jr-ca-newsletter

The presence of diffuse ionized gas (DIG), also frequently known as the warm ionized medium (WIM), is found to be ubiquitous in the Milky Way (e.g.,Reynolds et al. 1999). First detections of extraplanar DIG (eDIG) in halos of external galaxies have been made over a decade ago (Dettmar 1990; Rand et al. 1990). Since then, several detections have been reported (e.g.,Rand 1996; Rossa & Dettmar 2000), and the question was raised whether eDIG is a general scenario among all types of galaxies (starburst, non-starburst, dwarfs), or whether it is only an exception for actively star forming galaxies. The detections of eDIG in galactic halos is generally believed to be correlated with the star formation activity in the underlying galaxy disk, where the gas is being assumed to be photoionized by OB-stars (e.g.,Dove & Shull 1994). The gas and energy transport might be supplied by correlated SNe explosions according to the theoretical scenarios such as galactic fountains (Shapiro & Field 1976); chimneys (Norman & Ikeuchi 1989); superbubble outbreak (Mac Low & McCray 1988) or in the case of starburst galaxies by superwinds (Heckman et al. 1990). An observational approach to validate the general scenario, that the strength of the star formation activity in the underlying galaxy disk is correlated with the presence of eDIG in the halos, has been achieved with a H $\alpha$ imaging survey of nearby edge-on spiral galaxies, performed for the northern hemisphere objects with the CAHA 2.2m telescope at Calar Alto, and the southern hemisphere objects were studied with the Danish 1.54m telescope at ESO (La Silla/Chile).

2. Target selection, observations, and data reduction

The edge-on galaxies were selected according to the following criteria. We chose late-type spiral galaxies (Sb-Sc) with an inclination of $i \geq 80^\circ$ , and optical diameters of $3' \leq D_{25} \leq 12'$ . The galaxies have radial velocities $v_{\rm rad} \leq 6000\,{\rm km\,s^{-1}}$ , with the majority having $v_{\rm rad} \leq 4000\,{\rm km\,s^{-1}}$ . In order to cover a broad SF activity, we selected our survey galaxies yielding a wide range of FIR flux values. Moreover, the availability of the desired H $\alpha$ filters according to redshifts of the galaxies was a further constraint. The galaxies were selected from the RC3 catalog (de Vaucouleurs et al. 1991), and from the Flat Galaxy Catalog (FGC) by Karachentsev et al. (1993).

The observations for the northern hemisphere objects have been carried out with CAFOS in imaging mode at the CAHA 2.2m telescope. The used CCD chip was a 2048 $\times$ 2048 pixel SITe#1d, with a pixel size of 24 $\mu$ m. The pixel scale was 0.53 $''\,\rm {pix}^{-1}$ , and the resulting field of view was 18.1 $'\times18.1'$ . The used H $\alpha$ filters were the CA #658/10 and the CA #665/17 filters with a $\Delta\lambda$ = 98Å and 168Å, respectively. The used R-band filter (Johnson R) was the CA #641/158 with $\Delta\lambda$ = 1575Å. The data reduction was performed in the usual manner using various IRAF packages. In order to obtain a continuum free H $\alpha$ +[NII] image, the R-band image was scaled and subtracted from the H $\alpha$ image.

3. Results

Finally, about 120 galaxies did fulfill all of our selection critera. We did observe 65 of these galaxies, given the object visibility at our various scheduled observing runs. Adding 9 observed galaxies from a recent investigation, acting as a first sub-sample (Rossa & Dettmar 2000), our survey consists of 74 galaxies in total. Although this is the largest H $\alpha$ survey to date investigating edge-on galaxies in this context, it should be mentioned that it is not a statistically complete sample. In Fig.1 we show two galaxies, observed with CAFOS, acting as representative examples of our survey galaxies. Further details appear elsewhere (Rossa 2001; Rossa & Dettmar 2001a,b).

**Figure 1:** Continuum subtracted H $\alpha$ images. Left: **NGC5290**, right: **NGC5965**; orientation: N is up, and E is to the left
$\begin{figure} \begin{center} \hspace{-2.0cm} \psfig {figure=pic-gal-2-new.ps,width=18cm,clip=t}\end{center}. \end{figure}$

In order to link the observed eDIG morphology with SF activity, we have used FIR properties as diagnostic means. We make a detailed comparison of the FIR properties of our survey galaxies with the starburst galaxy sample studied by Lehnert & Heckman (1995). Important FIR parameters which characterize the strength of star formation within galaxies, such as the SFR per unit area ( $L_{\rm FIR}/D^2_{25}$ ), and the FIR color ( $S_{60}/S_{100}$ ), are shown for our studied galaxies in the following two histograms.

We have plotted the histogram of the $S_{60}/S_{100}$ ratio (Fig.2) for the 62 galaxies which have IRAS detections at 60 $\mu$ m and 100 $\mu$ m. In comparison with the starburst sample, studied by Lehnert & Heckman (1995), our survey covers the fainter end of the $S_{60}/S_{100}$ distribution with a peak between 0.2 and 0.35. Their sample covers the IRAS warm galaxies with $S_{60}/S_{100} \geq 0.4$ with a local peak at 0.55 and highest values reaching almost 1.4. The combined histogram of these both surveys is shown in Fig.2. Here the vertical dotted line at 0.4 marks the border for IRAS warm galaxies. The covered FIR luminosities of our survey galaxies range from a few 10 $^8\,\rm {L_{\odot}}$ to $10^{11}\,\rm {L_{\odot}}$ (see Fig.3). With the covered distribution of the $S_{60}/S_{100}$ ratio (lower values) it is warranted that the fainter end is well traced, in order to derive a minimal energy threshold.

**Figure 2:** Histogram showing the $S_{60}/S_{100}$ distribution of the combined samples (starburst sample by Lehnert & Heckman (1995) and our non-starburst sample). The vertical dashed line represents the division between normal and IRAS warm galaxies.
$\rotatebox {270}{\resizebox{8.7cm}{!}{\includegraphics{hist-ratio-comb.ps}}}$

**Figure 3:** Histogram showing the distribution of the FIR luminosity (logarithmically), expressed in units of solar luminosities ( ${\rm L_{\odot}}$ ) of the 62 galaxies of our survey with IRAS detections.
$\rotatebox {270}{\resizebox{8.7cm}{!}{\includegraphics{hist_firlum.ps}}}$

The final results of our H $\alpha$ survey are shown in Fig.4, where we have plotted the SFR per unit area ( $L_{\rm {FIR}}/D^2_{25}$ ) logarithmically as a function of the FIR flux ratio $S_{60}/S_{100}$ (a measure of the dust temperature). There exists a good correlation between the two properties for starburst galaxies as well as for normal (quiescent) galaxies.

**Figure 4:** Diagnostic DIG diagram for all survey galaxies with IRAS detections (62), together with the starburst galaxies, studied by Lehnert & Heckman (1995). Here the ratio of the 60 $\mu$ m and 100 $\mu$ m fluxes ( $S_{60}/S_{100}$ ), is plotted logarithmically versus the SFR per unit area ( $L_{\rm{FIR}}/ D^2_{25}$ ), logarithmically expressed in units of $10^{40}\rm{\,erg\,s^{-1}\, kpc^{-2}}$ . The horizontal dashed line marks the threshold for IRAS warm galaxies at $S_{60}/S_{100} \geq 0.4$ . The starburst galaxies (all showing eDIG) are denoted by filled lozenges, whereas all galaxies of our survey with eDIG detections are denoted by filled squares. The open squares represent galaxies with no eDIG detections.
$\rotatebox {270}{\resizebox{8.7cm}{!}{\includegraphics{newest-ddd_log.ps}}}$

It was found that the presence of eDIG is depending on a minimal energy threshold, expressed in units of the SFR per unit area. Edge-on galaxies ( $i \geq 80^\circ$ ), which exceed a value of $\dot{E}_A^{\rm {thres}}=(3.2\pm0.5)\times10^{40}\,\rm {erg\,s^{-1}\,kpc^{-2}}$ are found to possess extraplanar emission in their halos. While galaxies with a smaller ratio of $L_{\rm {FIR}}/D^2_{25}$ are generally found to be absent of eDIG, however, galaxies with a significantly enhanced $S_{60}/S_{100}$ ratio of $\geq 0.35$ also possess eDIG. All studied galaxies with $S_{60}/S_{100}$ $\geq 0.40$ show extended emission. Those galaxies with high values of both $L_{\rm {FIR}}/D^2_{25}$ , and $S_{60}/S_{!00}$ show most often a layer of eDIG, while galaxies with significantly lower values of either one of the two diagnostic ratios reveal only individual (localized) filaments or plumes.

In our extended H $\alpha$ survey we have investigated 74 edge-on galaxies, observed at Calar Alto and at ESO (La Silla/Chile). For 30 galaxies (40.5%) we found positive eDIG detections. 44 galaxies (59.5%) show no detectable eDIG. While a good correlation between $L_{\rm {FIR}}/D^2_{25}$ and $S_{60}/S_{100}$ is found, the SN energy injection rate per SF area does not correlate as tightly with $S_{60}/S_{100}$ . This can be interpreted in the following way. Galaxies which possess an intrinsically large SF radius (as determined from our H $\alpha$ images), show a significant decrease of $L_{\rm {FIR}}/\pi\,r^2_{\rm {SF}}$ , relative to the ones with smaller values of $r_{\rm {SF}}$ , as the energy injection rate per SF area is $\propto r_{\rm SF}^{-2}$ . However, if local SF regions contribute significantly to the H $\alpha$ flux, there is a good chance that they obey the energy requirement, and thus may also show eDIG, at least locally, Therefore a mix of galaxies with eDIG detections is expected, according to their energy injection rate per SF area.

The detected extraplanar distances of our survey galaxies reach typically values of $\vert z\vert\sim1-2$ kpc, while individual filaments and diffuse clouds reach distances of up to $\vert z\vert\geq6$ kpc in a few cases. For a sub-sample of our survey galaxies we have derived electron densities of typically $n_{\rm {e}}=0.003\,\rm {cm^{-3}}$ , and DIG masses of $M_{\rm {DIG}}\sim10^7\, \rm {M}_{\odot}$ .

The dust component (extraplanar dust) has also been investigated. As a by-product of our continuum subtraction process the R-band images have been used to investigate the high- $\vert z\vert$ dust in the galaxy halos. For this purpose we have applied an unsharp-mask technique to enhance the dust structures against the galaxy background. In this investigation we found that 48 galaxies (65%) did not show extended dust emission at high $\vert z\vert$ . In 26 galaxies (35%) we found high- $\vert z\vert$ dust filaments. A correlation was found in 66 cases (89%), where either eDIG and eDust is present or neither eDIG and eDust was detected. In only 8 galaxies (11%) we found an anti-correlation.

4. Conclusions

The presence of galactic gaseous halos is often regarded as the result of strong SF activity in the underlying galaxy disks. Although detailed kinematical information has mostly eluded us to confirm an associated outflow, many pieces of evidence that have been gathered in the last couple of years, however, strengthen the concept of a disk-halo interaction. Several theoretical scenarios have been suggested for the gas transport from the galaxy disk into the halo, among them the galactic fountains (Shapiro & Field 1976; de Avillez 2000), chimneys (Norman & Ikeuchi 1989), superbubble outbreak (Mac Low & McCray 1988), or superwinds (Heckman et al. 1990).

Past observations have indicated that several edge-on galaxies show DIG at large extraplanar distances, whereas on the other side examples were found where no eDIG has been detected (e.g.,Rand 1996). The study of the DIG in external galaxies was still in its infancy about ten years ago. This new H $\alpha$ survey has contributed significantly to the question on the occurence of gaseous halos in normal spiral galaxies. This first systematical approach has investigated galaxies with a broad distribution of FIR luminosities. In the Diagnostic DiG Diagram (DDD) the correlation between $S_{60}/S_{100}$ and $L_{\rm {FIR}}/D^2_{25}$ is valid for the normal galaxies as well as for the starburst galaxies. The normal galaxies represent the fainter extension of the starburst regime with an obvious soft transition. While the SFR per unit area is a good tracer for eDIG in this diagram, the diagnostic diagram with $\dot{E}_A$ (SNe energy injection per SF area) shows a less obvious correlation. Individual galaxies which have not been investigated yet in the Disk-Halo Interaction (DHI) context, can be selected according to the DDD, to be used in a comparison study for multifrequency studies.

We have presented the individual results for the H $\alpha$ survey galaxies. From the 74 investigated edge-on spiral galaxies we have detected eDIG in 30 galaxies, that is 40.5%. We can therefore conclude, that the presence of DIG in halos of galaxies is not a unique case for only a few galaxies, rather it is found to be ubiquitous in galaxies, which exceed a certain level of SFR per unit area ( $(3.2\pm0.5)\times10^{40}\,\rm {erg\,s^{-1}\, kpc^{-2}}$ ), or at a fainter threshold in combination with enhanced dust temperatures. It can thus also be concluded that eDIG is not a common feature among all spiral galaxies, as many of them do not show eDIG (at the level of the observed sensitivities). The presence of eDIG is depending on the SF activity on both local and global scales. As it was shown in the case of ESO274-1, even a small value of $L_{\rm {FIR}}/D_{25}^2$ in combination with a relatively high value of S $_{60}$ /S $_{100}$ can lead to strong local outflows (Rossa 2001).

The morphology of eDIG shows a wide variety ranging from individual plumes, filaments to pervasive layers. Several of our eDIG detected galaxies bear a more or less intense layer of extended emission with typical extraplanar distances of 1.5-2kpc. Individual filaments of some galaxies (e.g.,NGC4388, NGC5775) reach distances of up to $\sim6$ kpc. In the case of NGC4700 a good correlation between extended H $\alpha$ emission and radio continuum (radio halo) is found, which further strengthens the disk-halo interaction scenario.

Acknowledgements
It is our sincere pleasure to thank Dr. Francisco Prada for carrying out some of the observations of the northern hemisphere objects at Calar Alto in an emergency case. The authors would like to thank Deutsches Zentrum für Luft- und Raumfahrt (DLR) for financial support of this research project through grant 50OR9707. Additional funding for the observing trips to Calar Alto is acknowledged from the DFG. This research has made extensive use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

5. References

de Avillez M.A., 2000, MNRAS 315, 479
Dettmar R.-J., 1990, A&A 232, L15
de Vaucouleurs G., de Vaucouleurs A., Corwin H.G.,Jr., Buta R.J., Paturel G., Fouque P., 1991, Third Reference Catalogue of Bright Galaxies (RC3), Springer Verlag
Dove J.B., Shull J.M., 1994, ApJ 430, 222
Heckman T.M., Armus L., Miley G.K., 1990, ApJS 74, 833
Karachentsev I.D., Karachentseva V.E., Parnovskij S.L., 1993, AN 314, 97
Lehnert M.D., Heckman T. M., 1995, ApJS 97, 89
Mac Low M.-M., McCray R., 1988, ApJ 324, 776
Norman C.A., Ikeuchi S., 1989, ApJ 345, 372
Rand R.J., 1996, ApJ 462, 712
Rand R.J., Kulkarni S.R., Hester J.J., 1990, ApJ 352, L1
Reynolds R.J., Haffner L.M., Tufte S.L., 1999, in: ``New Perspectives on the Interstellar Medium'', ASP Conf. Series Vol. 168, eds. A.R. Taylor, T.L. Landecker, and G. Joncas, p. 149
Rossa J., 2001, PhD thesis, Ruhr-University Bochum, Germany
Rossa J., Dettmar R.-J., 2000, A&A 359, 433
Rossa J., Dettmar R.-J., 2001a, in prep.
Rossa J., Dettmar R.-J., 2001b, in prep.
Shapiro P.R., Field G.B., 1976, ApJ 205, 762