Changeset 40567 for trunk/doc/release.2015/ps1.detrend/detrend.tex
- Timestamp:
- Dec 5, 2018, 6:15:07 AM (8 years ago)
- File:
-
- 1 edited
-
trunk/doc/release.2015/ps1.detrend/detrend.tex (modified) (35 diffs)
Legend:
- Unmodified
- Added
- Removed
-
trunk/doc/release.2015/ps1.detrend/detrend.tex
r40563 r40567 177 177 in detail in \cite{2012ApJ...750...99T}. 178 178 179 180 179 The Pan-STARRS 1 Science Survey uses the 1.4 gigapixel GPC1 camera 181 180 with the PS1 telescope on Haleakala Maui to image the sky north of 182 181 $-30^\circ$ declination. The GPC1 camera is composed of 60 orthogonal 183 transfer array (OTA) devices, each of which is an $8\times{}8$ grid of 184 readout cells. The large number of cells parallelizes the readout 185 process, reducing the overhead in each exposure. However, as a 186 consequence, many calibration operations are needed to ensure the 187 response is consistent across the entire seven square degree field of 188 view. 182 transfer array (OTA) devices arranged in an $8\times{}8$ grid, 183 excluding the four corners. Each of the 60 devices is itself an 184 $8\times{}8$ grid of readout cells. The large number of cells 185 parallelizes the readout process, reducing the overhead in each 186 exposure. However, as a consequence, many calibration operations are 187 needed to ensure the response is consistent across the entire seven 188 square degree field of view. 189 189 190 190 \note{DS notes fonts are not consistent for keywords, etc} 191 192 \note{DS: captions need to be clear re: illustrated effect}193 194 \note{need to define PV3 (and PV0-2) here. see datasystem.tx}195 191 196 192 %The Processing Version 3 (PV3) reduction represents the third full … … 262 258 section \ref{sec:discussion}. 263 259 264 \note{describe gpc1 camera layout before the following paragraph} 260 As mentioned above, the GPC1 camera is composed of 60 orthogonal 261 transfer array (OTA) devices arranged in an $8\times{}8$ grid, 262 excluding the four corners. Each of the 60 devices is itself an 263 $8\times{}8$ grid of readout cells consisting of $590 \times 598$ 264 pixels. We label the OTAs by their coordinate in the camera grid in 265 the form `OTAXY', where X and Y each range from 0 - 7, e.g., OTA12 would 266 be the chip in the $(1,2)$ position of the grid. Similarly, we 267 identify the cells as `xyXY' where X and Y again each range from 0 - 268 7. 265 269 266 270 Image products presented in figures have been mosaicked to arrange … … 277 281 and pixel $(590,1)$ to the top left of their position. For mosaics of 278 282 the full field of view, the OTAs are arranged as they see the sky, 279 with the cells arranged as in the single OTA images (Figure \ref{fig:optical ghosts}). The lower left280 corner is the empty location where OTA70 would exist. Toward the 281 right, the OTA labels decrease in $X$ label, with the empty OTA00 282 located in the lower right. The OTA $Y$ labels increase upward in the 283 mosaic.283 with the cells arranged as in the single OTA images (Figure 284 \ref{fig:optical ghosts}). The lower left corner is the empty 285 location where OTA70 would exist. Toward the right, the OTA labels 286 decrease in $X$ label, with the empty OTA00 located in the lower 287 right. The OTA $Y$ labels increase upward in the mosaic. 284 288 285 289 %%\textit{Note: These papers are being placed on the arXiv.org to … … 358 362 \label{sec:dark} 359 363 364 \begin{figure} 365 \centering 366 \begin{minipage}{0.45\hsize} 367 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123o_M_OS_NL_XY23.png} 368 \end{minipage}% 369 \begin{minipage}{0.45\hsize} 370 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123o_to_DARK_XY23.png} 371 \end{minipage} 372 \caption{{\bf Dark Correction:} An example of the dark model application to exposure o5677g0123o, OTA23 (2011-04-26, 43s \gps{} filter). The left panel shows the image data mosaicked to the OTA level, and has had the static mask applied, the overscan subtracted, and the detector non-linearity corrected. The right panel, shows the same exposure with the dark applied in addition to the processing shown on the left, removing the amplifier glows in the cell corners.} 373 \label{fig:dark image} 374 \end{figure} 375 360 376 The dark current in the GPC1 detectors has significant variations 361 377 across each cell. The model we make to remove this signal considers … … 381 397 \subsubsection{Time evolution} 382 398 399 \begin{figure} 400 \centering 401 \includegraphics[width=0.9\hsize,angle=0,clip]{images/B_profile_v1.pdf} 402 \caption{Example showing a profile cut across exposure o5676g0195, 403 OTA67 (2011-04-25, 43s \gps{} filter). The entire first row of 404 cells (xy00-xy07) have had a median calculated along each pixel 405 column on the OTA mosaicked image. Arbitrary offsets have been 406 applied so the curves do not overlap. The top curve (in purple) 407 shows the initial raw profile, with no dark model applied. The 408 next curve (in green) shows the smoother profile after applying 409 the appropriate B-mode dark model. Applying the (incorrect) 410 A-mode dark instead results in the third (blue) curve, which shows 411 a significant increase in gradients across the cells. The fourth 412 (red) curve is the result of applying the PATTERN.CONTINUITY 413 correction along with the B-mode dark model. Although this 414 creates a larger gradient across the mosaicked images, it 415 decreases the cell-to-cell boundary offsets. The bottom (black) 416 curve shows the final image profile after all detrending and 417 background subtraction (no offset applied). The bright source at 418 the cell xy00 to xy01 transition is a result of a large optical 419 ghost which, due to the area covered, increases the median level 420 more than the field stars.} 421 \label{fig:dark switching} 422 \end{figure} 423 383 424 The dark model is not consistently stable over the full survey, with 384 425 significant drift over the course of multiple months. Some of the … … 432 473 as it is correctable with a small number of dark models, this does not 433 474 significantly impact detrending. 434 435 \begin{figure}436 \centering437 \begin{minipage}{0.45\hsize}438 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123o_M_OS_NL_XY23.png}439 \end{minipage}%440 \begin{minipage}{0.45\hsize}441 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123o_to_DARK_XY23.png}442 \end{minipage}443 \caption{An example of the dark model application to exposure o5677g0123o, OTA23 (2011-04-26, 43s \gps{} filter). The left panel shows the image data mosaicked to the OTA level, and has had the static mask applied, the overscan subtracted, and the detector non-linearity corrected. The right panel, shows the same exposure with the dark applied in addition to the processing shown on the left, removing the amplifier glows in the cell corners.}444 \label{fig:dark image}445 \end{figure}446 447 \begin{figure}448 \centering449 \includegraphics[width=0.9\hsize,angle=0,clip]{images/B_profile_ex.png}450 \caption{Example showing a profile cut across exposure o5676g0195, OTA67 (2011-04-25, 43s \gps{} filter). The entire first row of cells (xy00-xy07) have had a median calculated along each pixel column on the OTA mosaicked image. Arbitrary offsets have been applied to shift the curves to not overlap. The top curve (in purple) shows the initial raw profile, with no dark model applied. The next curve (in green) shows the smoother profile after applying the appropriate B-mode dark model. Applying the A-mode dark instead results in the blue curve, which shows a significant increase in gradients across the cells. The orange curve shows the result of the PATTERN.CONTINUITY correction. Although this creates a larger gradient across the mosaicked images, it decreases the cell-to-cell level changes. The final yellow curve shows the final image profile after all detrending and background subtraction, and has not had an offset applied. The bright source at the cell xy00 to xy01 transition is a result of a large optical ghost, which due to the area covered, increases the median level more than the field stars.}451 \label{fig:dark switching}452 \end{figure}453 475 454 476 \subsubsection{Video Dark} … … 496 518 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123o_VIDEODARK_VDim_VDdark_XY22.png} 497 519 \end{minipage} 498 \caption{ An example of the video dark model application to exposure o5677g0123o, OTA22 (2011-04-26, 43s \gps{} filter), which has a video cell located in cell xy16. The left panel shows the image data mosaicked to the OTA level, and has had the static mask applied, the overscan subtracted, the detector non-linearity corrected, and a regular dark applied. The right panel, shows the same exposure with a video dark applied instead of the standard dark. The main impact of this change is the improved correction of the corner glows, which are over subtracted with the standard dark.}520 \caption{{\bf Video Dark:} An example of the video dark model application to exposure o5677g0123o, OTA22 (2011-04-26, 43s \gps{} filter), which has a video cell located in cell xy16. The left panel shows the image data mosaicked to the OTA level, and has had the static mask applied, the overscan subtracted, the detector non-linearity corrected, and a regular dark applied. The right panel, shows the same exposure with a video dark applied instead of the standard dark. The main impact of this change is the improved correction of the corner glows, which are over subtracted with the standard dark.} 499 521 \label{fig:video_darks} 500 522 \end{figure} … … 627 649 measurements and the corresponding measurements on the science image 628 650 provides the scale factor multiplied to the fringe before it is 629 subtracted from the science image. 651 subtracted from the science image. An example of the fringe correction can be seen in Figure~\ref{fig: fringe example}. 630 652 631 653 \begin{figure} … … 637 659 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5220g0025o_fringe_XY53.png} 638 660 \end{minipage} 639 \caption{Example of the \yps{} filter fringe pattern on exposure o5220g0025o OTA53 (\yps{} filter 30s). The left panel shows the OTA mosaic with all detrending except the fringe correction, while the right shows the same including the fringe correction. Both images have been smoothed with a Gaussian with $\sigma = 3$ pixels to highlight the faint and large scale fringe patterns. 640 } 661 \caption{{\bf Fringing:} Example of the \yps{} filter fringe pattern 662 on exposure o5220g0025o OTA53 (\yps{} filter 30s). The left panel 663 shows the OTA mosaic with all detrending except the fringe 664 correction, while the right shows the same including the fringe 665 correction. Both images have been smoothed with a Gaussian with 666 $\sigma = 3$ pixels to highlight the faint and large scale fringe 667 patterns. } 641 668 \label{fig: fringe example} 642 669 \end{figure} … … 725 752 \colhead{Description (static values listed in bold)}} 726 753 \startdata 727 {\bf DETECTOR & 0x0001 &A detector defect is present.} \\728 {\bf FLAT & 0x0002 &The flat field model does not calibrate the pixel reliably.} \\729 {\bf DARK & 0x0004 &The dark model does not calibrate the pixel reliably.} \\730 {\bf BLANK & 0x0008 &The pixel does not contain valid data.} \\731 {\bf CTE & 0x0010 &The pixel has poor charge transfer efficiency.} \\754 {\bf DETECTOR } & {\bf 0x0001} & {\bf A detector defect is present.} \\ 755 {\bf FLAT } & {\bf 0x0002} & {\bf The flat field model does not calibrate the pixel reliably.} \\ 756 {\bf DARK } & {\bf 0x0004} & {\bf The dark model does not calibrate the pixel reliably.} \\ 757 {\bf BLANK } & {\bf 0x0008} & {\bf The pixel does not contain valid data.} \\ 758 {\bf CTE } & {\bf 0x0010} & {\bf The pixel has poor charge transfer efficiency.} \\ 732 759 SAT & 0x0020 & The pixel is saturated. \\ 733 760 LOW & 0x0040 & The pixel has a lower value than expected. \\ … … 840 867 corrector lens), and then back down onto the focal plane. Due to the 841 868 extra travel distance, the resulting source is out of focus and 842 elongated along the radial direction of the camera focal plane. These 843 optical ghosts can be modeled in the focal plane coordinates ($L,M$) 844 which has its origin at the center of the focal plane. In this 845 system, a bright object at location ($L,M$) on the focal plane creates a 846 reflection ghost on the opposite side of the optical axis near 847 ($-L,-M$). The exact location is fit as a third order polynomial in the 848 focal plane $L$ and $M$ directions (as listed in Table 869 elongated along the radial direction of the camera focal 870 plane. Figure~\ref{fig:optical ghosts} shows an example exposure with 871 several prominent optical ghosts. 872 873 These optical ghosts can be modeled in the focal plane coordinates 874 ($L,M$) which has its origin at the center of the focal plane. In 875 this system, a bright object at location ($L,M$) on the focal plane 876 creates a reflection ghost on the opposite side of the optical axis 877 near ($-L,-M$). The exact location is fit as a third order polynomial 878 in the focal plane $L$ and $M$ directions (as listed in Table 849 879 \ref{tab:ghost_centers}). An elliptical annulus mask is constructed 850 880 at the expected ghost location, with the major and minor axes defined … … 887 917 \end{deluxetable} 888 918 889 \begin{deluxetable}{l c}890 \tablecolumns{ 2}919 \begin{deluxetable}{lrr} 920 \tablecolumns{3} 891 921 \tablewidth{0pc} 892 922 \tablecaption{Optical Ghost Magnitude Limits} 893 \tablehead{\colhead{Filter} &\colhead{$m_{inst}$}}923 \tablehead{\colhead{Filter} & \colhead{$m_{inst}$} & \colhead{Approx apparent mag ($3\pi$)}} 894 924 \startdata 895 \gps{} & -16.5 \\896 \rps{} & -20.0 \\897 \ips{} & -25.0 \\898 \zps{} & -25.0 \\899 \yps{} & -25.0 \\900 \wps{} & -20.0 \\925 \gps{} & -16.5 & 12.2 \\ 926 \rps{} & -20.0 & 8.9 \\ 927 \ips{} & -25.0 & 3.7 \\ 928 \zps{} & -25.0 & 3.4 \\ 929 \yps{} & -25.0 & 2.5 \\ 930 \wps{} & -20.0 & 10.2 \\ 901 931 \enddata 902 932 \label{tab:ghost_magnitudes} … … 905 935 \begin{figure} 906 936 \centering 907 \includegraphics[width=0.9\hsize,angle=0,clip]{images/full_fpa_ghosts.jpg} 908 \caption{Example of the full GPC1 field of view illustrating the sources and destinations of optical ghosts on exposure o5677g0123o (2011-04-26, 43s \gps{} filter). The bright stars on OTA33 and OTA44 result in nearly circular ghosts on the opposite OTA. In contrast, the trio of stars on OTA11 result in very elongated ghosts on OTA66.} 937 % \includegraphics[width=0.9\hsize,angle=0,clip]{images/full_fpa_ghosts.jpg} 938 \includegraphics[width=0.9\hsize,angle=0,clip]{images/full_fpa_ghosts.png} 939 \caption{{\bf Ghosts:} Example of the full GPC1 field of view illustrating the sources and destinations of optical ghosts on exposure o5677g0123o (2011-04-26, 43s \gps{} filter). The bright stars on OTA33 and OTA44 result in nearly circular ghosts on the opposite OTA. In contrast, the trio of stars on OTA11 result in very elongated ghosts on OTA66.} 909 940 \label{fig:optical ghosts} 910 941 \end{figure} … … 917 948 telescope. Sources brighter than $m_{inst} = -21$ ($\rps \lesssim 918 949 7.5$) that fell on this reflective surface resulted in light being 919 scattered across the detector surface in a long narrow glint. This 920 surface was physically masked on 2010-08-24, removing the possibility 921 of glints in subsequent data, but images that were taken prior to this 922 date have an advisory dynamic mask constructed when a reference source 923 falls on the focal plane within one degree of the detector edge. This 924 mask is 150 pixels wide, with length $L = 2500 \left(-20 - 925 m_{inst}\right)$ pixels. These glint masks are constructed by 926 selecting sufficiently bright sources in the reference catalog that 927 fall within rectangular regions around each edge of the GPC1 camera. 928 These regions are separated from the edge of the camera by 17 929 arcminutes, and extend outwards an additional degree. 950 scattered across the detector surface in a long narrow glint. 951 Figure~\ref{fig:optical glints} shows an example exposure with 952 a prominent optical glint. 953 954 This reflective surface in the camera was physically masked on 955 2010-08-24, removing the possibility of glints in subsequent data, but 956 images that were taken prior to this date have an advisory dynamic 957 mask constructed when a reference source falls on the focal plane 958 within one degree of the detector edge. This mask is 150 pixels wide, 959 with length $L = 2500 \left(-20 - m_{inst}\right)$ pixels. These 960 glint masks are constructed by selecting sufficiently bright sources 961 in the reference catalog that fall within rectangular regions around 962 each edge of the GPC1 camera. These regions are separated from the 963 edge of the camera by 17 arcminutes, and extend outwards an additional 964 degree. 930 965 931 966 \begin{figure} 932 967 \centering 933 \includegraphics[width=0.9\hsize,angle=0,clip]{images/glint_example_o5379g0103o.jpg} 934 \caption{Example of a glint on exposure o5379g0103o (2010-07-02, 45s \ips{} filter). The source star out of the field of view creates a long reflection that extends through OTA73 and OTA63.} 968 % \includegraphics[width=0.9\hsize,angle=0,clip]{images/glint_example_o5379g0103o.jpg} 969 \includegraphics[width=0.9\hsize,angle=0,clip]{images/full_fpa_glints.png} 970 \caption{{\bf Glints:} Example of a glint on exposure o5379g0103o (2010-07-02, 45s \ips{} filter). The source star out of the field of view creates a long reflection that extends through OTA73 and OTA63.} 935 971 \label{fig:optical glints} 936 972 \end{figure} … … 1219 1255 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0124o_wbt_XY11.png} 1220 1256 \end{minipage} 1221 \caption{ Example of OTA11 cell xy50 on exposures o5677g0123o (left) and o5677g0124o (right). The top panels show the image with all appropriate detrending steps, but without burntool, and the bottom show the same with burntool applied. There is some slight over subtraction in fitting the initial trail, but the impact of the trail is greatly reduced in both exposures.}1257 \caption{{\bf Persistent Charge:} Example of OTA11 cell xy50 on exposures o5677g0123o (left) and o5677g0124o (right). The top panels show the image with all appropriate detrending steps, but without burntool, and the bottom show the same with burntool applied. There is some slight over subtraction in fitting the initial trail, but the impact of the trail is greatly reduced in both exposures.} 1222 1258 \label{fig:burntool images} 1223 1259 \end{figure} … … 1226 1262 \begin{figure} 1227 1263 \centering 1228 \begin{minipage}{0.45\hsize} 1229 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123o_XY11_bt_trail.png} 1230 \end{minipage}% 1231 \begin{minipage}{0.45\hsize} 1232 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0124o_XY11_bt_trail.png} 1233 \end{minipage} 1234 1235 \caption{Example of a profile cut along the y-axis through a bright star on exposure o5677g0123o OTA11 in cell xy50 (left panel) and on the subsequent exposure o5677g0124o (right panel). In both figures, the green points show the image corrected with all appropriate detrending steps, but without burntool applied, illustrating the amplitude of the persistence trails. The red points show the same data after the burntool correction, which reduces the impact of these features. Both exposures are in the \gps{} filter with exposure times of 43s} 1264 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5677g0123n4o_XY11_bt_trail.pdf} 1265 1266 \caption{{\bf Burntool Correction:} Example of a profile cut along 1267 the y-axis through a bright star on exposure o5677g0123o OTA11 in 1268 cell xy50 (left panel) and on the subsequent exposure o5677g0124o 1269 (right panel). In both figures, the blue pluses show the image 1270 corrected with all appropriate detrending steps, but without 1271 burntool applied, illustrating the amplitude of the persistence 1272 trails. The red circles show the same data after the burntool 1273 correction, which reduces the impact of these features. Both 1274 exposures are in the \gps{} filter with exposure times of 43s} 1275 1236 1276 \label{fig:burntool plot} 1237 1277 \end{figure} … … 1261 1301 effects. 1262 1302 1303 % An example of this data is shown in Figure~\ref{fig: nonlinearity}. 1304 1263 1305 We store the average flux measurement and deviation from the linear 1264 1306 fit for each exposure time for each region on all detector cells in 1265 the linearity detrend look-up tables. An example of this data is 1266 shown in Figure~\ref{fig: nonlinearity}. When this correction is 1307 the linearity detrend look-up tables. When this correction is 1267 1308 applied to science data, these lookup tables are loaded, and a linear 1268 1309 interpolation is performed to determine the correction needed for the … … 1284 1325 rejected. 1285 1326 1286 \begin{figure} 1287 \centering 1288 \includegraphics[width=0.9\hsize,angle=0,clip]{images/linearity_XY27_xy16.png} 1289 \caption{Example of the linearity correction as a fraction of observed flux for OTA27, cell xy16.} 1290 \label{fig: nonlinearity} 1291 \end{figure} 1327 % this figure does not really clarify anything 1328 % \begin{figure} 1329 % \centering 1330 % \includegraphics[width=0.9\hsize,angle=0,clip]{images/linearity_XY27_xy16.png} 1331 % \caption{Example of the linearity correction as a fraction of observed flux for OTA27, cell xy16.} 1332 % \label{fig: nonlinearity} 1333 % \end{figure} 1292 1334 1293 1335 \subsection{Pattern correction} … … 1336 1378 sky noise does not fully obscure the row-by-row noise. 1337 1379 1380 %% GPC1 tuning describe in email from Peter Onaka 2009.11.30, 1381 %% with notes in GPC1TuningTestLog.pdf 1382 1338 1383 This correction was required on all cells on all OTAs prior to 1339 2009-12-01, at which point a modification of the camera electronics1340 reduced the scale of the row-by-row offsets for the majority of the 1341 OTAs. \czw{describe modification} As a result, we only apply this1342 correction to the cells where it is still necessary, as shown in 1343 Figure \ref{fig: pattern row cells}. A list of these cells is in 1344 Table\ref{tab:pattern_row_cells}.1384 2009-12-01, at which point a modification of the camera clocking phase 1385 delays reduced the scale of the row-by-row offsets for the majority of 1386 the OTAs. As a result, we only apply this correction to the cells 1387 where it is still necessary, as shown in Figure \ref{fig: pattern row 1388 cells}. A list of these cells is in Table 1389 \ref{tab:pattern_row_cells}. 1345 1390 1346 1391 Although this correction largely resolves the row-by-row offset issue … … 1391 1436 \includegraphics[width=0.9\hsize,angle=0,clip]{images/o5379g0103o_wpt_XY57.png} 1392 1437 \end{minipage} 1393 \caption{ Example of the PATTERN.ROW correction on exposure o5379g0103o OTA57 cell xy01 (\ips{} filter 45s). The left panel shows the cell with all appropriate detrending except the PATTERN.ROW, and the right shows the same cell with PATTERN.ROW applied. The correction reduces the correlated noise on the right side, which is most distant from the read out amplifier. There is a slight over subtraction along the rows near the bright star.}1438 \caption{{\bf Correlated Noise:} Example of the PATTERN.ROW correction on exposure o5379g0103o OTA57 cell xy01 (\ips{} filter 45s). The left panel shows the cell with all appropriate detrending except the PATTERN.ROW, and the right shows the same cell with PATTERN.ROW applied. The correction reduces the correlated noise on the right side, which is most distant from the read out amplifier. There is a slight over subtraction along the rows near the bright star.} 1394 1439 \label{fig: pattern row example} 1395 1440 \end{figure} … … 1657 1702 the input OTA images, with some reduction in accuracy. 1658 1703 1704 Examples of a warped signal, variance, and mask image are illustrated 1705 in Figures~\ref{fig:warp image} through \ref{fig:warp mask}. 1706 1659 1707 \begin{figure} 1660 1708 \centering 1661 1709 \includegraphics[width=0.9\hsize,angle=0,clip]{images/warp_2046019_sci.png} 1662 \caption{Example of the warp image for skycell skycell. 2047.0051663 centered at ($\alpha,\delta$) = (1 79.763, 32.1899) for exposure1664 o 4985g0073o, (2009-06-03, 30s \zps{} filter). The data from six1710 \caption{Example of the warp image for skycell skycell.1146.095 1711 centered at ($\alpha,\delta$) = (11.934, -4.197) for exposure 1712 o5104g0266o, (2009-09-30, 60s \rps{} filter). The data from four 1665 1713 OTAs contribute to this image, although they are all truncated by 1666 1714 the skycell boundaries. This skycell image is aligned such that 1667 1715 north points to the top of the image, and east to the left. The 1668 contributing OTAs are from the right half of the detector, with 1669 OTA24 contributing the most pixels, and originally have the 1670 positive y axis pointing to the southwest in this warped image, 1671 with the positive x axis to the northwest.} 1716 contributing OTAs are OTA20, OTA21, OTA30, OTA31.} 1672 1717 \label{fig:warp image} 1673 1718 \end{figure} … … 1677 1722 \includegraphics[width=0.9\hsize,angle=0,clip]{images/warp_2046019_var.png} 1678 1723 \caption{Example of the warp variance image for skycell 1679 skycell. 2047.005 of exposure o4985g0073o, the same as in Figure1724 skycell.1146.095 of exposure o5104g0266o, the same as in Figure 1680 1725 \ref{fig:warp image}. This variance map retains information about 1681 1726 the higher flux levels that were found in burntool corrected 1682 1727 persistence trails, which appear here as streaks along the 1683 original OTA y axis. The amplifier glows that are corrected in1684 the dark model are also more visible in the corners of the cells1685 in OTA24. As both of these effects are corrected in the science1728 original OTA y axis. The dark glows that are corrected in the 1729 dark model are also more visible, especially on certain cell 1730 edges. As both of these effects are corrected in the science 1686 1731 image, there are no significant features visible there.} 1687 1732 \label{fig:warp variance} … … 1691 1736 \centering 1692 1737 \includegraphics[width=0.9\hsize,angle=0,clip]{images/warp_2046019_mask.png} 1693 \caption{Example of the warp mask image for skycell skycell. 2047.0051694 of exposure o 4985g0073o, the same as in Figure \ref{fig:warp1738 \caption{Example of the warp mask image for skycell skycell.1146.095 1739 of exposure o5104g0266o, the same as in Figure \ref{fig:warp 1695 1740 image}. This mask image shows the many small defects removed 1696 1741 from the image, along with larger advisory trails on corrected 1697 1742 burntool trails. The saturated cores of the bright stars are also 1698 masked, along with the diffraction spikes found on these stars. 1699 In addition OTA24 shows the precautionary crosstalk bleed masks 1700 for the two brightest stars applied to all cells within the same 1701 row.} 1743 masked, along with the diffraction spikes found on these stars. A 1744 ghost mask is visible just below the center as an elliptical 1745 region. 1746 % In addition OTA24 shows the precautionary crosstalk bleed masks 1747 % for the two brightest stars applied to all cells within the same 1748 % row. 1749 \label{fig:warp mask} 1750 } 1702 1751 \end{figure} 1703 1752 … … 1716 1765 sources. 1717 1766 1767 As part of the stacking process, the collection of input pixels for a 1768 given output stack pixel are checked for consistency and outliers are 1769 rejected. Varying image quality makes a pixel-by-pixel check for 1770 outliers challenging in the vicinity of brighter stars. Pixels in the 1771 wings of bright stars are liable to be over-rejected as the image 1772 quality changes because the flux observed at a given position varies 1773 as its location on the stellar profile changes. To avoid this effect, 1774 we convolve all input images to a common PSF before making the 1775 pixel-by-pixel comparison. This PSF-matching technique allows us to 1776 detect inconsistent pixels even in the sensitive wings of bright objects. 1777 1718 1778 For the $3\pi$ survey, the stacked image is comprised of all warp 1719 1779 frames for a given skycell in a single filter. The source catalogs 1720 1780 and image components are loaded into the \IPPprog{ppStack} program to 1721 1781 prepare the inputs and stack the frames. 1722 1723 \note{need to point out that we are convolving to a matched PSF}1724 1782 1725 1783 Once all files are ingested, the first step is to measure the size and … … 1770 1828 convolution kernels can be calculated for each image. To calculate 1771 1829 the convolution kernels, we use the algorithm described by 1772 \cite{1998ApJ...503..325A} and \cite{2000.alard} to perform optimal 1773 image subtraction. These `ISIS' kernels \citep[named after the 1774 software package described by][]{1998ApJ...503..325A} are used with 1775 FWHM values of 1.5, 3.0, and 6.0 pixels and polynomial orders of 6, 4, 1776 and 2. Regions around the sources identified in the input images are 1777 extracted, convolved with the kernel, and the residual with the target 1778 PSF used to update the parameters of the kernel via least squares 1779 optimization. Stamps that significantly deviate are rejected, 1780 although the squared residual difference will increase with increasing 1781 source flux. To mitigate this effect, a parabola is fit to the 1782 distribution of squared residuals as a function of source flux. 1783 Stamps that deviate from this fit by more than $2.5\sigma$ are 1784 rejected, and not used on further kernel fit iterations. This process 1785 is repeated twice, and the final convolution kernel is returned. 1830 \cite{1998ApJ...503..325A} and extended by \cite{2000A&AS..144..363A} 1831 to perform optimal image subtraction. These `ISIS' kernels 1832 \citep[named after the software package described 1833 by][]{1998ApJ...503..325A} are used with FWHM values of 1.5, 3.0, 1834 and 6.0 pixels and polynomial orders of 6, 4, and 2. Regions around 1835 the sources identified in the input images are extracted, convolved 1836 with the kernel, and the residual with the target PSF used to update 1837 the parameters of the kernel via least squares optimization. Stamps 1838 that significantly deviate are rejected, although the squared residual 1839 difference will increase with increasing source flux. To mitigate 1840 this effect, a parabola is fit to the distribution of squared 1841 residuals as a function of source flux. Stamps that deviate from this 1842 fit by more than $2.5\sigma$ are rejected, and not used on further 1843 kernel fit iterations. This process is repeated twice, and the final 1844 convolution kernel is returned. 1786 1845 1787 1846 This convolution may change the image flux scaling, so the kernel is … … 1812 1871 identify discrepant input values that should be excluded. 1813 1872 1814 \note{clarify 'should' below, e.g., with a histogram}1815 1816 1873 If only a single input is available, the initial stack contains the 1817 1874 value from that single input. If there are only two inputs, the 1818 average of the two is used. These cases shouldoccur only rarely in1875 average of the two is used. These cases are expected to occur only rarely in 1819 1876 the $3\pi$ survey, as there are many input exposures that overlap each 1820 1877 point on the sky. For the more common case of three or more inputs, a … … 1879 1936 distribution is likely to be unimodal), or if there are insufficient 1880 1937 inputs for this mixture model analysis, the input values are passed to 1881 an Olympic \note{define} weighted mean calculation. We reject $20\%$ of the number 1882 of inputs through this process. The number of bad inputs is set to 1883 $N_\mathrm{bad} = 0.2 \times N_\mathrm{input} + 0.5$, with the 0.5 term 1884 ensuring at least one input is rejected. This number is further 1885 separated into the number of low values to exclude, $N_\mathrm{low} = 1886 N_\mathrm{bad} / 2$, which will default to zero if there are few 1887 inputs, and $N_\mathrm{high} = N_\mathrm{low} - N_\mathrm{bad}$. 1888 After sorting the input values to determine which values fall into the 1889 low and high groups, the remaining input values are used in a weighted 1890 mean using the image weights above. 1938 an ``Olympic'' weighted mean calculation (both the lowest and highest 1939 values are ignored in calculating the weighted mean). We reject 1940 $20\%$ of the number of inputs through this process. The number of 1941 bad inputs is set to $N_\mathrm{bad} = 0.2 \times N_\mathrm{input} + 1942 0.5$, with the 0.5 term ensuring at least one input is rejected. This 1943 number is further separated into the number of low values to exclude, 1944 $N_\mathrm{low} = N_\mathrm{bad} / 2$, which will default to zero if 1945 there are few inputs, and $N_\mathrm{high} = N_\mathrm{low} - 1946 N_\mathrm{bad}$. After sorting the input values to determine which 1947 values fall into the low and high groups, the remaining input values 1948 are used in a weighted mean using the image weights above. 1891 1949 1892 1950 A systematic variance term is necessary to correctly scale how … … 1924 1982 pixels. The ISIS kernel used in the previous step is again used to 1925 1983 determine the largest square box that does not exceed the limit of 1926 $0.25 \times \sum_{x,y} kernel^2$. This square box is then convolved with1927 the rejected pixel mask to reject the neighboring pixels. This final 1928 list of rejected pixels is passed to the final combination, which 1929 creates the final stack values from the weighted mean of the1984 $0.25 \times \sum_{x,y} kernel^2$. This square box is then convolved 1985 with the rejected pixel mask to reject the neighboring pixels. This 1986 final list of rejected pixels is passed to the final combination, 1987 which creates the final stack values from the weighted mean of the 1930 1988 non-rejected pixels. Six total images are constructed for this final 1931 1989 stack: the image, its variance, a mask, a map of the exposure time per 1932 1990 pixel, that exposure time map weighted by the input image weight, and 1933 a map of the number of inputs per pixel. 1934 1935 These convolved stack products are not retained, as the convolution is 1991 a map of the number of inputs per pixel. Examples of each output 1992 image type for the stacking process are shown in 1993 Figures~\ref{fig:stack image} through \ref{fig:stack exp wtimage}. 1994 1995 The convolved stack products are not retained, as the convolution is 1936 1996 used to ensure that the pixel rejection uses seeing-matched images. 1937 1997 This prevents any differences in the input PSF shape from skewing the … … 1980 2040 \centering 1981 2041 \includegraphics[width=0.9\hsize,angle=0,clip]{images/stack_3956997_sci.png} 1982 \caption{Example of the stack image for skycell skycell. 2047.0051983 centered at ($\alpha,\delta$) = (1 79.763, 32.1899) in the \zps{}1984 filter, stack\_id 3 775944. This stack includes 25 input images,1985 including o 4985g0073othe warp image in Figure \ref{fig:warp1986 image}, and has a combined exposure time of 870s. Combining2042 \caption{Example of the stack image for skycell skycell.1146.095 2043 centered at ($\alpha,\delta$) = (11.934, -4.197) in the \rps{} 2044 filter, stack\_id 3956997. This stack includes 39 input images 2045 including o5104g0266o, the warp image in Figure \ref{fig:warp 2046 image}, and has a combined exposure time of 1880s. Combining 1987 2047 such a large number of input images removes the inter-cell and 1988 2048 inter-chip gaps, providing a fully populated image. In addition, … … 1997 2057 \includegraphics[width=0.9\hsize,angle=0,clip]{images/stack_3956997_mask.png} 1998 2058 \caption{Example of the stack mask image for skycell 1999 skycell.2047.005 centered at ($\alpha,\delta$) = (179.763, 2000 32.1899) in the \zps{} filter, stack\_id 3775944. The entire 2001 frame is largely unmasked after combining inputs, with the only 2002 remaining masks falling on the cores of bright stars, and in small 2003 regions around the brightest objects where the overlapping of 2004 diffraction spike masks have removed all inputs.} 2005 2059 skycell.1146.095 centered at ($\alpha,\delta$) = (11.934, -4.197) 2060 in the \rps{} filter, stack\_id 3956997. The entire frame is 2061 largely unmasked after combining inputs, with the only remaining 2062 masks falling on the cores of bright stars, and in small regions 2063 around the brightest objects where the overlapping of diffraction 2064 spike masks have removed all inputs.} 2006 2065 \label{fig:stack mask image} 2007 2066 \end{figure} … … 2010 2069 \centering 2011 2070 \includegraphics[width=0.9\hsize,angle=0,clip]{images/stack_3956997_var.png} 2012 \caption{Example of the stack variance image for skycell 2013 skycell. 2047.005 centered at ($\alpha,\delta$) = (179.763,2014 32.1899) in the \zps{} filter, stack\_id 3775944. The variance2071 \caption{Example of the stack variance image for skycell 2072 skycell.1146.095 centered at ($\alpha,\delta$) = (11.934, -4.197) 2073 in the \rps{} filter, stack\_id 3956997. The variance 2015 2074 map for this stack is reasonably smooth, with the mottled pattern 2016 2075 from the inter-chip and inter-cell gaps printing through. Some … … 2025 2084 \includegraphics[width=0.9\hsize,angle=0,clip]{images/stack_3956997_num.png} 2026 2085 \caption{Example of the stack number image for skycell 2027 skycell. 2047.005 centered at ($\alpha,\delta$) = (179.763,2028 32.1899) in the \zps{} filter, stack\_id 3775944. This map shows2086 skycell.1146.095 centered at ($\alpha,\delta$) = (11.934, -4.197) 2087 in the \rps{} filter, stack\_id 3956997. This map shows 2029 2088 the number of inputs contributing to each pixel of the output 2030 2089 stack. Again, the pattern of the inter-chip and inter-cell gaps 2031 is visible, along with the mask pattern of regions with CTE 2032 problems (visible in the upper right corner). } 2090 is visible, along with other mask features. } 2033 2091 2034 2092 \label{fig:stack num image} … … 2039 2097 \includegraphics[width=0.9\hsize,angle=0,clip]{images/stack_3956997_exp.png} 2040 2098 \caption{Example of the stack exposure time image for skycell 2041 skycell.2047.005 centered at ($\alpha,\delta$) = (179.763, 2042 32.1899) in the \zps{} filter, stack\_id 3775944. As all input 2043 warps had the same 30s exposure time, this map essentially 2044 recreates the number map, with units of seconds of exposure 2045 instead of number of inputs contributing to a given pixel.} 2046 2099 skycell.1146.095 centered at ($\alpha,\delta$) = (11.934, -4.197) 2100 in the \rps{} filter, stack\_id 3956997. Since the input 2101 exposures had exposures times of 40 and 60 seconds, the pattern 2102 observed here similar to, but subtly different from the number 2103 map.} 2047 2104 \label{fig:stack exp image} 2048 2105 \end{figure} … … 2052 2109 \includegraphics[width=0.9\hsize,angle=0,clip]{images/stack_3956997_expwt.png} 2053 2110 \caption{Example of the stack weighted exposure image for skycell 2054 skycell. 2047.005 centered at ($\alpha,\delta$) = (179.763,2055 32.1899) in the \zps{} filter, stack\_id 3775944. This map shows2111 skycell.1146.095 centered at ($\alpha,\delta$) = (11.934, -4.197) 2112 in the \rps{} filter, stack\_id 3956997. This map shows 2056 2113 the weighted average exposure time, as described in the text. It 2057 2114 is similar to the simple exposure time map, but shows how some 2058 2115 input exposures have their contributions weighted down due to the 2059 2116 observed larger image variances.} 2060 2061 2062 2117 \label{fig:stack exp wtimage} 2063 2118 \end{figure} … … 2226 2281 University (ELTE), and the Los Alamos National Laboratory. 2227 2282 2228 \note{ApJ, etc latex macros have an extra comma}2229 2230 2283 \bibliography{lib}{} 2231 2284 \bibliographystyle{apj}
Note:
See TracChangeset
for help on using the changeset viewer.
