THE ASTROPHYSICAL JOURNAL, 511 :282È288, 1999 January 201999. The American Astronomical Society. All rights reserved. Printed in U.S.A.(

NICMOS NARROWBAND IMAGES OF OMC-11

A. S. B. SCHULTZ,2 SEAN W. J. COLGAN, E. F. ERICKSON, M. J. KAUFMAN,3 AND D. J. HOLLENBACH

NASA-Ames Research Center, Mo†ett Field, CA 94035 ; schultz=ssa1.arc.nasa.gov

C. R. OÏDELL

Department of Space Physics and Astronomy, MS-108, Rice University, 6100 South Main Street, Houston, TX 77005-1892

AND

E. T. YOUNG AND H. CHEN

Steward Observatory, University of Arizona, Tucson, AZ 85721Received 1998 June 25 ; accepted 1998 August 25

ABSTRACTWe present images of a 90@@] 90@@ Ðeld centered on the Becklin-Neugebauer object (BN) in OMC-1,

taken with the Near-Infrared Camera and Multiobject Spectrometer (NICMOS) aboard the HubbleSpace Telescope. The observed lines are 1È0 S(1), Paa, [Fe II] 1.64 km, and the adjacent continua.H2The region is rich in interesting structures. The most remarkable are streamers of emission thatH2extend from 15A to 50A from IRc2, seen here in unprecedented detail. Unlike the northern ““ Ðngers,ÏÏH2these inner structures do not exhibit signiÐcant [Fe II] emission at their tips, which we suggest is due tolower excitation. These observations also show that the morphological details of the Paa and [Fe II]emission (both imaged for the Ðrst time in this region) bear a striking resemblance to that of the Ha and[S II] emission previously observed with WFPC2. This implies that these IR and optical lines are pro-duced by radiative excitation on the surface of the molecular cloud. The Paa morphology of HH 202 isalso very similar to its Ha and [O III] emission, again suggesting that the extended Paa emission in thisobject is photoexcited by the Trapezium, as has been suggested for the optical emission. We Ðnd evi-dence of shock-excited [Fe II] in HH 208, where it again closely follows the morphology of [S II]. Thereis also coincident with the [S II] and [Fe II] emission, which may be associated with HH 208.H2Finally, we note some interesting continuum features : di†use ““ tails ÏÏ trailing from IRc3 and IRc4, moreextensive observations of the ““ crescent ÏÏ found by Stolovy et al., and new observations of a similar ovalobject nearby. We also Ðnd a V-shaped region that may be the boundary of a cavity being cleared byIRc2.Subject headings : H II regions È infrared : ISM: lines and bands È ISM: individual (OMC-1) È

ISM: jets and outÑows È stars : preÈmain-sequence

1. INTRODUCTION

As the nearest region (d \ 450 pc) undergoing recentmassive star formation, the Orion giant molecular cloud isthe target of intense study at many wavelengths. On theedge of the OMC-1 molecular cloud is a blister H II regionexcited by the Trapezium stars. The visible nebula is thethin layer of photoionized gas on the surface of the molecu-lar cloud. Between the H II region and the molecular cloudlies a photodissociation region (PDR) that is also excited bythe Trapezium. A lid of neutral gas covers the entire nebula(van der Werf & Goss 1989).

Within the molecular cloud behind the H II region liesBN/KL, a location of particular interest both because of thecandidate protostars identiÐed there and because of theassociated energetic outÑows. The energetics of the regionseem to be dominated by a few luminous but deeply em-bedded sources whose bolometric luminosities approach105 (e.g., Genzel & Stutzki 1989). The outÑows appearL

_

1 Based on observations with the NASA/ESA Hubble Space Telescope,obtained at the Space Telescope Science Institute, which is operated by theAssociation of Universities for Research in Astronomy, Inc. under NASAcontract NAS 5-26555.

2 SETI Institute.3 Current address : Department of Physics, San Jose State University,

San Jose, CA 95014.

to originate from the vicinity of the radio continuumsources ““ I ÏÏ and ““ n,ÏÏ located B8A southeast of the Becklin-Neugebauer object (BN) (Menten & Reid 1995). With theexception of the bright near-IR sources IRc2, BN, and IRc9,the extended continuum features in the region appear to bedue to reÑected starlight (Minchin et al. 1991) seen throughinhomogeneities in the surrounding dust cloud.

Here we introduce and brieÑy discuss new [Fe II] (1.64km), Paa (1.87 km), 1È0 S(1) (2.12 km) and continuumH2images of a region centered on BN/KL. These images wereobtained with the Near-Infrared Camera and MultiobjectSpectrometer (NICMOS) on board the Hubble SpaceTelescope. The images have a pixel size of and show Ðne0A.2details in all of the Ðlters.

emission arises either from shocked gas [where itH2traces both dissociative km s~1) and non-(vSZ 50

dissociative km s~1) shocks], or from UV Ñuor-(vS[ 50

escence. [Fe II] emission arises from dissociative shocks, aswell as photoionized gas. Paa emission is typically fromrecombinations in H II regions, but can also result fromhigh-velocity shocks.

2. OBSERVATIONS AND DATA REDUCTION

The data were obtained with NICMOS camera 3 in twoseparate orbits of the Hubble Space Telescope on 1998January 12 and 16, during the camera 3 campaign, in which

282

NICMOS NARROWBAND IMAGES OF OMC-1 283

FIG. 1.ÈSchematic of the near-IR morphology of the OMC-1 region. Colors are as indicated. Dust disk is from Wright et al. (1992).

the telescope focus was changed to optimize camera 3observations. The total area of each frame is 50@@] 50@@,with a plate scale of per pixel. The di†raction limit0A.2(h \ 1.22j/D) is at 2 km. Images were obtained in three0A.2line Ðlters, plus three nearby continuum Ðlters (all Ðlters are1% bandpass) : [Fe II] 1.64 km (F164N), 1.66 km (F166N),Paa 1.87 km (F187N), 1.90 km (F190N), 1È0 S(1) 2.12H2km (F212N), and 2.15 km (F215N).

The data were taken in the NICMOS MULTIACCUMmode ; total integration time per frame was 32 s for F212Nand F215N, 16 s for F187N and F190N, and 88 s for F164Nand F166N. The data were processed with the standard

NICMOS CALNICA pipeline, which includes bias subtrac-tion, dark subtraction, Ñat-Ðelding, cosmic-ray removal,and Ñux calibration. The continuum images were then sub-tracted from the emission-line images.

The and Paa images are each composed of sevenH2frames, covering a total of approximately 95@@] 95@@. The[Fe II] images are composed of four frames, covering thecentral square arcminute of the and Paa images. In allH2cases the position angle was 356¡ east from north. Aftermosaicking, the various Ðlters were registered to the orien-tation and position of the F215N Ðlter using star positions(with an accuracy of B0.2 pixel). The resulting 3 p upper

284 SCHULTZ ET AL. Vol. 511

FIG. 2.ÈLarge Ðeld of OMC-1, centered on BN at (J2000). Red is green is Paa, and blue is 2.15 km continuum.05h35m14s.12, [05¡22@23A.2 H2,

limits in Jy in a 1 square arcsecond area are 1.0] 10~5 forF215N, 1.3] 10~5 for F212N, 2.5] 10~5 for F190N,3.4] 10~5 for F187N, 0.5] 10~5 for F166N, and0.6] 10~5 for F164N.

3. RESULTS

Overviews of the region are shown in Figures 1È3. Figure1 is a schematic highlighting the important features in theregion. Figure 2 shows our large Ðeld, which includes H2,Paa, and 2.15 km continuum. The smaller Ðeld, comprisedof Paa, and [Fe II] images, is shown in Figure 3.H2,The interesting structures in our images can be dividedroughly into three groups : shocked emission features (the

streamers, HH 208), radiative emission features (theH2““ wisp,ÏÏ HH 202, proplyds, and the proplyd cloud), and

continuum features (IRc3 and IRc4, the crescent, and the““ V ÏÏ).

3.1. Shock Excitation FeaturesStreamers3.1.1. H2

The most striking feature of the data is the remarkablearray of streamers emanating from the general vicinityH2of BN (Fig. 2). Allen & Burton (1993, hereafter AB) estab-lished the presence of similar structures, which they termed““ Ðngers,ÏÏ up to 2@ (0.25 pc) north of our Ðeld. Schild, Miller,& Tennyson (1997) presented high spatial resolution imagesof the 2 km continuum and several lines of showingH2,many of the same features seen in our images. AB inter-preted the Ðngers as bow shocks formed in a stellar outÑow,when ““ bullets ÏÏ of ejected matter strike the ambient

No. 1, 1999 NICMOS NARROWBAND IMAGES OF OMC-1 285

FIG. 3.ÈSmall Ðeld of OMC-1, centered on BN, showing (red), PaaH2(green), and [Fe II] (blue) emission. See Fig. 1 for comparison of large andsmall Ðelds of view.

medium. Based on the cooling time, AB concluded thatH2the Ðngers were created in a single event, rather than by aprecessing jet or succession of jets.

Stone, Xu, & Mundy (1995, hereafter SXM) suggest adi†erent creation mechanism, in which the bullets do notoriginate at the source of the outÑow, but instead arecreated and accelerated in situ in the outÑow, producing thebow shocks. The bullets are the result of Rayleigh-Taylorinstabilities formed when a poorly collimated wind is sud-denly accelerated into the ambient medium, either by asecond, faster wind behind it, or by a decrease in the densityof the ambient material. One prediction of SXMÏs model isthat a region of clumpy emission will form behind the bowshocks.

McCaughrean & Mac Low (1997) have presented imagesof the emission in the region and argued that the clumpyH2emission they Ðnd in the inner regions (i.e., surroundingBN) is that predicted by SXMÏs model. Our images revealthat much of the emission in that region is also made up ofclosely spaced bow shocksÈsome of them superposedÈwhich would lead to a clumpy appearance at lower spatialresolution. However, the region to the north and northwestof BN is a large area of Ñocculent emission, extending toroughly 45A from BN, which does not exhibit deÐnite bowshock structures. The emission is brightest in this region,H2which contains IRc9 and OMC Pk 1 (Beckwith et al. 1978).This may be the clumpy emission predicted by SXM.

Our images (see Fig. 2) reveal many of the streamers to bemore complex than simple bow shocks ; they exhibit deli-cate internal structures that may be only the superposition

of separate streamers. These are the ““ nested arcs ÏÏ seen inthe inner streamers by Stolovy et al. (1998) ; we see thesestructures out to 50A. The six Ðngers found by AB wereH2all capped with [Fe II] emission. This is understandablesince the tip of the bow shock represents the region of great-est shock velocity and therefore greatest excitation. Our[Fe II] images cover only the central square arcminute ofthe area covered by our 2 km observations, but within thatregion we Ðnd that only two of the 15 streamers haveH2[Fe II] tips.

This is unlikely to be an extinction e†ect, even though thestreamers are embedded in the molecular cloud ; rather, theabsence of [Fe II] in the inner streamers is probably due tolower excitation. Unlike the northern Ðngers found by AB,the streamers in our image show the emission out to theH2tips of the bow shocks. This is consistent with otheranalyses of the shocked emission, which imply B40 km s~1nondissociative shocks (Harwit et al. 1998).

Shock models have been presented that predict the rela-tive strengths of and [Fe II] emission from dissociativeH2shocks (e.g., Hollenbach & McKee 1989). For the twostreamers in which we observe both [Fe II] and emis-H2sion, we Ðnd the observed ratios are consistent only withslow dissociative shocks km s~1). If the bow shocks(v

S[ 35

were produced in a single outÑow incident, as postulated byAB, the distant northern Ðngers may have traveled fartherbecause of their larger velocity. It is also possible that thenorthern Ðngers originated in a di†erent outÑow episode, oreven from a di†erent source, from the inner structures weobserve.

3.1.2. HH 208

Most of the shocked [Fe II] emission seen in our imagesis in HH 208. We Ðnd that the [Fe II] emission is almostidentical in morphology to the [S II] jj6716/6730 emissionseen by OÏDell et al. (1997b), which implies that the [Fe II]and [S II] emission come from the same region : the photo-dissociation region (PDR) near the ionization front. The[Fe II] emission cannot come from the H II region becausethere Fe` would be converted to Fe`` ; the [S II] emissioncannot come from very deep in the PDR because of extinc-tion. Hence emission from both species probably originatesin the PDR close to the far side (relative to the observer) ofthe ionization front. The presence of coincident withH2[Fe II] and [S II] may also argue for the placement ofHH 208 within the PDR, where radiative heating allows[Fe II] and [S II] to be emitted from shocks at lower veloci-ties than would be possible in colder gas ; these shocks canexcite without dissociating it.H2We see at least one other example of an knot with aH2cap of [Fe II] (which is also seen in [S II] by OÏDell et al.1997b) to the southeast of BN. We believe this to be aHerbig-Haro object ; the [Fe II] cap is the higher excitationbow shock. For the same reasons applied to HH 208 above,this is also likely to be situated in the near side of the PDR.

3.2. Radiative Excitation FeaturesThe general appearance of the Paa and [Fe II] emission is

strikingly similar to that of Ha and [S II] (OÏDell et al.1997b). The maxima and minima of the Ha emission arereproduced in Paa, with only a few important di†erences.The Ha emission drops o† sharply north of BN, whereas thePaa decreases only toward the northwest of BN. There is aparticularly strong region of Paa emission B40A southeast

N

W

10"

286 SCHULTZ ET AL. Vol. 511

FIG. 4.È2.15 km continuum image of OMC-1, showing the same Ðeld as Fig. 2

of BN that is surrounded by six proplyds (see ° 3.2.3). Thedi†erences between Ha and Paa are likely to be due todi†erential extinction.

3.2.1. T he ““W isp ÏÏ

One interesting structure found in Paa and [Fe II] emis-sion is the bright ““ wisp ÏÏ immediately to the northeast ofBN (Figs. 1 and 3). This object also appears in Ha, [S II],and [N II] j6583 (OÏDell et al. 1997b). It was Ðrst observedwith WFPC1 by Hester et al. (1991), who suggested it was ajet. The higher resolution WFPC2 images of OÏDell et al.(1997b), and the lack of high-velocity gas (OÏDell, Wen,& Hester 1991) show that this explanation is unlikely, butit is still unclear whether the gas is shock excited or photo-excited.

The appearance of the wisp in hydrogen emission is dif-ferent from that in the forbidden lines ; the hydrogen-

emitting region is shorter and thicker than the forbiddenline emission region. Furthermore, the [S II] and [Fe II]images exhibit a second linear strand of emission, runningparallel to the main body and joining it in a faint emissionknot at the western end. The [N II] image does not showthis. Finally, the forbidden line emission region is o†setslightly to the northeast from the hydrogen emission. Thismorphology is consistent with photoexcitation from thesouth, since the more easily excited species (hydrogen, inthis case) would be produced closer to the Trapezium.

3.2.2. HH 202

Another feature seen in both optical and infrared wave-lengths is HH 202, a large tubular structure in the south-western corner of the Ðeld (Figs. 1 and 2). This object wasrecently studied by OÏDell et al. (1997b), who note thatalthough the lower excitation lines of [S II] and [N II] are

N

W

5"

No. 1, 1999 NICMOS NARROWBAND IMAGES OF OMC-1 287

FIG. 5.È1.66 km continuum in OMC-1, showing detail in the region around BN

concentrated in the head of the long tube, [O III] j5007 andHa emission is seen along the entire tube. OÏDell et al.(1997a) Ðnd that the [O III] emission is blueshifted. Ourobservations show that Paa, as well as very faint continuumemission at 1.90 and 2.15 km, also extends the length of thetube.

OÏDell et al. (1997a, 1997b) suggest that the [O III] emis-sion along the tube does not come entirely from shocks, butis partly photoexcited. They hypothesize a jet originatingfrom a source within the H II region and striking the fore-ground neutral lid, creating a tube that is photoionizedfrom within by the Trapezium. Photoionization changes S`to S``, so that the gas is unable to cool by [S II] emission(which its shock velocity would otherwise dictate), but mustcool by emission of higher excitation lines such as [O III].We suggest this radiative heating also produces the Paaemission. The jet terminates inside the neutral lid, produc-ing the north and south knots that are bright in [S II](OÏDell et al. 1997a, 1997b). We also Ðnd Paa emission inthe southern knot, with the same general position and mor-phology as the Ha emission seen there by OÏDell et al.(1997b). The higher excitation lines of [O III] and hydrogenare not seen in the northern knot.

3.2.3. Proplyds

The following proplyds all exhibit compact Paa emission,suggesting that they are inside the ionized region : 113-153,114-155, 115-155, 116-156, 131-247, 135-220, 138-207, 142-301, 154-240, 154-225, 163-249, 163-222, 165-235, 168-235,170-249, and 173-236 (nomenclature of OÏDell & Wen1994). 142-301 is a striking object resembling an ice creamcone to the east of HH 202. We also Ðnd one compactobject at the head of HH 202 that exhibits strong Paa emis-sion but has only a faint optical counterpart (source P1 inFig. 1).

3.3. Continuum Features3.3.1. IRc3 and IRc4

Both of these objects possess long, curving, di†use ““ tails ÏÏof continuum emission extending to the southwest of themain objects (Fig. 4). The brightest parts of the tails extendup to 13A (0.03 pc) from the main objects, but there is muchfainter emission up to 28A (0.06 pc) away. The more distantfeatures seem to emerge from behind curving strands ofobscuration that lie to the southwest of BN (as well as to thesouth ; these are best seen in Fig. 2). Each of the bright tailsis coincident with a streamer of emission, which suggestsH2

288 SCHULTZ ET AL.

the tails may be tunnels punched through the cloud byejected matter that are now being illuminated from withinby the exciting source.

This may also be the origin of the extended continuumsource SEBN (Aitken et al. 1985), approximately 25A eastand 10A south of BN. Our images show that this objectexhibits a Ñocculent structure not as clearly delineated inprevious images (Fig. 4). The region has a large polarization(B30% at K) whose orientation clearly points to illumi-nation from the vicinity of IRc2 (Minchin et al. 1991).

3.3.2. T he Crescent

In agreement with Stolovy et al. (1998), we detect the““ crescent ÏÏ feature in all three continuum images, with someassociated emission. No [Fe II] or Paa emission is seenH2in this feature. The sensitive F166N image, however (Fig. 5),shows that the ““ tail ÏÏ of the crescent extends up to 6A to thenorthwest, further than seen by Stolovy et al. (1998).2A.5Another, similar oval feature can be seen very faintly to thesoutheast of the crescent. Its blunt end can be seen almosttouching the northeast extensions of BN.

3.3.3. T he V

Approximately northeast of BN and apparent in all7A.5three of the continuum images is the base (source ““ q ÏÏ ofLonsdale et al. 1982) of a V-shaped structure that opens tothe north (Fig. 5). The extended ridge of dust emissionmapped at 450 km and 3.5 mm by Wright et al. (1992) liesalong the line joining BN and the apex of the V. The V isalso adjacent to the northeast extension of BN, which has asimilarly sharp edge.

The V appears faintly in emission, but not in [Fe II]H2or Paa. The eastern branch contains a compact knotH2near its base and seems to terminate in a long feature.H2This linear structure appears to pass through a ring orbubble of emission and continue on the far side. SchildH2et al. (1997) suggested that this is a jet that struck a clump ofdenser material, producing an expanding density wave.

The coincident dust and emission suggest that the V isH2a photodissociation region. However, there is no knownfar-UV emission from the center of the nebula and noextended radio sources that would indicate a photoionizedregion obscured from our point of view but with an unob-scured path to the V. This renders a PDR interpretationunlikely. We propose a scenario in which the emission inH2

the V is shock excited. In this picture the supersonic outÑowfrom the (unknown) central source clears a channel throughthe molecular cloud. Note that the angular extent of thegeneral northern outÑow region is bordered by theH2eastern leg of the V and by the ““ jet.ÏÏ Thus the jet mayH2delineate the conÐnement of the outÑow by the molecularcloud. The continuum emission then results from the reÑec-tion of IR photons produced elsewhere in the cloud. Thepolarimetry of Minchin et al. (1991) seems to conÐrm thereÑection hypothesis and indicates that the V is likely to beilluminated by IRc2. In this picture the V shows the extentto which the outÑow has been able to clear a path at itsouter edges. The shocked emission it exhibits is from theH2impact of the outÑow on the walls of the cavity.

4. CONCLUSIONS

We have seen that the near-IR emission in OMC-1 iscomplex, arising from structures at all levels of the nebula,from the H II region near the neutral lid, to deep within theunderlying molecular cloud. The most striking features arethe streamers, many of which we see here in unprece-H2dented detail. We Ðnd that most of the inner streamers donot exhibit detectable [Fe II] emission, as the more distantnorthern ““ Ðngers ÏÏ do (Allen & Burton 1993). We suggestthat this is due to excitation, rather than extinction.

Our continuum images reveal numerous extended emis-sion features that seem to be reÑection from the continuumsources within the cloud, presumably including the outÑowsource. To the north of BN we see a V-shaped structure thatlies at the eastern limit of the streamers and may deÐneH2the boundary of a cavity cleared by the outÑow.

On the near side of the molecular cloud, within the PDR,we Ðnd shocked structures that exhibit both atomic ([Fe II])and molecular emission, such as HH 208 and the(H2)[Fe II]Ècapped knot to the southeast of BN. TheH2[Fe II] morphology of these objects is very similar to their[S II] morphology as seen by OÏDell et al. (1997b), implyingthat the objects are not buried deeply in the PDR.

The visible surface of the Orion Nebula, which is excitedradiatively by the Trapezium stars, is seen in Paa and[Fe II], with a morphology strikingly similar to the Ha,[S II], [O I], and [N II] emission found there by, e.g., OÏDell,et al. (1997b). Embedded in this surface are a number ofcompact objects, proplyds, which exhibit Paa emission.

REFERENCESAitken, D. K., Bailey, J. A., Roche, P. F., & Hough, J. H. 1985, MNRAS,

215, 815Allen, D. A., & Burton, M. G. 1993, Nature, 363, 54 (AB)Beckwith, S., Persson, S. E., Neugebauer, G., & Becklin, E. E. 1978, ApJ,

223, 464Genzel, R., & Stutzki, J. 1989, ARA&A, 27, 41Harwit, M., Neufeld, D. A., Melnick, G. J., & Kaufman, M. J. 1998, ApJ,

497, L105Hester, J. J., et al. 1991, ApJ, 369, L75Hollenbach, D. J., & McKee, C. 1989, ApJ, 342, 306Lonsdale, C. J., Becklin, E. E., Lee, T. J., & Stewart, J. M. 1982, AJ, 87, 1819McCaughrean, M. J., & Mac Low, M.-M. 1997, AJ, 113, 391Menten, K. M., & Reid, M. J. 1995, ApJ, 445, L157

Minchin, N. R., et al. 1991, MNRAS, 248, 715OÏDell, C. R., Hartigan, P., Bally, J., & Morse, J. A. 1997a, AJ, 114, 2106OÏDell, C. R., Hartigan, P., Lane, W. M., Wong, S. K., Burton, M. G.,

Raymond, J., & Axon, D. J. 1997b, AJ, 114, 730OÏDell, C. R., & Wen, Z. 1994, ApJ, 436, 194OÏDell, C. R., Wen, Z., & Hester, J. J. 1991, PASP, 103, 82Schild, H., Miller, S., & Tennyson, J. 1997, A&A, 318, 608Stolovy, S. R., et al. 1998, ApJ, 492, L151Stone, J. M., Xu, J., & Mundy, L. G. 1995, Nature, 377, 315 (SXM)van der Werf, P. P., & Goss, W. M. 1989, A&A, 224, 209Wright, M., Sandell, G., Wilner, D. L., & Plambeck, R. L. 1992, ApJ, 393,

225