by the *srctool* are less than the value which I directly count. The question is how *srctool* calculate the GTI and where to find the GTI information?
B.T.W. It seems that *srctool* cannot correctly calculate the exposure? For my region, the EXPOSURE keyword in derived spectrum file is about 2000 s but the exposure map shows that the exposure in any point is about 100 s. I guess that it may relate to the calculation of exposure from GTI. The GTI generated by *srctool* may describe photons in the total region but actually these photons are not observed simultaneously.

As stated in the documentation, "The EXPOSURE keyword of the spectrum is given by the total time the source is on a detector (providing NOSRCGTI is not used), after deadtime corrections.". The EXPOSURE is the total amount of time the source is in the field of view. This is different from the expmap exposure, which is the equivalent on-axis exposure for a particular sky position. srctool corrects the ARF for the fact that only part of a source may be visible at any one time (computed via the source model).

2) Also I am confused by the BACKSCAL keyword. As I know, this keyword describes the relative area of spectrum-extracting region. But when I use a 5' circular source region and a 10'-15' annulus background region, in the source spectrum, the REGAREA is ~ 0.02 deg^2 which matches the absolute area. So the BACKSCAL is ~ 0.015 which is relative to the corresponding area of CCD size? But the REGAREA and BACKSCAL of the background spectrum is only 0.0037 and 0.0026. It seems that the description on the eRASS task website "intersection of the source region with the detector during the source-specific GTI" cannot fully explain this. Since the small BACKSCAL value will lead to a background-dominated source spectra, which seems not the case here.

I don't understand the comment here, but maybe this is related to the misunderstanding of the EXPOSURE above. The BACKSCAL gives the average source area visible on the detector during the exposure. Effectively, the BACKSCAL gives the scaling factor for the instrumental background, while the AREASCAL gives the scaling for the XRB background model.

Also I want to estimate the contribution from instrumental background. So I generate a spectrum file using the information provided by FWC data. Should I modify the BACKSCAL parameters here?

The BACKSCAL should be used to normalise the FWC data (BACKSCAL is in deg^2).

3) Will the *exttype* parameter in *srctool* affect the results? I guess that this parametes will control the weight of each point in my region to calculate the mean exposure and then ARF. Is there any other differences may bring by the change of *exttype*?

As stated above, the ARF takes account that only part of the source is visible at any one time via the source model. See the description for the ARF file.

Jeremy

[baotong]

Hi Jeremy,

Thank you for your reply. It is very helpful.

There is still one question. When I extract the spectra from a large area, the photon counts given by *srctool* do not match the value I directly count from the event file. The total number of photon counts in the spectrum file is about 6,000 and when I consider the same region, I count to ~38,000 photons. What will lead to this discrepancy? I check all 38,000 photons I find and they are all located within the GTI given by the spectrum file.

B.T.W is there any way to find out if one specific point is on a detector or not? In other words, for exmaple, the total exposure as shown in the keywords is 2,800 s for my region. But for any location in this region, the GTI should only sum up to ~40*6=240 s (am I right?), how can I check the time range of this 240 s?

Cheers,

Ziqian Hua

[JeremySanders]

There is still one question. When I extract the spectra from a large area, the photon counts given by *srctool* do not match the value I directly count from the event file. The total number of photon counts in the spectrum file is about 6,000 and when I consider the same region, I count to ~38,000 photons. What will lead to this discrepancy? I check all 38,000 photons I find and they are all located within the GTI given by the spectrum file.

I don't know why the number of photons should be different, unless there is a problem with the regions. To do a check you could use todo='ALL' and inspect the event files produced (these are produced only for the individual TMs). The event files are not for scientific use, but are good to check the extraction looks ok.

B.T.W is there any way to find out if one specific point is on a detector or not? In other words, for exmaple, the total exposure as shown in the keywords is 2,800 s for my region. But for any location in this region, the GTI should only sum up to ~40*6=240 s (am I right?), how can I check the time range of this 240 s?

The EXPOSURE is the total time where any part of of the source region is on the detector. It should also be the same as the per-source GTI produced in the output files (ignoring DEADCOR corrections and weighting the TM times). I don't know any way of checking this easily, though you can test roughly if a region is on the detector at any time by comparing the distance from the CORRATTx pointing detection and seeing if it is less than 30 arcmin.

Jeremy