Approved plots for 2009 data that can be shown by ATLAS speakers at conferences and similar events.
Please do not add figures on your own. Contact the responsible project leader in case of questions and/or suggestions.
Two dimensional distribution in the xz plane of primary vertices with at least 4 tracks per vertex for run 142193 at √s = 900GeV. These are the primary vertices used as input to the beam spot maximum likelihood fit. The luminous region parameters extracted using primary vertices with at least 4 tracks per vertex are in excellent agreement with the parameters obtained when requiring at least 10 tracks per vertex (see below). The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Two dimensional distribution in the yz plane of primary vertices with at least 4 tracks per vertex for run 142193 at √s = 900GeV. These are the primary vertices used as input to the beam spot maximum likelihood fit. The luminous region parameters extracted using primary vertices with at least 4 tracks per vertex are in excellent agreement with the parameters obtained when requiring at least 10 tracks per vertex (see below). The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Two dimensional distribution in the xz plane of primary vertices with at least 10 tracks per vertex for run 142193 at √s = 900GeV. Compared to the figure above, the increased number of tracks per vertex improves the primary vertex resolution to about 100μm, so that the RMS width of the transverse distribution is dominated by the luminous size. The transverse luminous size extracted from the maximum likelihood fit is σ(x) = 200μm ± 1μm, the longitudinal size is σ(z) = 41.3mm ± 0.1mm. The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Two dimensional distribution in the yz plane of primary vertices with at least 10 tracks per vertex for run 142193 at√ s = 900GeV. Compared to the figure above, the increased number of tracks per vertex improves the primary vertex resolution to about 100μm, so that the RMS width of the transverse distribution is dominated by the luminous size. The transverse luminous size extracted from the maximum likelihood fit is σ(y) = 279μm ± 1μm, the longitudinal size is σ(z) = 41.3mm ± 0.1mm. The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Two dimensional distribution in the yx plane of primary vertices with at least 10 tracks per vertex for run 142193 at √s = 900GeV. With the requirement of 10 tracks per vertex, the primary vertex resolution of about 100μm is sufficiently small so that the RMS widths are dominated by the luminous size. The transverse luminous sizes extracted from the maximum likelihood fit are σ(x) = 200μm ± 1μm and σ(y) = 279μm ± 1μm . Both the RMS widths of the transverse primary vertex distributions and the extracted transverse luminous sizes show that the average luminous size σ(y) is about 40% larger than σ(x). The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Luminous centroid position in x over the course of run 142193 (LHC fill 911). Both the results from fits to samples of 10 minutes of data (points) and the result from the overall fit (line with error band) are shown. Errors are statistical only. 
eps version of the figure 
Luminous centroid position in y over the course of run 142193 (LHC fill 911). Both the results from fits to samples of 10 minutes of data (points) and the result from the overall fit (line with error band) are shown. Errors are statistical only. 
eps version of the figure 
Luminous centroid position in z over the course of run 142193 (LHC fill 911). Both the results from fits to samples of 10 minutes of data (points) and the result from the overall fit (line with error band) are shown. Errors are statistical only. 
eps version of the figure 
Luminous size σ(x) over the course of run 142193 (LHC fill 911). The data points shown are the results from fits to samples of 10 minutes of data. During the time period shown the luminous size σ(x) decreased by about 20%. Errors are statistical only. 
eps version of the figure 
Luminous size σ(y) over the course of run 142193 (LHC fill 911). The data points shown are the results from fits to samples of 10 minutes of data. During the time period shown the luminous size σ(y) increased by about 20%. Errors are statistical only. 
eps version of the figure 
Luminous size σ(z) over the course of run 142193 (LHC fill 911). The data points shown are the results from fits to samples of 10 minutes of data. Errors are statistical only. 
eps version of the figure 
Two dimensional distribution in the xz plane of primary vertices with at least 10 tracks per vertex for run 142308 at √s = 2.36TeV. Taking into account the degradation of the primary vertex resolution to about 135μm due to the SCT being off, one finds from the RMS width a slightly smaller luminous size σ(x) than at 900 GeV. The reduction in transverse width is much smaller than for σ(y). The longitudinal luminous size is also smaller than at 900 GeV. The transverse luminous size extracted from the maximum likelihood fit is σ(x) = 180μm ± 5μm, the longitudinal size is σ(z) = 27.0mm ± 0.4mm. (Taking the average luminous size from both runs taken at √s = 2.36TeV one finds very similar average transverse luminous sizes in x and y: σ(x) = 165μm ± 4μm and σ(y) = 163μm ± 4μm). The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Two dimensional distribution in the yz plane of primary vertices with at least 10 tracks per vertex for run 142308 at √s = 2.36TeV. In spite of the degradation of the primary vertex resolution to about 135μm due to the SCT being off, one can clearly see the smaller RMS width of the transverse primary vertex distribution that is expected at the higher energy. The longitudinal luminous size is also smaller than at 900 GeV. The transverse luminous size extracted from the maximum likelihood fit is σ(y) = 143μm± 5μm, the longitudinal size is σ(z) = 27.0mm± 0.4mm. The errors shown for the beam spot fit results are statistical only. 
eps version of the figure 
Luminous size σ(x) over the course of run 142308 (LHC fill 916) at √s = 2.36TeV. Both the results from fits to samples of 10 minutes of data (points) and the result from the overall fit (line with error band) are shown. Errors are statistical only. 
eps version of the figure 
Luminous size σ(y) over the course of run 142308 (LHC fill 916) at √s = 2.36TeV. Both the results from fits to samples of 10 minutes of data (points) and the result from the overall fit (line with error band) are shown. Errors are statistical only. 
eps version of the figure 
Luminous size σ(z) over the course of run 142308 (LHC fill 916) at√ s = 2.36TeV. Both the results from fits to samples of 10 minutes of data (points) and the result from the overall fit (line with error band) are shown. Errors are statistical only. 
eps version of the figure 
Evolution of the luminous size in x over several runs at √s = 900GeV from December 10 to December 12. Errors are statistical only. 
eps version of the figure 
Evolution of the luminous size in y over several runs at √s = 900GeV from December 10 to December 12. Errors are statistical only. 
eps version of the figure 
Evolution of the luminous size in z over several runs at √s = 900GeV from December 10 to December 12. Errors are statistical only. 
eps version of the figure 
Shift in z0 of tracks as a result of RF cogging. This plot was made from the best quality collision candidates recorded by ATLAS on 23 November, and selecting tracks with at least 6 SCT hits, more than 10 TRT hits and d0<10mm (distance of closest approach to the nominal beam line). The distribution in red is for luminosity blocks LB<140 (before RF adjustment) and in blue for LB>145 (after RF adjustment). The shift in z0 is consistent with the 12cm expected from beam pick up measurements of the arrival times of the beams (cf. the beam pickup system plots). 
eps version of the figure 
Scatter plot of hits on tracks in the first collision events. TRT hits are shown in blue, and SCT hits in red. The pixel detector was not switched on, and the SCT was running at reduced voltage. The solenoid magnetic field was zero, so tracks are straight. 
eps version of the figure 
Distribution of impact parameter, d0, of tracks in first collision events. 
eps version of the figure

Scatter plot of hits on barrel tracks in the first collision events with stable beams. TRT hits are shown in blue, SCT in red, and Pixel in green. The solenoid is at full field, and the curved tracks can clearly be seen in the TRT. 
eps version of the figure 
Unbiased x residual distribution, integrated over all hitsontracks in the pixel barrel. The x residual is defined as the measured hit position minus the expected hit position from the track extrapolation, projected onto the local x coordinate (the precision coordinate). The distribution is shown for the 900 GeV collisions run 141749, compared with a minimum bias Monte Carlo simulation sample reconstructed with perfect ID alignment. The distributions are normalised to equal area. A track selection requiring pT>2.0 GeV, d0 < 10mm, N(Pix+SCT)Hits>=6 and nTRTHits>=15 is applied. This result is achieved using the alignment constants calculated from cosmic ray samples. 
eps version of the figure 
Unbiased x residual distribution, integrated over all hitsontracks in the SCT barrel. The x residual is defined as the measured hit position minus the expected hit position from the track extrapolation, projected onto the local x coordinate (the precision coordinate). The distribution is shown for the 900 GeV collisions run 141749, compared with a minimum bias Monte Carlo simulation sample reconstructed with perfect ID alignment. The distributions are normalised to equal area. A track selection requiring pT>2.0 GeV, d0 < 10mm, N(Pix+SCT)Hits>=6 and nTRTHits>=15 is applied. This result is achieved using the alignment constants calculated from cosmic ray samples. 
eps version of the figure 
X and Z distribution of position of primary vertices for the run 142165. The pt cut for incoming tracks is lowered to 100 MeV. Tracks contributing less than chi2 value of 15 (per two degrees of freedom) to the reconstructed vertex are accepted to the fit. 
eps version of the figure 
X and Y distribution of position of primary vertices for the run 142165. The pt cut for incoming tracks is lowered to 100 MeV. Tracks contributing less than chi2 value of 15 (per two degrees of freedom) to the reconstructed vertex are accepted to the fit. 
eps version of the figure 
X distribution of position of primary vertices for the run 142165. The pt cut for incoming tracks is lowered to 100 MeV. Tracks contributing less than chi2 value of 15 (per two degrees of freedom) to the reconstructed vertex are accepted to the fit. 
eps version of the figure 
Y distribution of position of primary vertices for the run 142165. The pt cut for incoming tracks is lowered to 100 MeV. Tracks contributing less than chi2 value of 15 (per two degrees of freedom) to the reconstructed vertex are accepted to the fit. 
eps version of the figure 
Z distribution of position of primary vertices for the run 142165. The pt cut for incoming tracks is lowered to 100 MeV. Tracks contributing less than chi2 value of 15 (per two degrees of freedom) to the reconstructed vertex are accepted to the fit. 
eps version of the figure 
The beam spot is determined from an unbinned maximumlikelihood (ML) fit to
the distribution of selected primary vertices. The fit makes use of the
estimated error on the position of each reconstructed vertex, allowing
for a constant error scale factor. The fit can extract both the position
and the size of the beam spot.
This series of six plots show the beam spot position during a run, as a function of time, where time is measured in Luminosity Blocks (each lasting approximately two minutes). Beam spot stability  x position vs Luminosity Block 
eps version of the figure 
Beam spot stability  y position vs Lumi Block 
eps version of the figure 
Beam spot stability  z position vs Lumi Block 
eps version of the figure 
Beam spot stability  x width vs Lumi Block 
eps version of the figure 
Beam spot stability  y width vs Lumi Block 
eps version of the figure 
Beam spot stability  z width vs Lumi Block 
eps version of the figure 
During 2008 and 2009, a large sample of cosmic ray tracks was used to align the Inner Detector. Since these tracks tend to be vertical, they are much more useful for aligning the barrel than the end caps. After only a few days of collision data, the quality of the end cap alignment was dramatically improved, as shown in the following plots, where the position and orientation of end cap disks is updated. The most notable change is an improvement in the alignment residuals for SCT end cap C. The distributions are normalised to equal area. A track selection requiring pT>2.0 GeV, d0 < 10mm and N(Pix+SCT)Hits>=6 is applied.
SCT barrel local x residuals for a minimum bias Monte Carlo sample with perfect alignment (solid blue), collision data using the alignment based on cosmic rays (open black squares) and after a first update using collision data (open red circles). A double Gaus fit is performed but only the mean and sigma of the narrower Gaussian are shown. 
eps version of the figure 
SCT endcap A local x residuals for a minimum bias Monte Carlo sample with perfect alignment (solid blue), collision data using the alignment based on cosmic rays (open black squares) and after a first update using collision data (open red circles). A single Gaussian fit is performed. 
eps version of the figure 
SCT endcap C local x residuals for a minimum bias Monte Carlo sample with perfect alignment (solid blue), collision data using the alignment based on cosmic rays (open black squares) and after a first update using collision data (open red circles). A single Gaussian fit is performed. 
eps version of the figure 
Pixel barrel local x residuals for a minimum bias Monte Carlo sample with perfect alignment (solid blue), collision data using the alignment based on cosmic rays (open black squares) and after a first update using collision data (open red circles). A double Gaus fit is performed but only the mean and sigma of the narrower Gaussian are shown. 
eps version of the figure 
Pixel endcap A local x residuals for a minimum bias Monte Carlo sample with perfect alignment (solid blue), collision data using the alignment based on cosmic rays (open black squares) and after a first update using collision data (open red circles). A single Gaussian fit is performed. 
eps version of the figure 
Pixel endcap C local x residuals for a minimum bias Monte Carlo sample with perfect alignment (solid blue), collision data using the alignment based on cosmic rays (open black squares) and after a first update using collision data (open red circles). A single Gaussian fit is performed. 
eps version of the figure 
The ID alignment has been further improved by using the whole collision data taken in 2009. The validation plots below show the performance of the alignment which is used for the Feb10 reprocessing, using the following alignment tags:
Unbiased residual distribution in x, integrated over all hitsontracks in the pixel barrel for the MC perfect alignment (red) and the current alignment with collision data taken in 2009 (blue). The residual is defined as the measured hit position minus the expected hit position from the track extrapolation. Shown is the projection onto the local x coordinate, which is the precision coordinate. Tracks are selected to have pT > 2 GeV, d0<10mm and ≥ 6 silicon hits. The mean and FWHM/2.35 are quoted. 
eps version of the figure 
Unbiased residual distribution in x, integrated over all hitsontracks in the pixel EC’s for the MC perfect alignment (red) and the current alignment with collision data taken in 2009 (blue). The residual is defined as the measured hit position minus the expected hit position from the track extrapolation. Shown is the projection onto the local x coordinate, which is the precision coordinate. Tracks are selected to have pT > 2 GeV, d0<10mm and ≥ 6 silicon hits. The mean and FWHM/2.35 are quoted.

eps version of the figure 
Unbiased residual distribution in x, integrated over all hitsontracks in the SCT barrel for the MC perfect alignment (red) and the current alignment with collision data taken in 2009 (blue). The residual is defined as the measured hit position minus the expected hit position from the track extrapolation. Shown is the projection onto the local x coordinate, which is the precision coordinate. Tracks are selected to have pT > 2 GeV, d0<10mm and ≥ 6 silicon hits. The mean and FWHM/2.35 are quoted.

eps version of the figure 
Unbiased residual distribution in x, integrated over all hitsontracks in the SCT EC’s for the MC perfect alignment (red) and the current alignment with collision data taken in 2009 (blue). The residual is defined as the measured hit position minus the expected hit position from the track extrapolation. Shown is the projection onto the local x coordinate, which is the precision coordinate. Tracks are selected to have pT > 2 GeV, d0<10mm and ≥ 6 silicon hits. The mean and FWHM/2.35 are quoted.

eps version of the figure 