Deciphering the 3-D Orion Nebula-III: Structure on the NE boundary of the Orion-S Embedded Molecular Cloud
DDraft version January 11, 2021
Typeset using L A TEX preprint2 style in AASTeX63
Deciphering the 3-D Orion Nebula-III: Structure on the NE boundary of the Orion-SImbedded Molecular Cloud
C. R. O’Dell, N. P. Abel, and G. J. Ferland Department of Physics and Astronomy, Vanderbilt University, Nashville, TN 37235-1807 MCGP Department, University of Cincinnati, Clermont College, Batavia, OH, 45103 Department of Physics and Astronomy, University of Kentucky, Lexington, KY 40506
ABSTRACTWe have extended the work of Papers I and II of this series to determine at higherspatial resolution the properties of the embedded Orion-S Molecular Cloud that lieswithin the ionized cavity of the Orion Nebula and of the thin ionized layer that liesbetween the Cloud and the observer. This was done using existing and new [N II ] (658.3nm) and [O III ] (500.7 nm) spectra that map the central region of the Orion Nebula(the Huygens Region). However, it remains unclear how the surface brightness of theionized layer on the Orion-S Molecular Cloud and that of a foreground Nearer IonizedLayer are linked, as the observations show they must be. It is shown that the Cloudmodifies the outer parts of the Huygens Region in the direction of the extended hotX-ray gas.
Keywords:
ISM:bubbles-ISM:HII regions-ISM: individual (Orion Nebula, NGC 1976)-ISM:lines and bands-ISM:photon-dominated region(PDR)-ISM:structure INTRODUCTIONThis is the third of a series of papers [(O’Dellet al. 2020a) (henceforth Paper-I) and (O’Dellet al. 2020b) (henceforth Paper-II)] address-ing structures within the Huygens Region ofthe Orion Nebula revealed by high-velocity-resolution spectroscopy. Using spatially largesamples of spectra we determined in Paper-IIthat the region containing the Orion-S Cloudwas characterized by widely changing strengthsof velocity components in [O
III ] (500.7nm),a phenomenon most striking near a sub-regioncall the
Crossing . In this paper we investigatethis sub-region using higher spatial resolutiongroupings of spectra.The
Crossing and the Orion-S Cloud are es-pecially important features as they lie in the di-rection of the X-ray bright (G¨udel et al. 2008) portions of the nebula that are enclosed by therecently discovered [C II ] Outer Shell (Pabst etal. 2019, 2020). The Orion-S Cloud must inter-rupt the flow of stellar wind in that directionand also cast a radiation shadow. In Paper-IV(O’Dell et al. 2020c) we will explore the Herbig-Haro flow designated as HH 269 that arises fromwithin the Orion-S Cloud.This paper differs from Papers-I and II in deal-ing with sequences of higher spatial resolutionslit spectra of about 3 . (cid:48)(cid:48) . (cid:48)(cid:48) (cid:48)(cid:48) to 13 (cid:48)(cid:48) length rather than averages of spectraover 10 × (cid:48)(cid:48) samples. Although the slit spec-tra have lower total signal-to-noise ratios, theyare not blurred in spatial or spectral resolutionby fine-scale structure in the nebula. We shallsee that the sequences of spectra reveal patternsonly hinted at in the large sample studies. a r X i v : . [ a s t r o - ph . GA ] J a n Background of this study
A line-of-sight ray out of the Orion MolecularCloud (OMC) towards the observer first passesthrough a Photon Dominated Region (PDR)(Stacy et al. 1993; Goicoechea et al. 2015) andthen the overlying Main Ionization Layer (MIL)predominately photo-ionized by θ Ori C. TheMIL is stratified into two zones of ionization,ordered by increasing distance from the H II actual ionization front and the distance to thephotoionizing star θ Ori C. The zone closestto the ionization front is composed of He o +H + and emits the collisionally excited [N II ] (658.3nm) line used in this study. The further zoneis composed of He + +H + and emits the colli-sionally excited [O III ] (500.7 nm) line used inthis study. Material from the PDR is contin-uously lost as the MIL gas expands as photo-evaporation flow into the lower density regionsfurther out. The expanding gas is acceleratedaway from the ionization front, with the resultthat the [O
III ] photoevaporation flow velocityshould be greater than that of [N II ]. This pho-toevaporation flow has been well modeled byHenney et al. (2005) and are not affected bythe global motions seen in layers of gas beyond θ Ori C.The region immediately around θ Ori C is oflower density due to a stellar-wind blown bub-ble whose outer boundary is shown in projectionby a High Ionization Arc (O’Dell et al. 2020a).Within the bubble is the Orion-S MolecularCloud that is seen in absorption against theMIL radio continuum (Johnston et al. 1983;Mangum et al. 1993; van der Werf et al. 2013).The observer’s side of the Orion-S MolecularCloud is illuminated by θ Ori C, producing anionized layer that is the optically brightest por-tion of the Huygens Region (Paper-II). Proceed-ing outward it encounters a layer of ionized gasdesignated as the Nearer Ionized Layer (NIL)(Abel et al. 2019), Paper-I, and Paper-II. Fur-ther out still there are two layers of atomic gas and H (Abel et al. 2016). These were discov-ered in H I II ] 158 µ memission (Pabst et al. 2019, 2020).In Paper-I and Paper-II we drew on the high-spectral-resolution Spectroscopic Atlas of OrionSpectra (Garc´ıa-D´ıaz et al. 2008) (henceforth‘the Atlas’), compiled from a series of north-south spectra at intervals of 2 (cid:48)(cid:48) . The Atlas hasa velocity resolution of 10 km s − and a see-ing limited spatial resolution of about 2 (cid:48)(cid:48) . Weanalyzed high signal-to-noise (S/N) spectra of[N II ] and [O III ], averaged over spatial boxes of10 (cid:48)(cid:48) × (cid:48)(cid:48) and groupings of these Samples intolarger areas called Groups and Regions. InPaper-I we established the large-scale proper-ties of the Huygens Region (the well-studiedbright region usually identified with M42, theOrion Nebula), establishing that this region hasa series of large-scale structures, the inner-mostbeing the optically bright Main Ionization Front(MIF) and the outer-most being a Veil of atomicand molecular gas.In Paper-II we explored at low spatial resolu-tion the region to the SW of the dominant ion-izing star θ Ori C, establishing that the areaincluding the
Orion-S Cloud defies explana-tion by simple models. The
Low-Ionization-Group was marked by very different behaviorsof radial velocities. In [O
III ] these velocitiesgroup at two values, contrary to the expecta-tions of photoevaporation from an ionizationfront and is in contrast with the simpler behav-ior in [N II ].In the current paper we target the Low-Ionization-Group , which overlies the thirdstar formation region in the Huygens Region,at higher spatial resolution. It lies 50 (cid:48)(cid:48) at 234 ◦ from θ Ori C and must be associated with the
Orion-S Cloud (it lies at the NE corner ofthe 21-cm H I contour of the Orion-S Cloud .The
Orion-S Cloud is seen in radio absorp-tion lines (hence it must lie in front of a sourceof radio continuum). The young stars lie onthe east side of the Cloud and are the sourceof many collimated molecular and ionized out-flows (jets). Shocks associated with these jetsspan the Huygens Region. Extrapolation of thejets backwards gives their origins with varyingdegrees of accuracy. Good presentations of theradio sources and jets are in Fig. 1 of O’Dell etal. (2009) and Figures 15, 16, and 17 of O’Dellet al. (2008). A more recent study (O’Dell et al.2015) refines the idea that although the sourcesare imbedded and not seen in the optical, theylie close enough to the PDR that many of theirjets break out and become optically visible fea-tures. This means that the geometry of the neb-ula is very different near this star formation re-gion.1.2.
Nomenclature and adopted values
The list of terms and adopted values are pre-sented in Paper-II, but additional terms are nec-essary in this study and an amended list is givenbelow. • Slit spectra are narrow rectangular areas ofsamples analogous to a short slit spectrum. • Profiles are the data from a series of slit spec-tra ordered along a single direction. • Adjacent Samples are a set of spectra se-lected to avoid the High Ionization Arc and theregion surrounding the intersection of the Pro-files called the
Crossing . • Samples are areas of 10 (cid:48)(cid:48) × (cid:48)(cid:48) within whichspectra from a spatially resolved atlas of spectraof certain emission-lines have been averaged. • Regions are groupings of Samples. • Sectors are groupings of Adjacent Samples,Samples, and Regions grouped by orientationrelative to the
Crossing . • The adopted distance is 388 ± • The adopted velocity for the backgroundPDR is V PDR = 27.3 ± − (Goicoecheaet al. 2015). • All velocities are expressed in km s − in theHeliocentric system (Local Standard of Rest ve-locities are 18.1 km s − less). • Directions such as Northeast and Southwestare often expressed in short form as NE andSW. 1.3.
Outline of this paper
In Section 2 we describe the division of thecomponents of the resolved line profiles (Sec-tion 2.1), how the rectilinear arrays of spatiallyresolved spectra from the Atlas were used to cre-ate series of spectra of a few arcseconds widthat orientations useful for analyzing the Orion-S Cloud and groupings of the slit spectra into’Adjacent Samples’ of higher signal to noise ra-tio (Section 2.3. The location of the Adja-cent Samples and their grouping are describedin Section 2.3.1. Profiles previously presented(O’Dell 2018) are presented in Section 4, nowannotated for the current study. The methodsof analysis of the spectra are descibed in Sec-tion 3.An analysis of the profiles is presentedin Section 4. A compilation of the data fromPapers I and II, and this paper is given in Sec-tion 5. Section 6 discusses the observationalproperties and their relation to the 3-D prop-erties of the Orion-S Cloud and a summary ofour conclusions and recommendations appear inSection 7. Appendix A presents a detailed studyof an important small feature on the
Orion-SCloud . OBSERVATIONS2.1.
Characteristic Velocity Systems
In Paper-I and Paper-II we have describedhow the spectra were created and de-convolvedinto velocity components. We repeat their de-scription presented in descending velocity. V scat (ascribed to backscattering from dust particlesin the background Photon Dominated Region(henceforth PDR)), V new and V red , [O III] (theformer ascribed to material in the High Ioniza-tion Arc feature and the latter to material accel-erated towards the host Orion Molecular Cloudby the stellar wind of θ Ori C) or materialthat would normally be assigned to V long , [O III] except for when the V short , [O III] component ismuch stronger, V long , [O III] (ascribed to emis-sion from the ionized layer, the MIF, on theOrion Molecular Cloud (OMC) or the ionizedlayer of the Orion-S Cloud facing the observer,and V short , [O III] (usually weak and ascribed toa foreground Nearer Ionized Layer (the NIL )lying in the foreground of θ Ori C). Thesecomponents are seen in both the [N II ] and[O III ] emission-lines. In Paper-I and earlierstudies the V short , [N II] and V short , [O III] compo-nents were called V low and the V long compo-nents were called V mif . As in Paper-II, whendiscussed as emission from specific physical lay-ers, the terms V mif and V NIL are used.2.2.
Twin Strong Components are seen inSpectra within the
Crossing
In Papers-I and II, usually the strong MIFvelocity component dominated each spectrumalthough in Paper-II we found Samples where V short was stronger and V long was not detected.Typical spectra and the short-comings of de-convolution of the weaker secondary compo-nents were discussed and illustrated (O’Dell2018; O’Dell et al. 2020a). Such spectra wereencountered in the present study, but in theregions within the Crossing we often foundthat V short , [O III] and V long , [O III] components areboth strong.In Figure 1 we show a good example of a twinstrong component [O III ] spectrum. This isspectrum 29 in the South-North Profile (Fig-
V(heliocentric) (km/s)-50 0 5010020001002003004005000 ‘splot’ Solutionshowing individualline components.‘splot’ Solutionshowing total signalof fitted components.
Figure 1.
This spectrum illustrates [O
III ] spec-tra in the middle of the
Crossing region, as de-scribed in Section 2.2. The top panel shows theobserved line profile in black and the fitted com-ponents in color V short -Blue, V long -Orange, V scat -Red). In the lower panel, the observed line profileis in solid black and the composite of the three fit-ted components is a barely distinguishable dottedline. The solutions using IRAF are: V short , [O III] = 5.3 km s − ( S short , [O III] / S long , [O III] = 0.86), V long , [O III] = 21.0 km s − ( S long , [O III] = 1.00), V scat , [OIII] = 39.1 km s − ( S scat , [O III] / S long , [O III] = 0.23). ure 5) and was deconvolved using IRAF task‘splot’.2.3. Higher spatial resolution slit spectra
We use small spatial samples as we study the
Crossing , whereas in Papers I and II we used10 × (cid:48)(cid:48) samples. This allowed an appropriately IRAF is distributed by the National Optical AstronomyObservatories, which is operated by the Association ofUniversities for Research in Astronomy, Inc. under coop-erative agreement with the National Science foundation.
Figure 2.
This 194 (cid:48)(cid:48) × (cid:48)(cid:48) (0.36 × (cid:48)(cid:48) at PA (PositionAngle) =215 ◦ from θ Ori C. The background im-age is from HST WFPC2 images coded by color(F658N, [N II ] Red; F656N, H α , Green; F502N,[O III ], Blue) (O’Dell & Wong 1996). It showsthe location of the Adjacent and
Crossing
Sam-ples discussed in Section 2.3. The three
Crossing samples for each profile are not named because ofcrowding but are shown. The
Crossing circle hasa diameter of 30 (cid:48)(cid:48) and is centered at 5:35:13.95 -5:23:49.2 (2000). The SW-Sector white lines showthe boundaries for those samples to the SW thatexclude
Crossing
Samples. better spatial resolution in the
Crossing . Thiswas done using the same sequences of spectraas in O’Dell (2018) and newly created N-S se-quences of spectra.2.3.1.
Previously Used Slit Spectra Profiles
The first type of samples are taken fromO’Dell (2018) where a series of pseudo-slit spec-tra of approximately 3 . (cid:48)(cid:48) × (cid:48)(cid:48) –13 (cid:48)(cid:48) were gatheredinto congruent sequences called Profiles. Their locations are shown in Figure 2 and the Pro-files themselves in Figure 5. These are the sameas in O’Dell (2018) except that they are nowannotated for this study. Our analysis of theseProfiles is quite different than in O’Dell (2018)in that we now recognize the importance of thecenter of the Crossing and know when to groupspectra into higher S/N ratio data.Some groups are within the
Crossing (calledthe
Crossing -Samples) and others (called theAdjacent-Samples) were selected to be closeto the
Crossing but avoided highly struc-tured regions outside of the
Low-Ionization-Group . The Crossing-Samples do not overlapcompletely, thus their results need not be ex-actly the same, but collectively they are repre-sentative of conditions within the
Crossing .Upon examination of the results for the Pro-files, we identified three regions of similardata. The NE-Sector includes the North, NE,and East Adjacent-Samples, the
Crossing con-tains the central Adjacent-Samples, and theSW-Sector contains the South, SW, and WestAdjacent-Samples. Their locations are shownin Figure 2.We present the averaged data of the multiplesamples in Table 1. Reading left to right onesees a progression of samples essentially flowingfrom the NE to the SW of the
Crossing . Verti-cally the results are ordered with the [N II ] datain the first five lines, the [O III ] data in the nextfive, and the joint [N II ] and [O III ] data on thelast line.2.3.2.
Newly Created Slit Spectra Sequences
The second type of samples were created forthis study. Within the
Crossing , sequences of12 2 . (cid:48)(cid:48) (cid:48)(cid:48) in Right As-cension (RA) were made, as shown in Figures 3and 4. This rectilinear array has the advan-tage of higher spatial resolution (2 × (cid:48)(cid:48) ) and itbetter samples a nearly linear E-W feature inthe middle of the Crossing . This is a welldefined ’squiggly’ feature, called here the
Ex-
Table 1.
Data from the Profile Spectra used in this study*
NE-Sector Adjacents
Crossing
SW-Sector Adjacents V long , [N II] ± ± ± V short , [N II] ± ± ± V scat , [NII] ± ± ± V scat , [NII] - V long , [N II] ± ± ± V long , [O III] ± ± ± V short , [O III] ± ± ± V new , [O III] ± V red , [O III] — 19 ± ± V scat , [OIII] ± ± ± V scat , [OIII] - V long , [O III] ± ± ± S long , [N II] / S long , [O III] † ± ± S long , [N II] / S short , [O III] ± ± ± S long , [N II] /) — — —( S long , [O III] + S short , [O III] ) 1.94 ± ± ± − .**Samples group around these values. † S long , [N II] / S long , [O III] = 1.00 corresponds to a calibrated surface brightness (erg s − cm − sr − )ratio of 0.13. †† Without the highly scattered N-S Profile (19.6 ± tended Ledge , that was discussed in O’Dell etal. (2015) (where it was called the West-Jet).We determined membership in the ExtendedLedge using higher surface brightnesses andvelocities as a guide and then derived aver-ages of regions to the north and south of this(the Crossing-North and Crossing-South Spec-tra groups, henceforth North-Array and South-Array), with the results shown in Tables 2, 3,and 4. We believe that the disadvantage of av-eraging over varying features is appropriate be-cause of the many samples in these groups.Within the
Extended Ledge there is a [N II ]bright E-W linear feature. Because of its lackof apparent motion in the plane of the sky thisis not part of a moving jet. It is most likelyto be a small portion of the MIF that is al-most along the observer’s line-of-sight. We nowrefer to this as the Ledge and it must be asmall local escarpment with the higher side tothe south. There are other features that arestrong in [O
III ] on both sides of the
Ledge with the ones on the west have measurable tan-gential motions. The
Extended Ledge and its location is shown in Figures 3 and 4. The ledgeis pointed out in Figure 6.The
Ledge is discussed in detail in Ap-pendix A .In addition to velocity components that fitinto the usual classifications, very blue and weak[O
III ] velocity components were seen in a few ofthe 34 (cid:48)(cid:48) -West, 42 (cid:48)(cid:48) -West, and 46 (cid:48)(cid:48) -West profiles.The data from these three regions are givenwithin Tables 2, 3, and 4. In addition, resultsfrom Papers I and II are given for regions out-side these samples. ANALYTIC TOOLS3.1.
Velocity variations when crossing tiltedionization layers
A velocity profile across an escarpment suchas the Bright Bar will produce a single velocitypeak, followed by a decrease to the velocity inthe region beyond the escarpment, as shown inFig. 6 of O’Dell (2018). When a profile crosses adiscreet cloud, one would expect to see a doublevelocity peak, the first on the side illuminatedby θ Ori C, the other on the far side of the -15.2-17.4-19.5-21.6-23.8-25.9-28.1-30.2-32.3-34.4-36.6-38.7 30 34 38 42 46
Figure 3.
This 46 (cid:48)(cid:48) × (cid:48)(cid:48) FOV is centered on thesymmetry axis of the Dark Arc (within the
Cross-ing which is shown with a dark circle). The boxesshow the locations of the sampled spectra as dis-cussed in Section 6. Those outlined in yellow definethe
Extended Ledge samples, those to the north(up) the Dark Arc-North-Samples, and those to thesouth the Dark Arc-South-Samples. The columnsare marked with the west displacement in arcsecfrom θ Ori C and the individual north-south spec-tra in arcseconds south of θ Ori C.The image wasmade with the HST WFPC2 camera in the [N II ]filter (F658N). -15.2-17.4-19.5-21.6-23.8-25.9-28.1-30.2-32.3-34.4-36.6-38.7 30 34 38 42 46 Figure 4.
Like Figure 3 except now showing the[O
III ] F502N image. cloud. The magnitude of the velocity variationsdepend on V PDR and V evap (the photoevapora-tion flow velocity). When the geometry is flat-on the observed V long will be V PDR - V evap . Asthe MIF becomes tilted the projection of the V evap components will be reduced and at anedge-on geometry one expects V long = V PDR . Ifthe tilt is not completely edge-on, the V long vari-ation will not be as great. As noted above, wewould see this up and down velocity variationonce in the profile of an escarpment and twicewhen the profile cross both sides of an isolatedcloud. The velocity pattern would be clearestwhen the profile cuts across the steepest part ofthe tilted front and would be less visible if theprofile crosses it at an angle. Likewise, becauseof the difference in the thickness of their emit-ting layers, we would expect to see the patternbetter in the thinner V mif , [N II] emitting layerthan the thicker V mif , [O III] emitting layer.The Bright Bar is the best example of a singlepeaked velocity profile indicating that it is theedge of an escarpment and provides an aid whenlooking at similar features. The ionized portionshows signal peak and velocity maximum thatis expected (O’Dell 2018) and infrared and ra-dio studies of neutral gas shows the stratifiedpresence of atoms and molecules consistent withand verifying PDR models Tielens et al. (1993);Goicoechea et al. (2016).The velocities and relative strengths of boththe [N II ] and [O III ] lines are presented in Ta-ble 1.3.2.
Velocity differences of the V scat , V long ,and V short components As noted previously (Paper-II) there is an ex-pected relation between an emitting layer ve-locity ( V emit ) and its backscattered componentvelocity ( V scat ) and V scat - V emit that is depen-dent upon the photoevaporation flow velocity( V evap ), V PDR , and the tilt of the MIF. Underthe assumption that the tilt is the same for theMIF and the V emit producing layer and that V evap is constant throughout the area being ex-amined, a flat-on view would have V scat - V emit (cid:39) × V evap (Henney 1998) If the tilt increasestowards an edge-on configuration, the line-of-sight component of V evap will decrease, produc-ing a decrease in the observed V scat - V emit andan increase in the observed V emit . Applicationof this information discussed in Section 6.3.3.3. Relative strength of the red-shiftedbackscattered component
The ratio of signals (S) of the redshiftedshoulder component of an observed line com-pared with a lower velocity component informsthe question of what emission-line componentis being backscattered. The observed ratio S scat / S comp must be much less that unity, re-flecting the fact that the effective albedo mustbe low, unless there is a strong backscatteringphase function. A large ratio means that this V comp is not the source. If the source and thePDR are widely separated, the ratio will be un-usually small.3.4. Surface Brightness variations
For a photoevaporating ionization front theSurface Brightness (SB) in an H I recombina-tion line varies as the incident ionizing radia-tion (Baldwin et al. 1991; Osterbrock & Ferland2006). For a face-on flat MIF, this means thatthere would be a monotonic decrease in the SBat increasing distances (in the plane-of-the-sky)from the sub- θ Ori C position. A concave MIFwould have a slower decrease in SB with increas-ing angular distance and a convex MIF wouldhave a more rapid decrease. The general con-cave structure of the inner Huygen’s Region iswell established. Features within the Huygen’sRegion, such as the Bright Bar, are explainedas steep rises in the MIF. The high SB there isexplained by both the ionizing flux increasingdue to the tilt and the fact that one is lookingat the emitting layer edge-on. More recently (O’Dell 2018), the same geom-etry has been applied to explain why the bright-est part of the nebula occurs to the NE of theOrion-S Cloud. Although the increase in surfacebrightness is certainly due in part to the tilt, theproximity to θ Ori C of the NE boundary tothe Orion-S Cloud must also be very important.In Paper-I we established that the distance of θ Ori C from the MIF ionization boundary wasin the range 0.1 – 0.2 pc. The central valueof 0.15 pc corresponds to 81 (cid:48)(cid:48) if projected ontothe plane-of-the-sky. The peak surface bright-ness SB in [N II ] occurs at 33 (cid:48)(cid:48) (0.061 pc) from θ Ori C in the plane-of-the-sky. In Section 4.1we pointed out that the Cloud is no closer tothe observer than the plane including θ Ori Cor 0.05 pc (27 (cid:48)(cid:48) ) beyond that plane. The cor-responding range of physical distances between θ Ori C and the NW edge of the Cloud is 0.061– 0.079 pc, both are closer to θ Ori C than the θ Ori C to the MIF ionization boundary. Evenwithout consideration of the enhancement dueto looking along an emitting layer seen edge-on,this explains why the SB in the Huygens Regionis highest there.Of course the expectations for the SB becomesmore complex when dealing with [N II ] and[O III ] emission that occur in different zoneswithin the ionized hydrogen layer, but even fortheir emission the first order expectations re-main the same after consideration that the [N II ]emitting layer is thinner and closer to the ac-tual ionization front than the [O III ] emittinglayer. However, if a region is lower ionization,then [N II ] emission will be enhanced relative to[O III ] emission. ANALYSIS OF THE PROFILES4.1.
Analysis of the Profiles in and near the
Crossing
The northeast side of the
Orion-S Cloud shows the velocity and ionization changes char-acteristic of an ionization front viewed edge-on(Mesa-Delgado et al. 2011; O’Dell 2018; O’Dellet al. 2020b), illuminated by θ Ori C. The tran-sition is shown in Figure 2 where the [O
III ]dominated region transitions to an [N II ] dom-inated region along a SE-NW line. To the SWof this line lies the Orion-S Cloud ionized onthe observer’s side by θ Ori C. Our
Cross-ing
Samples are taken in the region of the DarkArc feature (c.f. Section 4.2.4) and overlap theNE 21-cm absorption boundary of the Orion-SCloud (van der Werf et al. 2013). The
Orion-S Cloud is also seen in absorption in H CO(Johnston et al. 1983; Mangum et al. 1993).The velocities of these features are the same asthe OMC in this direction (Tatematsu et al.1998; Peng et al. 2012; Troland et al. 2016).Since they are seen in absorption against an ion-ized gas continuum, the common interpretationis that this is a cloud lying within the main cav-ity of the nebula with portions of the MIF lyingbehind it. In Paper-I we established that the
Orion-S Cloud lies at the same distance fromthe observer as θ Ori C or no more than 0.05pc beyond it.4.2.
Velocity Variations
Given the guidelines described in Section 3.1we can use the velocity variations in the profilesshown in Figure 5 to determine the geometry ofthe
Crossing .4.2.1.
What do velocity variations in [N II ] tellus? Since the [N II ] emission comes from a thinlayer close to an ionization front, it is notsurprising the V long , [N II] changes demonstratemore continuous profiles than V long , [O III] . Inthis section we will discuss the V long , [N II] pro-files and have always assigned the strongestcomponent to V long , [N II] .It has been established (Mesa-Delgado et al.2011; O’Dell 2018) that the Orion-S Crossing is immediately southwest of an ionization frontviewed more nearly edge-on, like the Bright Bar. The question then becomes whether its velocityprofiles show a single peak (hence it is an es-carpment) or a double peak (hence it is a cloudilluminated on both sides.Each profile shows a slightly different varia-tion in V long , [N II] , reflecting the fact that theytrace different paths across the Crossing andnearby areas. The E-W Profile shows a broadpeak in V long , [N II] occurring at spectrum 13 anda small local rise at spectrum 20, which lies onthe east boundary of the Cloud’s V long , [N II] ,the first at spectrum 28 (closest to a surfacebrightness maximum) and a second at spectrum21 (outside of the Crossing , but well insideof the southern crossing of the High IonizationArc so that it is not the source of this velocitypeak). Further north there is a pair of veloc-ity peaks (spectra 42 and 48) as one crosses theHigh Ionization Arc, indicating that this is ashell of material. The NE-SW Profile shows apeak in V long , [N II] near spectrum 10, with oth-ers at spectra 13 and 20. The peak at spectrum13 may not be real as the surrounding spectra V long , [N II] lines are broad and may be blendsof higher and lower velocity components. Thiswould mean that there is the peak at spectrum10, followed by a continuous reduction until theweak peak at spectrum 20.Taken together, the V long , [N II] profiles indi-cate the crossing of a tilted ionization frontnear the center of the Crossing , followed bya flattening of the front, then crossing the outerboundary of this higher feature at about 30 (cid:48)(cid:48) from the center of the
Crossing .It is surprising that crossing the edge of araised feature (in the middle of the
Cross-ing ) occurs displaced from the transition regionshown in Figure 2. It is as if this is a localstructure superimposed on the broader
Low-Ionization-Group .0 South-North Profile
Black--[NII]Red-[OIII]SouthNorth
Spectrum Number u CN SC W NE-SW Profile Black-[NII]Red-[OIII]Spectrum Number Black--[NII]Red-[OIII]
Northeast Southwest C0 CNE SWC
Black--[NII]Red-[OIII]East-WestProfile Spectrum Number
Black--[NII]Red-[OIII]WestEast C0 C E WC24135 v LedgeLedge0 C Figure 5.
Like Figures 10–12 from O’Dell (2018) except annotated for this study. Large filled boxes long-components , small filled boxes scattered-components , horizontal boxes V new and V red , [O III] , and largefilled diamonds short-components. Black indicates [N II ] and red [O III ]. The spectra spacing is 3 . (cid:48)(cid:48) . (cid:48)(cid:48) θ Ori C indicates the spectrum lying closest to θ Ori C. Orion-S
Crossing indicates spectra within the
Crossing . The low red lines indicate the spectraincluded in the Adjacent Samples. The letter W indicates where the FWHM of V long , [N II] ≥ − or when the FWHM of V long , [O III] ≥ − . The open red circles depict the sum of the S short , [O III] and S long , [O III] components in the Crossing . We can safely conclude that the profiles areacross a raised feature (closer to the observer)because we see the expected local maximum inthe surface-brightness ( S long , [N II] ) at each of thefirst V long , [N II] peaks (crossing a depressed fea-ture would produce S long , [N II] minima becauseof shadowing of θ Ori C radiation).The V short , [N II] behavior is harder to track be-cause it is a usually a weak feature on the blueshoulder of the much stronger V long , [N II] com-ponent. The limitations of similar spectra arediscussed quantitatively in O’Dell (2018) andPaper-I. The lower signal to noise ratio of thesmaller spectra as compared with the 10 × (cid:48)(cid:48) samples used in Papers I and II, means that wecannot now use the V long data except where itssignal becomes strong. These are discussed inSection 6.1. 4.2.2. What do velocity variations in [O
III ] tellus?
Velocity variations in [O
III ] are much morecomplex that in [N II ]. The V short , [O III] com-ponent is often strong and important, whereasin [N II ] it was always weak as compared with V long , [N II] . Examining the different V long , [O III] profiles reveals the complex situation. The pro-files in Figure 5 show components classified us-ing the observed V r as a guideline, with the red-der assigned to V long and the bluer to V short .Proceeding from east to west in the E-W pro-file, we see that as one reaches the Cross-ing each spectrum suddenly has two significantcomponents ( S short , [O III] (cid:39) S long , [O III] ) and theyhave two very different velocities. After pass-ing through the Crossing S short , [O III] becomes1much less than S long , [O III] , but then increasesand finally matches S long , [O III] , although the ve-locities of these two components remain aboutthe same. V long , [O III] has broad peaks at thesame positions as V long , [N II] . The V short , [O III] values are accurate (i.e. not questionable be-cause of being a weak component on a strongerline’s shoulder). This belief is reinforced by thefact that V short , [O III] changes little as the com-ponents become of equal signal.Proceeding south in the S-N Profile we againsee two velocity peaks at spectra 42 and 48 asthe shell of the High Ionization Arc is crossed.In this region we see higher velocity components(labeled V new in Figure 5 that are probably as-sociated with the Arc. Moving further souththe velocity of V long , [O III] changes dramaticallyupon entering the Crossing . Suddenly thestronger component V long , [O III] drops to veloc-ity values usually assigned to V short , [O III] andremain there for the remainder of the profile.They are redder (20 ± − ) and weakercomponents (labeled V red , [O III] in Figure 5 thathave a wide velocity dispersion and velocitiessimilar to the V long , [N II] components (22 ± − ) in the other adjacent samples in theSW-Sector (Table 1).Proceeding SW in the NE-SW Profile we firstsee little V long , [O III] structure until reachingthe Crossing , at which point the V long , [O III] and V short , [O III] components become compara-ble signals. Results from spectra 12—18 are am-biguous because the strongest line is broad andthe lower V long , [O III] values there are in betweenthe previous V long , [O III] and V short , [O III] values.The few narrow lines at spectra 20–22 fall at ve-locities usually associated with the V long , [O III] and V short , [O III] systems.4.2.3. What do the velocity differences tell usabout the
Crossing geometry?
Examination of V scat - V long in Table 1 indi-cates that for the backscattered [N II ] compo-nent, the well-defined difference of 16 km s − is close to the expected value of 14 ± − fora flat-on viewing angle.The difference for [O III ] may change slightly,as the line-of-sight moves from the NE-SectorAdjacent-Samples (21 ± − ), through the Crossing (20 ± − ), to the SW-SectorAdjacent-Samples (18 ± − ). All of theseare unimportantly lower than the predictedrange 20 ± − .The agreements of observed and expected ve-locity differences indicate that the MIF emis-sion is in fact being backscattered by the nearbyunderlying PDR. The smaller dispersion of the[N II ] values indicates that there is not a bigchange in the viewing angle in the three regions.This is consistent with our intentionally not in-cluding the region of known high tilt on the NEboundary of the Orion-S Cloud. The same canbe said for the [O III ] emission, although withless certainty because of the larger probable er-rors.4.2.4.
Observed Surface Brightness variations
The geometry of the nebula that produced thevelocity variations described in Sections 4.2.1and 4.2.2 should also produce variations in theapparent surface brightness (SB) of the nebula.In this section we examine the SB variations us-ing the signal in the Atlas as in O’Dell (2018),where the conversion to power units is also ex-plained.None of the profiles cross through θ Ori C,but their minimum distances (the θ Ori C-tangent) from that star should be marked by alocal peak in SB. These θ Ori C-tangent points(spectrum 36 for the S-N Profile, spectrum -1for the NE-SW Profile, spectrum 3 for the E-WProfile) are shown in Figure 5. In the absence ofstructure in the region covered by the profiles,we would expect to see a single peak in the SB atthe θ Ori C-tangent location with a monotonicdrop in the SB in both directions away fromthis point. We see peaks in both ions near eachof the θ Ori C-tangent points and a continu-2ous drop into the NE-Sector, interrupted onlyin the S-N profile where it crosses the high ion-ization arc and an even more removed feature tothe north, seen only in S long , [O III] . Variations in V long , [N II] and V long , [O III] accompany the HighIonization Arc passage.We also see variations in the SB in both ions asone traces outward (into the SW-Sector) fromthe θ Ori C-tangent point. If these profiles sim-ply traced across an escarpment, one would ex-pect to see a monotonic rise to a SB maximumas the observed V long increases. This should bemore obvious in [N II ] because its emitting layeris thinner. The passage of the escarpment ac-counts for the S long , [N II] peaks in each profileat S-N Profile (spectrum 34), NE-SW Profile(spectrum 7), and E-W Profile (spectrum 9).Small increases in V long , [N II] can be attributedto the passage across the escarpment.It should also be noted that unlike [N II ],the sum of the S long , [O III] and S low , [O III] com-ponents in the Crossing approximate the total S long , [O III] value needed to provide a smooth in-terpolation of the SB over the Crossing as indi-cated by the open red circles in Figure 5. Thisis consistent with the dark features being lessconspicuous in [O
III ]. The comparable strengthof both components makes their measurementmore accurate than when S low , [O III] / S long , [O III] is low. The fact that the low components arestrong raises the possibility that in the Cross-ing the V short , [O III] emitting layer is competingwith the V long , [O III] emitting layer for the higherionization energy photons necessary to produce[O III ] as discussed in detail in Paper-II.The feature called here the
Extended Ledge appears to be a small escarpment region, withthe south side being higher. It is denoted inFigure 6. Although it looks like a jet, there areno tangential velocities (O’Dell et al. 2015) onits east end but several in [O
III ] and a few in[N II ] beginning at its west end. It is studied indepth in Paper-IV. The point of passage across this east-west ori-ented feature (Figure 6 is marked by the word Ledge in each panel of Figure 5. It is drawnwith a line in the E-W Profile because the pro-file passes along the axis of the feature. Ateach marked position there is an associated localpeak in S long , [N II] , attributable to the feature.Similarly, there is an increase in V long , [N II] atthese locations. This important feature is dis-cussed further in Appendix A.It is probable that a feature known as theDark Arc (O’Dell & Yusef-Zadeh 2000; O’Dellet al. 2015) also plays a role in the SB profiles.In Figure 6 we present our best resolution colorimage. The Dark Arc and Dark Box low SBfeatures are seen in both ions, especially [N II ].They are discussed in detail in Section 3.2.1 ofO’Dell et al. (2015). Although they are notunderstood, they are thought to be regions oflow emissivity near the ionized layer on the ob-server’s side of the Cloud, possibly small scaleescarpments produced by one or more of themany stellar outflows in this region. In the S-Nand NE-SW Profiles the Dark Arc feature oc-curs at the SB minimum. In the E-W Profilethe dip occurs at the spectrum including theeast boundary of the Dark Arc, where it is nearaligned nearly N-S.Examination of the [O III ] profiles showsthat the clear division into V short , [O III] and V long , [O III] components begins in the same sam-ples that dip in [N II ] SB. The V long , [O III] SBdrops to lower than interpolation from adja-cent spectra and the new V short , [O III] spectraare of comparable strong to the V long , [O III] com-ponents. Most significant is that the totalstrength of the V long , [O III] and V short , [O III] components fall onto a smooth interpola-tion of adjacent values. A COMPILATION OF DATA FOR THELARGE SCALE FEATURES ANDGROUPS3
Figure 6.
This 100 (cid:48)(cid:48) × (cid:48)(cid:48) (0.19 × Crossing (black circle) region andencloses the optically brightest part of the Huy-gens Region. The two small low surface bright-ness features (Dark Arc and Dark Box) discussedin Section 4.2.4 are labeled. The background im-age is from HST WFPC2 images coded by color(F658N, [N II ] Red; F656N, H α , Green; F502N,[O III ], Blue) (O’Dell & Wong 1996). The outerH I absorption contours with heavy red lines arefrom (van der Werf et al. 2013). The white rectan-gles indicate the location of the Central sections ofthe three profiles. Averaged results for the samples describedabove and nearby regions from Papers I and IIare given in Tables 2, 3, and 4 In addition tothe results of this study. The arrangement (topto bottom rows) is basically from NE to SW.We have not used the results from the SW-Region discussed in Paper-I because it overlapswith multiple smaller groups identified here andin Paper-II. Two groups rendered in italics areaffected by the slightly larger FWHM of the V long , [O III] component . In those (the Low-Ionization-Group and the
Outside-Group )this width means that there is no chance of see- ing a weak companion component and their val-ues must be used cautiously.Where there are two lines of entries for a groupin [O
III ], the property that showed the biggestdivision in entries is marked ***. The composi-tion of those subgroups was used to derive theother characteristics. When there was no obvi-ous difference in the two groups, a single value isgiven. The line containing the higher V long , [O III] subgroup is always located higher in Tables 3and 4. DISCUSSIONThe three compiled tables of data can be usedto investigate the structure and properties of aNE-SW swath across the Huygens Region.Examination of the results in Tables 2, 3,and 4 illustrates how the unusual behavior ofthe [O
III ] velocity components found over alarge area in Paper-II are first encountered inthe
Crossing , and that within the
Crossing the behavior originates near a narrow east-westfeature, the
Extended Ledge . To the NEthe V short , [O III] component is weak as comparedwith V long , [O III] , but this is reversed as onepasses the Crossing .These results refine the discovery in Paper-II that the V short , [O III] component appeared infive of the nine large samples that includedall or part of the Crossing . In addition,the V short , [O III] component appeared withouta V long , [O III] component in fifteen other Sam-ples to the south and southwest of the Cross-ing . This indicates that although the two ve-locity systems may originate in the
Crossing ,they occur throughout the
Low-Ionization-Group .In contrast, the V short , [N II] component is al-ways weak as compared with V long , [N II] . In Ta-ble 5 we see that S short , [N II] / S long , [N II] is 0.14,0.25, and 0.11 in the three Supergroups.We see in these tables that the [O III ] prop-erties within samples often break down intogroupings (subgroups), reflecting the fact that4
Table 2.
Data from All [N II ] Sources* Group Name V long , [N II] V short , [N II] V scat , [NII] - V long , [N II] S scat , [N II] / S long , [N II] NE-Region 22 ± ± ± ± ± ± ± ± ± ± ± ± Low Ioniz.-Group ± ± ± ± North-Array ± ± ± ± Extended Ledge ± ± ± ± South-Array ± ± ± ± ± ± ± ± ± ± ± ± − . Table 3.
Data from All [O
III ] Sources-Velocities*
Group Name V long , [O III] V short , [O III] S short , [O III] / S long , [O III] V scat , [OIII] - V long , [O III] V scat , [OIII] - V short , [O III] NE-Region 18 ± ± ± ± ± ± ± ± ± ± ± ± Low Ioniz.-Group †† ± ± —** ± ± North-Array ± ± ± ± ± North-Array ± ± ± ± Extended Ledge ± ± ± ± ± South-Array ± ± ± ± ± South-Array ± ± ± ± ± ± ± ± ± ± ± ± Outside-Group †† — ± >> — ± * All velocities are Heliocentric velocities in km s − .** V long , [O III] and V short , [O III] were not found in the same samples.***Samples group around these values. † S long , [N II] / S long , [O III] = 1.00 corresponds to a calibrated surface brightness (erg s − cm − sr − ) ratio of 0.13. †† Results are affected by the unusually large FWHM of the V long , [O III] components. within the scale of the features there are a va-riety (at least two) of characteristics. The sub-grouping begins in the North-Array and extendsinto the SW-Sector.The data in Tables 2, 3, and 4 can also beused to derive other properties of the regionssampled, as shown in the following sub-sections.6.1. Supergroups and the Relation of the shortvelocity features to the
NIL
The Nearer Ionized Layer (the NIL) lies acrossmuch of the Huygens Region and is best char-acterized in Paper-I, where its velocity and rel- ative strength is discussed and modeled, usingdata from near θ Ori C. It is appropriate tosee if the V short components we examine in thisstudy are part of the NIL or have been alteredby the conditions on the ionized layer atop theOrion-S Cloud. We have done this using thevelocities and signals.We gathered the groups into ’Super-Groups’of NE-Supergroup (NE-Region, Inside-Group, NE-Sector),
Crossing-Supergroup (North-Array,
Extended Ledge , South-Array), and
SW-Supergroup (SW-Sector).Within each Supergroup we created a subgroup5
Table 4.
Data from All [O
III ] Sources-Signal Ratios*
Group Name S scat , [O III] S scat , [O III] S scat , [O III] S long , [N II] S long , [N II] S long , [N II] / S long , [O III] / S short , [O III] / S both , [O III] / S long , [O III] / S short , [O III] / S both , [O III] † NE-Region 0.06 ± ± ± ± ± ± Low Ioniz.-Group †† ± ± —** ± ± —** North-Array ± ± ± ± ± ± North-Array ± ± ± ± ± ± Extended Ledge ± ± ± ± ± ± Extended Ledge ± ± South-Array ± ± ± >> ± ± South-Array ± ± ± >> ± ± ± ± ± ± ± ± ± ± ± ± ± ± Outside-Group †† — ± — — ± —* All velocities are Heliocentric velocities in km s − .** V long , [O III] and V short , [O III] were not found in the same samples.***Samples group around these values. † S long , [N II] / S long , [O III] = 1.00 corresponds to a calibrated surface brightness (erg s − cm − sr − ) ratio of0.13. †† Results are affected by the unusually large FWHM of the V long , [O III] components. Table 5.
Data for Supergroups*
Property
NE-Supergroup Crossing-Supergroup SW-SupergroupV long , [N II] ± ± ± V short , [N II] ± ± ± V long , [O III] ± ± ± V short , [O III] ± ± ± AveS long , [N II] ±
90 412.5 ±
157 129 ± AveS short , [N II] ± ±
19 13.8 ± AveS long , [O III] ±
47 71.4 ±
15 30.2 ± AveS short , [O III] ± ± ± AveS long , [N II] / AveS long , [O III] AveS long , [N II] / AveS short , [O III] − .**Samples group around these values. Parentheses enclose the number of samples within eachsubgroup. where velocities and signals existed for all com-ponents. Within these subgroups we deter-mined the average V short , [N II] and V short , [O III] ,with the results shown in Table 5.We also created Average signals for the sub-groups in both the long and short componentsof both ions. These are preceded with Ave inTable 5 and are more accurate measures of the surface brightness, whereas signals over samplesare a mix of features from different parts of thesamples. The ± symbols do not indicate uncer-tainties, rather, they are the 1- σ spreads of thesamples. The averaged values are used in ourdiscussion of ionization changes (Section 6.2. V short , [N II] shows the widest variation in valuebetween the Supergroups. The Crossing- Supergroup ’s value of 6 ± ±
2) between the ad-jacent Supergroups. If this marginal evidenceis accepted, then it is an indication that the[N II ] emitting layer of the NIL that lies infrom of the Orion-S Cloud has been affected,but in the opposite sense expected from photo-evaporation flow from the ionized surface of the
Orion-S Cloud (that would be a more negative V short , [N II] ). In our discussion of the full pro-files passing over the Crossing (Section 4.2.1),we saw that this region’s V long , [O III] values in-dicate that it is a raised region, which is con-sistent with its being part of the ionized layeron the observer’s side of the Orion-S Cloud .If the higher value of V short , [N II] is accepted itcould be due to interactions with the Orion-SCloud ’s surface, thus arguing for a small sepa-ration of positions. In this case the more posi-tive velocity would have to arise from flow awayfrom the NIL and the observer.We see that the V short , [O III] is essentially con-stant across the Supergroups at 5 ± − .This places it at a characteristic velocity of theNIL and argues that the velocity of the [O III ]component of the NIL is not affected by the
Orion-S Cloud .6.2.
Changes of ionization in and near the
Crossing
Since we see remarkable changes in the S short , [O III] / S long , [O III] ratio as one moves fromthe NE to the SW it is important to determine ifthese are due to one component increasing whilethe other decreases, or vice versa. This is bestdone with the Ave signals given in Table 5.There we see that
AveS long , [N II] has de-creased more than a factor of 0.75 when go-ing from the NE to the SW side of the cross-ing, while AveS long , [O III] has dropped by afactor of 0.28. This means that the ratio of AveS long , [N II] / AveS long , [O III] has changed from1.6 to 4.3, as shown in Table 5. This supportsa conclusion that the ionization drops with in- creasing distance from θ Ori C. Much of thischange of ratio is due to the dramatic drop in
AveS long , [O III] SW of the
Crossing while the
AveS short , [O III] has increased five-fold from the NE-Supergroup values6.3.
What do we learn from theBack-scattering?
We can apply the methods described in Sec-tions 3.1 and 3.3 to the interpretation of theregions immediately to the NE of the
Cross-ing (the NE-Sector) , the
Crossing itself, andthe regions immediately to the SW (the SW-Sector). For this we have employed the spectrain the profiles, as described in Section 4. Theresults of our observations and predictions areshown in Table 6.As noted in Section 3.3, the ratio S scat / S comp should be much less than unity and this can beused to verify that one has correctly identifiedthe source of the V scat component. This con-dition is satisfied for all of the [N II ] samplesand for most of the [O III ] samples. However,in the
Crossing we see that the more numer-ous V long , [O III] component satisfies the condi-tion while in V short , [O III] the less numerous do.The same pattern applies in the SW-Sector (al-though the population of the groups are aboutequal).It is informative to compare the observedvelocity difference ( V scat - V comp ) with the pre-dicted value. For a flat-on region the observed V comp will be V PDR - V expansion from which thephoto-ionization flow velocity for each emission-line can be calculated. We have used V PDR =27 km s − and from this calculated a predicted V scat - V comp . This value should decrease in amore highly tilted region.In [N II ], the long component is dominant(c.f. Table 2. The agreement of V scat , [NII] - V long , [N II] observed and predicted values is goodfor the NE- and SW-Sectors, which are nothighly tilted, and the difference in the Cross-ing indicates the complexity of the region.7
Table 6.
Predicted and Observed Back-scattering Velocities*
Velocity Component Predicted ( V scat - V comp ) Observed ( V scat - V comp ) S scat / S comp NE-SectorV long , [N II] ± ± ± V long , [O III] ± ± ± V short , [O III] ± CrossingV long , [N II] ± ± ± V long , [O III] ± ± ± ± ± V short , [O III] ± ± ± ± ± SW-SectorV long , [N II] ± ± ± V long , [O III] ± ± ± ± ± V short , [O III] ± ± ± ± ± − .**No V short , [O III] component is seen**Samples group around these values. Parentheses enclose the number of samples within each subgroup. The S scat , [N II] / S long , [N II] values are always lowenough to confirm that V long , [N II] is the sourceof the backscattered light.In the NE-Sector S short , [O III] / S long , [O III] issmall( to the point that V short , [O III] is not de-tected). The observed and predicted V scat , [OIII] - V long , [O III] are in good agreement.In the Crossing , we see that the observed andpredicted velocity difference is very different for[N II ]. This probably reflects the complexity ofthis region owing to the many smaller regionswith large tilts.Also in the Crossing we see that a larger sam-ple satisfies the S scat , [O III] - S long , [O III] require-ment, but that its observed velocity differentagree’s less well than the smaller group. How-ever, the probable errors are such that it maybe indistinguishable. Neither V short , [O III] groupagrees with the predictions. Again, this indi-cates the complexity of the Crossing .In the SW-Sector the observed and predictedvelocity differences for [N II ] are excellent.Again the V long , [O III] component breaks downinto favored and unlikely scatters, but bothcomponents agree equally well with the predic-tions. Like the Crossing , the observed and pre-dicted values are highly discrepant. In summary, we see that in [N II ] the observedand predicted velocity differences are good inthe NE and SW-Sectors, but poor in the com-plex Crossing . In [O
III ] we see that the agree-ment is good in the NE and SW-Sectors, but inthe
Crossing there are no good agreements.6.4.
A different V evap velocity for [O III ] The predicted values of V scat - V comp in the pre-vious section were calculated from inferred val-ues of V evap and that the velocity separationwould be twice V evap . The argument can bereversed to use the observed velocity differenceto derive V evap . This tells us that V evap , [N II] isthe same in all three Supergroups, about 8 ± − . Using the lower S scat , [O III] / S long , [O III] groups gives V evap , [O III] = 10 ± − . Theslightly larger value for [O III ] is consistent withthe idea (Henney et al. 2005)that the [O
III ]emitting zone is further from the MIF and thushaving been subjected to more acceleration. CONCLUSIONS • The [N II ] emission arises from a narrowlayer along an ionization front. The V long , [N II] component is always much brighter than the V short , [N II] component. In the region nearest θ Ori C it clearly lies along the MIF and in8the
Crossing along the ionized surface of the
Orion-S Cloud that faces the observer. To theSW its location is uncertain as it could be eitherover the main body of the
Orion-S Cloud oron the MIF of the nebula beyond the embedded
Orion-S Cloud . Strong backscattering of itsradiation indicates that it is always close to adusty PDR. • The V short , [N II] component arises from theNIL, the layer of gas lying on the observer’s sideof θ Ori C and the
Orion-S Cloud . Its surfacebrightness is the same in the
NE-Supergroup and the
SW-Supergroup , even though the lat-ter is much more distant in the plane of the sky.The surface brightness is unexpectedly muchhigher in the
Crossing-Supergroup . Its ve-locity is almost the same when sweeping fromNE to SW, but there is a possible increase overthe
Crossing-Supergroup . If real, this wouldindicate that the NIL is closer to θ Ori C thanwe calculated in Paper-I. • The V long , [O III] component arises froma thicker, more highly ionized region than V long , [N II] . Its velocity is almost constantwhen passing through the Supergroups .Like V long , [N II] , the NE-Supergroup emissionarises from the MIF beyond θ Ori C and the
Crossing-Supergroup on the observer’s sideof the
Orion-S Cloud . Again, the physicallocation is uncertain in the
SW-Supergroup .Its surface brightness decreases monotonicallywith increasing distance from θ Ori C. The lackof an increase in the
Crossing-Supergroup in-dicates that its emitting layer is thicker thanthe physical high point indicated by the [N II ]profiles. The [O III ] profiles indicate that atthe
Crossing the surface begins to drop andbeyond the
Crossing V short , [O III] is dominant.Within the Crossing we see wide variations in S short , [O III] / S long , [O III] and note that the surfacebrightness variations suggest that the EUV ra-diation is being split between the S short , [O III] and S long , [O III] emitting regions. • V short , [O III] is essentially constant across theNE-SW sweep. Taken alone this would indicatethat it always arises from the NIL. However,the surface brightness of V short , [O III] leaps in the Crossing-Supergroup and remains high intothe
SW-Supergroup . This behavior remainsunexplained. • Within the peak of the rise upon which the
Crossing is centered there is a highly tilted lowionization region facing the north. • The
Orion-S Cloud produces changes toits SW, which is the direction of the hot gasgiving rise to X-ray emission. However, there isno obvious link to the Outer Shell that coversthe near side of the Extended Orion Nebula. • There is a pressing need to refine the NILmodeling approach that we presented in Paper-I and to apply it to various positions along theNE-SW sweep.ACKNOWLEDGEMENTSThe observational data were obtained fromobservations with the NASA/ESA HubbleSpace Telescope, obtained at the Space Tele-scope Science Institute (GO 12543), which isoperated by the Association of Universities forResearch in Astronomy, Inc., under NASA Con-tract No. NAS 5-26555; the Kitt Peak NationalObservatory and the Cerro Tololo Interameri-can Observatory operated by the Associationof Universities for Research in Astronomy, Inc.,under cooperative agreement with the NationalScience Foundation; and the San Pedro M´artirObservatory operated by the Universidad Na-cional Aut´onoma de M´exico. We have madeextensive use of the SIMBAD data base, oper-ated at CDS, Strasbourg, France and its mir-ror site at Harvard University and NASA’s As-trophysics Data System Bibliographic Services.GJF acknowledges support by NSF (1816537,1910687), NASA (ATP 17-ATP17-0141), andSTScI (HST-AR- 15018).9 -2024135640200 -30 -40
Distance South of 0 C (arcseconds) - -20402010 -30 -40 Distance South of 0 C (arcseconds) - Panel A Panel B
Figure 7.
This figure is similar to Figure 5 except that it shows data from a profile at an RA displacement38 (cid:48)(cid:48) west of θ Ori C passing through the center of the
Ledge feature in the center of the Crossing. Thesymbols mean the same as in Figure 5 but the red ellipses indicate S scat , [O III] /( S short , [O III] + S long , [O III] ). InPanel A Velocities and Signals are shown. Weak but clear V blue , [O III] components are also shown in theircorrect displacements but at velocities 10 km s − greater than measured. In Panel B scattered light ratiosand velocity differences are shown. APPENDIX A. A PROFILE ACROSS THE LEDGE FEATURE AT THE CENTER OF THE CROSSINGAt the center of the 38 (cid:48)(cid:48) west profile lies a narrow E-W feature within the larger feature previouslycalled the West-jet. It does not have any measured radial and tangential velocities ((O’Dell et al.2015), Paper-IV), thus arguing that the name originally assigned to it (the West-jet) is misleading.Thesmall V long , [N II] peak there and the large increase in S long , [N II] mean that it is a small ionized regionseen nearly edge on, therefore a designation as the Ledge is more descriptive. The higher side of the
Ledge lies to the south, otherwise there would be a shadowed zone in that direction.Figure 7 shows the results along the RA38 profile. Proceeding south in the upper portion of PanelA we see that there are dips in S long , [N II] and S long , [O III] along the Dark Arc, with the minimum of0 S long , [O III] being slightly further north, at which point S short , [O III] begins to increase. The large risein S long , [N II] but continuation of a decreasing trend in S long , [O III] south of the Dark Arc is consistentwith the much higher spatial resolution images in Figures 3 and 4. This indicates that the [O III ]emitting layer is thicker that the size of the Dark Arc in the plane of the sky. S long , [O III] decreasessouth of the Ledge while S short , [O III] remains strong, indicating that the S short , [O III] component hasbecome dominant.In the lower portion of Panel A one sees that the V long , [N II] component slowly increases until the Ledge is reached. This can be interpreted as an increasing tilt of the ionized layer removing more ofthe photoevaporation flow velocity and after the peak velocity at the
Ledge , the drop in V long , [N II] and V long , [O III] indicate the south side of the Ledge is flatter than the north. The disappearance ofthe V short , [N II] component south of the Ledge indicates that the gas that had been producing it isno longer there, probably by having become more ionized.In the lower portion of Panel B we see that V scat , [NII] - V long , [N II] is about the same ( 16 ± − ) asfound in Table 1 for the Crossing and the slight shift to a larger value south of the
Ledge is consistentwith this region being flatter. In [O
III ] the velocity is again similar to the
Crossing value in Table 1(20 ± − ) but it is unclear why the difference decreases to the south, although this may be dueto the V long , [O III] feature becoming weak there. Similarly, the high values in S scat , [O III] / S short , [O III] (upper portion of Panel B) occur in the north, where S short , [O III] is weak and becomes appropriatelylow in the south where S short , [O III] is strong. This means that even in [O III ] the V scat , [OIII] isbackscattered light from the nearest strong emitting layer. This is illustrated by the series of data-points giving the S scat , [O III] /( S short , [O III] + S long , [O III] ) values. The local rise in the Ledge samplesprobably means that our simple backscattering model breaks down there.In the upper panel of the upper portion of Panel B we see that the ratio S scat , [N II] / S long , [N II] isalways small, as one would expected from V long , [N II] radiation being backscattered from a nearbyPDR. Again, [O III ] is more complex. However, in the north region, where V long , [O III] is strongest, the S scat , [O III] / S long , [O III] values are small-again like backscattering. Then the ratio becomes impossiblylarge to the south, indicating that it is not the source of V scat , [OIII] ; but, this is the region where V long , [O III] has become weaker than V short , [O III] .In the lower portion of Panel A we show the locations of four very blue (about -20 km s − ) andweak (about 1% of S long , [O III] ) [O III ] velocity components.The averaged results differ by no statistically significant amounts from the lower resolution profileresults given under the
Crossing heading in Table 1. There are two surprising results here; that[N II ] strong Ledge shows no difference in its radial velocity from that of the surrounding nebulaand that the dark features do not contain velocity differences.REFERENCES
Abel, N. P., Ferland, G. J., O’Dell, C. R., &Troland, T. H. 2016, ApJ, 819, 136Abel, N. P., Ferland, G. J., O’Dell, C. R. 2019,ApJ, 881, 130Baldwin, J. A., Ferland, G. J., Martin, et al. 1991,ApJ, 374, 580Eisner, J. A., & Carpenter, J. M. 2006, ApJ, 641,1162 Garc´ıa-D´ıaz , Ma.-T., Henney, W. J., L´opez, J.A., & Doi, T. 2008, RMxAA, 44, 181Goicoechea, J. R., Teyssier, D., Etxaluze, M., etal. 2015, ApJ, 812, 75Goicoechea, J. R., Pety, J., Quadrado, S., et al.2016, Nature, 537, 207G¨udel, M., Briggs, K. R., Montmerle, T., et al.2008, Science, 319, 3091