Alex S. Hill & L. Matthew Haffner · 2011. 3. 8. · poster by Haffner et al. At each position, we...
1
Ionized gas in the Smith Cloud Alex S. Hill & L. Matthew Haffner University of Wisconsin-Madison Abstract We present new observations of 10 4 K ionized gas in the Smith Cloud, a high velocity cloud thought to be interacting with the interstellar medium of the Galaxy 3 kpc below the midplane. We have obtained Hα, [N II], and [S II] spectra of the cloud with the Wisconsin H-Alpha Mapper (WHAM). Previously, we have shown that the ionized component has a mass of > 10 6 M ⊙ (comparable to the mass of the neutral gas in the cloud), with relatively bright Hα emission in di- rections with faint H I 21 cm emission. Detections of [S II] and [N II] constrain the temperature and metallicity of the infalling ionized gas. We report new WHAM [N II] and [S II] spectra in four directions along the axis of the cloud in order to explore the kinematics and metallicity distribution across the cloud. Introduction The infall of gas is a significant driver of galaxy evolution, and is thought to be partially responsible for maintaining the observed star formation rate in the Milky Way of 1 M ⊙ yr -1 over the past ~10 Gyr with roughly constant metal- licity. High velocity clouds (HVCs) are likely examples of the active infall of material, often with low metalicities. The Smith Cloud is perhaps the best studied HVC, with all three components of both the position in the Galaxy and the velocity well known. Stellar absorption distance brackets place the cloud at a heliocentric distance of 9.8 – 15.1 kpc (Wakker et al. 2008), or 2.9 ± 0.6 kpc below the Galactic mid- plane and 7.6 kpc from the Galactic center. Lockman et al. (2008) argued that the cloud is interacting with the inter- stellar medium of the Galactic disk. The cloud has a come- tary morphology with a bright core and a diffuse, trailing tail., moving towards the plane at 73 ± 26 km s -1 with an H I mass of > 10 6 M ⊙ and a comparable H + mass. This has al- lowed considerable speculation about the origins of the cloud (Bland-Hawthorn et al. 1998, Lockman et al. 2008, Nichols & Bland-Hawthorn 2009). However, relatively little is known about the physical conditions within the cloud. Recently, we have presented Wisconsin Hα Mapper (WHAM) observations of ionized gas in the Smith Cloud (Hill et al. 2009). In that work, we presented a map of the Hα emission from the cloud (Fig. 1), finding considerable Hα emission and H + mass in regions with relatively faint H I. We also obtained [N II] λ6583 and [S II] λ6716 spectra in the direction with the brightest Hα and H I emission (Fig. 2). With these data, we estimated the nitrogen abundance of the ionized gas in the cloud to be 0.15 – 0.44 times Solar assum- ing a temperature in the ionized gas of 12,000 – 8,000 K. Here, we report new [S II] and [N II] observations at four lo- cations in the cloud with the aim of constraining the kine- matics and abundance distribution of the ionized gas across the cloud. Observations We used WHAM (see adjacent poster by Haffner et al.) to obtain Hα, [N II] λ6583, and [S II] λ6716 spectra. These ob- servations were obtained in 2009 from the Cerro Tololo In- teramerican Observatory (CTIO). Pointed spectra: For all pointed spectra, we alternated 120 s ON source observations and 120 s OFF source observations for a total of 480 s ON source and 480 s OFF source. Hausen et al. (2002) constructed an empirical template of the faint lines near Hα, which are stable relative to each other but vary in intensity, partly dependent upon the zenith angle. To identify the appropriate level of the atmospheric template for our observations, we fit the OFF spectra of each line to a single atmospheric template, which we then subtracted from the ON spectra to obtain spectra consisting purely of astro- nomical emission. The atmospheric template used here is the template we used at Kitt Peak; if the atmospheric spectrum at CTIO is different, this will introduce an uncertainty in the observed velocity and intensity of faint lines. For Hα spec- tra, we also subtracted the geocornal Hα line. The multi- wavelength spectra reported for the tip of the cloud by Hill et al. (2009) are shown in Fig. 2. We show H I spectra from Lockman et al. in four directions along the main axis of the cloud in Fig. 3. Pointed [N II] and [S II] spectra obtained at CTIO in the same four directions are shown in Figs. 4 and 5. Maps: We obtained a spectroscopic Hα map of the Smith Cloud using the “block” method described in the nearby poster by Haffner et al. At each position, we obtained an Hα spectrum with an exposure time of 60 s. The resulting spec- tra were fit with three Gaussian components, each with a fixed width of 30 km s -1 , corresponding to local gas, the Sag- ittarius Arm, and the Smith Cloud. The Smith Cloud com- ponent was fixed at +96.2 km s -1 , the velocity of the [S II] emission from the Smith Cloud in Fig. 2; the resulting Smith Cloud intensities are mapped in Fig. 1. Velocity calibration: The velocity calibration of the new [N II] and [S II] data is estimated based on the instrument configuration (Madsen 2004). The recalibration of the veloc- ity scale at CTIO is not yet complete, so the absolute velocity scale has an uncertainty of up to ±10 km s -1 . Results Kinematics: With typical widths of 30 km s -1 and blending from the foreground Sagittarius Arm near v LSR = +40 km s -1 , the Hα line is not well suited to measuring the velocity of components along the Smith Cloud sightline. The nar- rower [S II] line is better suited for this task: note the rela- tively limited blending of the Smith Cloud [S II] line in Fig. 2. H I position-velocity diagrams presented by Lock- man et al. (2008) show a systemic velocity gradient along the cloud. Figs. 4 – 6 show [N II], [S II], and Hα spectra at four points along the cloud with the H I centroid for the Smith Cloud. In the ‘mid 2’ and ‘tail’ directions, the [S II] emission is at a velocity similar to the H I, indicating that the ionized gas traces the neutral gas. Abundance: Using the [S II] pointed data reported here and the nearest Hα pointing from the map, we estimate the [S II] / Hα line ratios. The [S II] / Hα line ratio in the tail is 0.30 ± 0.06, similar to the value of 0.32 ± 0.05 ob- served in the tip by Hill et al. (2009), suggesting that the ionization conditions of the sulfur abundance do not change appreciably across the cloud. We do not detect [N II] emission from the cloud in any direction other than the tip, although there are a number of atmospheric lines at velocities near the expected emission from the Smith Cloud. Our preliminary CTIO atmospheric template makes a sig- nificant [N II] detection difficult. References Bland-Hawthorn, J., et al. 1998, MNRAS, 299, 611 Hausen, N. R., et al. 2002, ApJ, 565, 1060 Hill, A. S., Haffner, L. M., & Reynolds, R. J. 2009, ApJ, 703, 1832 Lockman, F. J., et al. 2008, ApJ, 679, L21 Madsen, G. J. 2004, PhD thesis, Univ. Wisconsin-Madison Nichols, M. & Bland-Hawthorn, J. 2009, ApJ, 707, 1642 Wakker, B. P., et al. 2008, ApJ, 672, 298 Acknowledgments WHAM is funded by the National Science Foundation through grant AST 0607512 to the University of Wisconsin-Madison. Figure 4: [N II] spectra along the cloud. These summed, atmosphere-subtracted spectra were obtained at the locations of the green ‘×’ signs in Fig. 1. Dotted lines show the mean velocities of the Smith Cloud H I components in Fig. 3. http://www.astro.wisc.edu/~hill/hill_AAS_2010_poster.pdf AAS Washington, DC, January 2010 Local gas Sagittarius Arm Smith Cloud Figure 1: Map of Hα emission from the Smith Cloud. Contours indicate H I column densities from Lockman et al. (2008). Positions of deep spectra are shown with green ‘×’ signs, and the blue ‘×’ sign shows the OFF direction. The 1° WHAM beams are smoothed to produce this image. Shown in Galactic coordinates. From Hill et al. (2009). Figure 3: H I 21 cm spectra along the cloud. Green Bank Telescope data from Lockman et al. (2008) averaged over a 1° beam corresponding to the WHAM beams in Figs. 4 and 5. [email protected] Figure 5: [S II] spectra along the cloud. Identical to Fig. 4. Figure 6: Hα spectra along the cloud. Like Figs. 3 and 4 but spectra are 60 s exposures from the pointing in the map nearest the labeled position. Figure 2: Smith Cloud spectra obtained with WHAM from KPNO towards the region of brightest emission (the ‘tip’). Dashed lines show individual Gaussian fits to the local gas, Sagit- tarius Arm, and Smith Cloud components, and the solid line shows the sum of the three Gaussians. From Hill et al. (2009).
Transcript of Alex S. Hill & L. Matthew Haffner · 2011. 3. 8. · poster by Haffner et al. At each position, we...
Ionized gas in the Smith CloudAlex S. Hill & L. Matthew Haffner
University of Wisconsin-Madison
AbstractWe present new observations of 104 K ionized gas in the Smith Cloud, a high velocity cloud thought to be interacting with the interstellar medium of the Galaxy 3 kpc below the midplane. We have obtained Hα, [N II], and [S II] spectra of the cloud with the Wisconsin H-Alpha Mapper (WHAM). Previously, we have shown that the ionized component has a mass of > 106 M⊙ (comparable to the mass of the neutral gas in the cloud), with relatively bright Hα emission in di-rections with faint H I 21 cm emission. Detections of [S II] and [N II] constrain the temperature and metallicity of the infalling ionized gas. We report new WHAM [N II] and [S II] spectra in four directions along the axis of the cloud in order to explore the kinematics and metallicity distribution across the cloud.
IntroductionThe infall of gas is a significant driver of galaxy evolution, and is thought to be partially responsible for maintaining the observed star formation rate in the Milky Way of 1 M⊙ yr-1 over the past ~10 Gyr with roughly constant metal-licity. High velocity clouds (HVCs) are likely examples of the active infall of material, often with low metalicities.
The Smith Cloud is perhaps the best studied HVC, with all three components of both the position in the Galaxy and the velocity well known. Stellar absorption distance brackets place the cloud at a heliocentric distance of 9.8 – 15.1 kpc (Wakker et al. 2008), or 2.9 ± 0.6 kpc below the Galactic mid-plane and 7.6 kpc from the Galactic center. Lockman et al.
(2008) argued that the cloud is interacting with the inter-stellar medium of the Galactic disk. The cloud has a come-tary morphology with a bright core and a diffuse, trailing tail., moving towards the plane at 73 ± 26 km s-1 with an H I mass of > 106 M⊙ and a comparable H+ mass. This has al-lowed considerable speculation about the origins of the cloud (Bland-Hawthorn et al. 1998, Lockman et al. 2008, Nichols & Bland-Hawthorn 2009). However, relatively little is known about the physical conditions within the cloud.
Recently, we have presented Wisconsin Hα Mapper (WHAM) observations of ionized gas in the Smith Cloud (Hill et al. 2009). In that work, we presented a map of the Hα emission from the cloud (Fig. 1), finding considerable Hα emission and H+ mass in regions with relatively faint H I. We also obtained [N II] λ6583 and [S II] λ6716 spectra in the direction with the brightest Hα and H I emission (Fig. 2). With these data, we estimated the nitrogen abundance of the ionized gas in the cloud to be 0.15 – 0.44 times Solar assum-ing a temperature in the ionized gas of 12,000 – 8,000 K.
Here, we report new [S II] and [N II] observations at four lo-cations in the cloud with the aim of constraining the kine-matics and abundance distribution of the ionized gas across the cloud.
ObservationsWe used WHAM (see adjacent poster by Haffner et al.) to obtain Hα, [N II] λ6583, and [S II] λ6716 spectra. These ob-servations were obtained in 2009 from the Cerro Tololo In-teramerican Observatory (CTIO).
Pointed spectra: For all pointed spectra, we alternated 120 s ON source observations and 120 s OFF source observations for a total of 480 s ON source and 480 s OFF source. Hausen et al. (2002) constructed an empirical template of the faint lines near Hα, which are stable relative to each other but vary in intensity, partly dependent upon the zenith angle. To identify the appropriate level of the atmospheric template for our observations, we fit the OFF spectra of each line to a single atmospheric template, which we then subtracted from the ON spectra to obtain spectra consisting purely of astro-nomical emission. The atmospheric template used here is the template we used at Kitt Peak; if the atmospheric spectrum at CTIO is different, this will introduce an uncertainty in the observed velocity and intensity of faint lines. For Hα spec-tra, we also subtracted the geocornal Hα line. The multi-wavelength spectra reported for the tip of the cloud by Hill et al. (2009) are shown in Fig. 2. We show H I spectra from Lockman et al. in four directions along the main axis of the cloud in Fig. 3. Pointed [N II] and [S II] spectra obtained at CTIO in the same four directions are shown in Figs. 4 and 5.
Maps: We obtained a spectroscopic Hα map of the Smith Cloud using the “block” method described in the nearby
poster by Haffner et al. At each position, we obtained an Hα spectrum with an exposure time of 60 s. The resulting spec-tra were fit with three Gaussian components, each with a fixed width of 30 km s-1, corresponding to local gas, the Sag-ittarius Arm, and the Smith Cloud. The Smith Cloud com-ponent was fixed at +96.2 km s-1, the velocity of the [S II] emission from the Smith Cloud in Fig. 2; the resulting Smith Cloud intensities are mapped in Fig. 1.
Velocity calibration: The velocity calibration of the new [N II] and [S II] data is estimated based on the instrument configuration (Madsen 2004). The recalibration of the veloc-ity scale at CTIO is not yet complete, so the absolute velocity scale has an uncertainty of up to ±10 km s-1.
ResultsKinematics: With typical widths of 30 km s-1 and blending from the foreground Sagittarius Arm near vLSR = +40 km s-1, the Hα line is not well suited to measuring the velocity of components along the Smith Cloud sightline. The nar-rower [S II] line is better suited for this task: note the rela-tively limited blending of the Smith Cloud [S II] line in Fig. 2. H I position-velocity diagrams presented by Lock-man et al. (2008) show a systemic velocity gradient along the cloud. Figs. 4 – 6 show [N II], [S II], and Hα spectra at four points along the cloud with the H I centroid for the Smith Cloud. In the ‘mid 2’ and ‘tail’ directions, the [S II] emission is at a velocity similar to the H I, indicating that the ionized gas traces the neutral gas.
Abundance: Using the [S II] pointed data reported here and the nearest Hα pointing from the map, we estimate the [S II] / Hα line ratios. The [S II] / Hα line ratio in the tail is 0.30 ± 0.06, similar to the value of 0.32 ± 0.05 ob-served in the tip by Hill et al. (2009), suggesting that the ionization conditions of the sulfur abundance do not change appreciably across the cloud. We do not detect [N II] emission from the cloud in any direction other than the tip, although there are a number of atmospheric lines at velocities near the expected emission from the Smith Cloud. Our preliminary CTIO atmospheric template makes a sig-nificant [N II] detection difficult.
ReferencesBland-Hawthorn, J., et al. 1998, MNRAS, 299, 611Hausen, N. R., et al. 2002, ApJ, 565, 1060Hill, A. S., Haffner, L. M., & Reynolds, R. J. 2009, ApJ, 703, 1832Lockman, F. J., et al. 2008, ApJ, 679, L21Madsen, G. J. 2004, PhD thesis, Univ. Wisconsin-MadisonNichols, M. & Bland-Hawthorn, J. 2009, ApJ, 707, 1642Wakker, B. P., et al. 2008, ApJ, 672, 298
Acknowledgments WHAM is funded by the National Science Foundation through grant AST 0607512 to the University of Wisconsin-Madison.
Figure 4: [N II] spectra along the cloud. These summed, atmosphere-subtracted spectra were obtained at the locations of the green ‘×’ signs in Fig. 1. Dotted lines show the mean velocities of the Smith Cloud H I components in Fig. 3.
http://www.astro.wisc.edu/~hill/hill_AAS_2010_poster.pdf AAS Washington, DC, January 2010
Local gas
Sagittarius ArmSmith Cloud
Figure 1: Map of Hα emission from the Smith Cloud. Contours indicate H I column densities from Lockman et al. (2008). Positions of deep spectra are shown with green ‘×’ signs, and the blue ‘×’ sign shows the OFF direction. The 1° WHAM beams are smoothed to produce this image. Shown in Galactic coordinates. From Hill et al. (2009).
Figure 3: H I 21 cm spectra along the cloud. Green Bank Telescope data from Lockman et al. (2008) averaged over a 1° beam corresponding to the WHAM beams in Figs. 4 and 5.
Figure 5: [S II] spectra along the cloud. Identical to Fig. 4.
Figure 6: Hα spectra along the cloud. Like Figs. 3 and 4 but spectra are 60 s exposures from the pointing in the map nearest the labeled position.
Figure 2: Smith Cloud spectra obtained with WHAM from KPNO towards the region of brightest emission (the ‘tip’). Dashed lines show individual Gaussian fits to the local gas, Sagit-tarius Arm, and Smith Cloud components, and the solid line shows the sum of the three Gaussians. From Hill et al. (2009).
