Observations of Diffuse Ultraviolet Emission from Draco
N. V. Sujatha, Jayant Murthy, Rahul Suresh, Richard Conn Henry, Luciana Bianchi
aa r X i v : . [ a s t r o - ph . GA ] S e p GALEX
OBSERVATIONS OF DIFFUSE ULTRAVIOLETEMISSION FROM DRACO
N. V. SUJATHA, JAYANT MURTHY
Indian Institute of Astrophysics, Koramangala, Bangalore - 560 034, India [email protected], [email protected]
RAHUL SURESH
National Institute of Technology Karnataka, Surathkal, Managalore - 575 025, India [email protected]
RICHARD CONN HENRYandLUCIANA BIANCHI
Department of Physics and Astronomy, The Johns Hopkins University,Baltimore, MD 21218, USA [email protected], [email protected]
ABSTRACT
We have studied small scale (2 ′ ) spatial variation of the diffuse UV radiationusing a set of 11 GALEX deep observations in the constellation of Draco. Wefind a good correlation between the observed UV background and the IR 100 µ mflux, indicating that the dominant contributor of the diffuse background in thefield is the scattered starlight from the interstellar dust grains. We also findstrong evidence of additional emission in the FUV band which is absent in theNUV band. This is most likely due to Lyman band emission from molecularhydrogen in a ridge of dust running through the field and to line emissions fromspecies such as C IV (1550 ˚A) and Si II (1533 ˚A) in the rest of the field. Astrong correlation exists between the FUV/NUV ratio and the FUV intensity inthe excess emission regions in the FUV band irrespective of the optical depth ofthe region. The optical depth increases more rapidly in the UV than the IR andwe find that the UV/IR ratio drops off exponentially with increasing IR due tosaturation effects in the UV. Using the positional details of Spitzer extragalactic 2 –objects, we find that the contribution of extragalactic light in the diffuse NUVbackground is 49 ±
13 photons cm − sr − s − ˚A − and is 30 ±
10 photons cm − sr − s − ˚A − in the FUV band. Subject headings: dust, extinction - scattering - ultraviolet: ISM
1. INTRODUCTION
Studies of the diffuse ultraviolet (UV) sky have been an important part of interstellardust studies over the last four decades (Bowyer 1991; Henry 1991; Murthy 2009) but werelimited by the difficulty of observing faint diffuse sources near the limit of the instrumentalsensitivity. It has been generally agreed that the low and mid-latitude diffuse radiation isdominated by the scattering of starlight by interstellar dust but with a baseline at highgalactic latitudes, which was variously attributed to either high latitude dust (Bowyer 1991)or to an extragalactic source (Henry 2002).Just as the
Infrared Astronomy Satellite ( IRAS ) revolutionized the study of the dif-fuse infrared (IR) emission (Low et al. 1984), data from the
Galaxy Evolution Explorer ( GALEX ) have the potential to change our view of the diffuse UV sky. We have begunan ambitious effort to map the diffuse background in all
GALEX deep observations (expo-sure time ≥ ≤ τ ≤ µ m emission increasing by a factor of 2.In this work, we examine a set of observations of a region in Draco, where the opticaldepth is much lower ( τ < ◦ - 37 ◦ ) but is about 60 degrees away at a longitude of about 88 ◦ . The data arefrom the GALEX
Deep Imaging Survey (DIS), a few of them overlapping with the
Spitzer
First Look survey . Combining these two studies we present here the nature of diffuse UVradiation from low optical depth to high optical depth region. See http://ssc.spitzer.caltech.edu/fls/
2. OBSERVATION AND DATA ANALYSIS
The
GALEX spacecraft was launched in 2003 under NASA’s Small Explorer (SMEX)program with a primary science objective of observing star formation in galaxies at lowredshifts (Martin et al. 2005). Light from the sky is collected through a single 50 cm telescopeand separated into two bands (far ultraviolet (FUV): 1350 - 1750 ˚A; near ultraviolet (NUV):1750 - 2850 ˚A) using a dichroic mirror. Independent low noise delay-line detectors recordevery photon in each band with an overall effective spatial resolution of 5 – 7 ′′ in the skyover a 1.25 ◦ field. The data products from the mission include, amongst other files, FlexibleImage Transport System (FITS) (Wells et al. 1981) images of the FUV and NUV fields anda list of point sources in each field. A complete description of the data processing, thecalibration and the data products may be found in Morrissey et al. (2007).This work follows our study (Sujatha et al. 2009) on GALEX observations of diffuseemission in Region I and focuses on a set of 11 observations covering an area about 10square degrees in the constellation of Draco, with cumulative exposure times of 3,000 to50,000 seconds (Table 1). These observations were taken by the
GALEX team as part ofa program to map the
Space Infrared Telescope Facility ( SIRTF : now the
Spitzer SpaceTelescope ) First Light locations - hence the target name of “
SIRTFFL ”. This region (Fig. 1)contains the high velocity cloud (HVC) Complex C (Miville-Deschenes et al. 2005) at adistance of more than 800 pc but also, more relevant to our data, the nearby (60 pc) cloudLVC 88+36-2 (Lilienthal et al. 1991), seen as a ridge in the IR emission. This cloud wasfirst discovered to cast a shadow in the X-ray background (Burrows & Meadenhall 1991).Because of the then upcoming Spitzer observations, Lockman & Condon (2005) mapped theregion in the 21 cm line of H I , finding several components (Table 2). This wealth of detailhas proven invaluable to our understanding of the UV observations.Each observation is comprised of a number of visits spread over a period of months,or even years, all of which are coadded by the standard GALEX pipeline (Morrissey et al.2007) to produce a single image in each of the two bands. Point sources in each imagewere extracted by the
GALEX team using a standard point source extractor (SExtractor -Bertin & Arnouts (1996)) and a merged point source catalog was created. We note here thatthe exposure time in the FUV detector was often significantly less than that in the NUVbecause of intermittent power supply problems. Our processing uses the FITS image files andthe merged point source catalog from the
GALEX pipeline. These image files have been fullycalibrated and flat fielded but not background subtracted. Although the
GALEX programdoes provide files containing the background in each observation, these were made by fittinga multi-dimensional surface to the image and therefore show structure related to the pinningpoints of the surface. While perhaps adequate for their intended purpose of subtracting the 4 –background from point sources in the field, they introduce large scale artifacts which makethem unsuitable for the study of the diffuse radiation field.Following Sujatha et al. (2009), we created our own background files for each observationby blanking out the point sources in the merged
GALEX point source catalog and binningthe observation into 2 ′ pixels (80 × GALEX pixels). These images form the starting pointof our analysis. Because of edge effects, we only used the central 1.15 ◦ of the 1.25 ◦ field ofview for the analysis, rejecting about 20% of the total number of pixels. These backgroundfiles are comprised of the foreground emission (instrumental dark count, airglow and zodiacallight) and the astrophysical signal (atomic and molecular emission, dust scattered starlightand any extragalactic contribution).
3. FOREGROUND EMISSION
A large field of view imager such as
GALEX has distinct advantages in observations ofthe diffuse background in that stars can be easily identified and rejected. However, withoutspectra, we can only infer the contribution of the different components of the diffuse radiationfield. Instrumental dark count is negligible, contributing less than 5 photons cm − sr − s − ˚A − in either band (Morrissey et al. 2007) but airglow, primarily due to the O I lines at1356 ˚A and 2471 ˚A, is expected to contribute about 200 photons cm − sr − s − ˚A − to eitherband (Boffi et al. 2007). Although we cannot extract the airglow contribution directly, wehave been able to use the Telemetered Event Counter (TEC) of the spacecraft to track thetotal number of counts as a function of orbital time. Assuming the time dependent partof TEC present in both the GALEX bands as the total foreground emissions, a baselinehas been subtracted from each visit so that the count rate is zero at local midnight. Theremaining variable component of airglow (AG v ) is well fit with a quadratic as a function oftime from local midnight (Fig. 2).In addition, we have found that the baseline levels at local midnight are strongly cor-related with the 10.7 cm solar flux (Fig. 3) which is used as a proxy for solar-terrestrialinteractions (Chatterjee & Das 1995). Each observation is comprised of several visits, eachof which may have a different airglow and zodiacal light contribution. We have estimatedand subtracted the zodiacal light from each visit’s baseline level and found the y-interceptfor each observation, corresponding to stars in the field and the diffuse cosmic background.These values have been subtracted from the individual baseline levels and the resultant val-ues, assumed as the constant airglow (AG c ) in each visit, are plotted in Fig. 3. Combining c + AG v ) allow us to calculate the total airglow (AG) as a functionof local time (t, hours from local midnight) and solar 10.7 cm flux (SF, in 10 Jy) with thefollowing equations, with an uncertainty of about 50 photons cm − sr − s − ˚A − . F U V AG = 3 . SF + 24 . t + 11 . t (1) N U V AG = 3 . SF + 16 . t + 5 . t (2)This emission is consistent with an origin of the airglow in solar photons resonantly scatteredfrom geocoronal oxygen atoms (L. J. Paxton, personal communication). It should, however,be noted that Brune et al. (1978) observed a much lower level of airglow emission with ascaled GALEX contribution of about 50 photons cm − sr − s − ˚A − from their rocket-bornespectroscopic observation. It is possible that some part of what we have euphemisticallycalled “airglow”, may be due to some other contributor (Henry et al. 2010).The remaining foreground contributor, zodiacal light, is important only in the NUVband because of the rapidly fading solar spectrum at wavelengths shorter than 2000 ˚A.Although there is no UV map of the zodiacal light, we have used the distribution in thevisible with grey scattering (Leinert et al. 1998) to predict the zodiacal light in each visit .The foreground emission (Table 3), ranges from 20% to 50% of the total emission with anuncertainty of about 30 photons cm − sr − s − ˚A − , estimated using the spatial overlapbetween different observations. It should be emphasized that the foreground emission affectsonly the level of the offset and will not affect the spatial variability of the diffuse radiationfield. More interesting is the scatter in the data. For a photon counting instrument such as
GALEX , the instrumental scatter will be either due to photon noise or to errors in the flatfielding (calibration) of the instrument. We have empirically derived the instrumental scatterby dividing each observation into two sets of visits, which may well be separated by severalmonths. There is excellent agreement between this and the intrinsic photon noise (Fig. 4),confirming that the errors are dominated by poissonian rather than instrumental effects.As an independent test, we also took the overlap regions between different observationsand calculated the scatter between them. Although the scatter for the overlap regions issomewhat higher than the calculated values, this is due to the many fewer points in theoverlap regions and their location near the edge of the detector. We note here that all our calculator at http://tauvex.iiap.res.in/htmls/tools/zodicalc/
4. RESULTS AND DISCUSSION
The FUV and NUV images of the
Spitzer “First Look” field obtained after subtractionof the foreground emission are shown in Fig. 5 at a spatial resolution of 2 ′ . The UV imagesof Fig. 5 may be compared with the IR 100 µ m map (Fig. 1). There are several possi-ble contributors to the astrophysical UV emission, a significant one being, dust-scatteredstarlight which contributes to both the FUV and the NUV bands. This is reflected in thegood correlation between the FUV and NUV bands (Fig. 6) and between the two UV bandsand the IR 100 µ m fluxes (Fig. 7). This is in contrast with the essentially flat UV-IR curvesobtained by Sujatha et al. (2009) in Region I. The IR emission is due to thermal radiationfrom an optically thin layer of dust, as the cross-section of the grains is low in the IR. On theother hand, the cross-section of the grains is much higher in the UV and the optical depthtransitions from being optically thin in these Draco observations to being optically thick inRegion I.In Fig. 8, we have plotted the ratio between the UV bands and the IR to understandthe nature of diffuse UV emission with optical depth. There is a clear trend visible fromthe low optical depth Draco region to the high optical depth (in the UV) Region I with anempirical formula of F UV F IR = 415 e − . × F IR . It is interesting to note that the F UV /F IR ratio in our GALEX data follows a continuous curvevery similar to that found by Murthy et al. (2001) in Orion using data from the MidcourseSpace Experiment (MSX) even though the UV and the IR fluxes in Orion were each greaterby a factor of about 200, reflecting the intense radiation field there. However, quite differentvalues are cited in the literature for other regions with ratios ranging from near -50 to almost260 photons cm − sr − s − ˚A − (MJy sr − ) − with little dependence on the IR (Sasseen et al.(1995); Sasseen & Deharveng (1996)). It is likely that these relations are only apparent whenobserved at a high enough spatial resolution; the MSX data were at a resolution of 20 ′′ andour data are at a resolution of 2 ′ , while the other observations are at resolutions of 0.5 ◦ orworse. Since, both the IR and the UV vary on smaller scales, the measured F UV /F IR ratiomay not be a reliable estimator of the true ratio. In fact, Sasseen & Deharveng (1996) founda F UV /F IR ratio of 255 photons cm − sr − s − ˚A − (MJy sr − ) − for the slope using all theirdata, higher than any of the individual data sets. In general, we conclude that the F UV /F IR ratio in any region strongly depends on the local effects such as the proximity of hot stars 7 –near the scattering dust and the optical depth.Readily apparent in both Fig. 7 and Fig. 8 is the ridge of dust (LVC 88+36-2) run-ning through our field, where the FUV emission is proportionately greater than the NUV.Indeed, this reflects a general increase in the FUV/NUV ratio with the FUV surface bright-ness (Fig. 9) seen here and in Region I. The most likely explanation for this is that thereis an additional component in the FUV band which is not seen in the NUV. Sujatha et al.(2009) suggested that this is fluorescent Lyman band (1400 - 1700 ˚A) emission of molecu-lar hydrogen, a reasonable assumption in Region I where Martin et al. (1990) had alreadyobserved widespread H fluorescent emission.Assuming that the FUV/NUV ratio for dust scattering alone is constant with a value of0.8 (Fig. 9), we can estimate the level of excess emission in the field. The average error in thisratio, due to the scatter in the data, is estimated to be ± I ) (Fig. 10), there is a strong correlationin the ridge (LVC 88+36-2), where the excess emission is likely due to H fluorescence.We obtain a reasonable fit to the data following Martin et al. (1990) and calculate theemission assuming a plane-parallel slab with constant density (Fig. 11). Park et al. (2009)have observed atomic emission lines of both Si II (1533 ˚A) and C IV (1550 ˚A) around thenearby Draco molecular cloud which would effectively contribute about 50 photons cm − sr − s − ˚A − in the FUV band and it may be that some part of the emission outside theridge, where there is no correlation with H I , may be due to atomic lines instead. We have applied our standard three-parameter model of interstellar dust scattering(Sujatha et al. 2005) to the continuum dust scattered light in Draco. This model has beendescribed fully by Sujatha et al. (2005) and uses Kurucz models (Kurucz 1992) for the starsin the Hipparcos catalog (Perryman et al. 1997) to calculate the interstellar radiation field(Sujatha et al. 2004). This radiation is then scattered from dust in the line of sight, takinginto account self-extinction. The scattering function is from Henyey & Greenstein (1941)and depends only on the albedo ( a ) and the phase function asymmetry factor ( g =
6) grains in the UV, in agreement with the predictions for a mixture ofspherical carbonaceous and silicate grains (Draine 2003). On account of the uncertainity ofextragalactic contribution (EGL) in the data, we have considered it as a variable parameterin the model. A full treatment of the problem would take into account multiple scatteringand clumpiness in the ISM (see, for example, Gordon 2004) but, because the optical depth is 8 –low ( τ < I in the regionshow that the diffuse emission is correlated maximum with the LVC component of H I , whichis the local cloud at 60 pc, and the addition of any other components such as IVC or HVCto LVC reduces the correlation. The details (correlation coefficient, r ) are given in Table 4.Hence for these observations, we have assumed scattering from the local clouds at a distanceof 60 pc; very little contribution to the diffuse light comes from the more distant clouds.With these assumptions, we have placed 1 σ limits of 0.45 ± a ),0.56 ± g and 58 ±
18 photons cm − sr − s − ˚A − on the EGL in the NUV bandwith a reduced χ of 1.32. If we use the empirical ratio of 0.8 for the FUV/NUV ratio of thedust, the best fit NUV values translate into an albedo of 0.32 ± g of 0.51 ± σ error bar of about 40 photons cm − sr − s − ˚A − inthe model fit to the data compared to about 20 photons cm − sr − s − ˚A − from the photonnoise, probably reflecting the incompleteness of the model. Spitzer Space Telescope (Werner et al. 2004) made its 67 hours First Look Survey (FLS)near Draco in 2003 in order to characterize the starlight from distant galaxies in the regionin mid-infrared, using
Infrared Array Camera ( IRAC ; Fazio et al. (2004)) and the
MultibandImaging Photometer for Spitzer ( MIPS ; Rieke et al. (2004)). The
IRAC survey covered anarea of 3.8 deg centered on R.A. 17 h m s , Dec. +59 ◦ ′ ′′ at wavelengths 3.6, 4.5, 5.8and 8.0 micron, with flux density limits of 20, 25, 100 and 100 µ Jy (Lacy et al. 2005). Thisinstrument produced a band merged catalog of the survey containing 103,193 objects witha positional accuracy of about 0.25 ′′ for high signal-to-noise objects and about 1 ′′ at theflux density limits. The overlap area of IRAC survey is about 38% of the total
GALEX observed area in Draco. We have used this important positional details of
IRAC catalogedsources to estimate the observed EGL contribution in our diffuse maps. Note that the onlyexpected contribution of EGL in our diffuse maps are from the undetected faint galaxies bySExtractor, since we have removed all the detected sources using the
GALEX catalog fromeach of our field.We find that some
IRAC objects are showing enhancement in the UV intensities fromtheir local background, measured from 2 ′ bin. In Fig. 12, the average UV intensities of theseobjects measured using a diameter of 9 ′′ (6 pixels) are plotted against the corresponding local 9 –background. The UV intensities and the corresponding AB magnitudes of these sources areestimated after subtracting the local background. The total number of such objects detectedin the NUV field is 18,989 in the magnitude range 20.0 – 24.0. The number counts of theseobjects (Table 5) are shown in Fig.13. By integrating along the curve, we derived the EGLcontribution in the NUV map as 49 ±
13 photons cm − sr − s − ˚A − . The errorbar includeboth the uncertainities in the magnitude and the area overlapped by IRAC in the field. It isinteresting to note that this amount is in good agreement with the extracted value of EGLfrom the model. We have also found that an accurate estimation of number counts in theFUV band is difficult due to the excess emission in the field and hence we restricted ouranalysis to the NUV band. However, assuming an average ratio of 0.43 between the FUVand NUV sources derived from Xu et al. (2005) and Hammer et al. (2010) in the magnituderange 20.0 – 24.0, we estimated the EGL contribution as 30 ±
10 photons cm − sr − s − ˚A − in the FUV map from a total of 8165 objects.
5. CONCLUSIONS
We have completed an analysis of two sets of deep
GALEX observations: earlier nearthe Sandage reflection nebulosity (Region I) towards MBM 30 (Sujatha et al. 2009) and nownear the Draco Nebula. In both cases, we have found a good correlation between the signal inthe FUV band (1350 – 1750 ˚A) and the NUV band (1750 – 2850 ˚A) but with an additionalcomponent in the FUV which is not seen in the NUV. This was identified as fluorescentemission from the Lyman band of molecular hydrogen in Region I and in the nearby cloudLVC 88+36-2 in these observations, where the ratio was correlated with the H I columndensity. However, there was excess emission throughout the Draco region which was notcorrelated (or anti-correlated) with N(H I ) and this may be due either to H emission orto line emission from hot gas. While GALEX observations are invaluable in probing thediffuse background at unprecedented sensitivity and spatial resolution, spectra will still benecessary to fully understand the observations. However, we strongly recommend that theFUV/NUV ratio can be used to identify the atomic and molecular emission regions in the
GALEX survey fields all over the sky.The scattered light from the interstellar dust is consistent with an optically thin layer inthe Draco region transitioning to optically thick in the earlier Region I results, although thethermal emission in the infrared is optically thin in both cases. The F UV /F IR ratio followsan exponential curve across both regions, as would be expected for optically thick media.Interestingly, the F UV /F IR ratio in Orion follows exactly the same curve even though boththe UV and IR values are higher by a factor of almost 200 due to the intense radiation field. 10 –In general, we find that the F UV /F IR ratio strongly depends on the local effects such as theproximity of hot stars to the scattering medium and its optical depth.We have determined optical constants a (0.45 ± g (0.56 ± a (0.32 ± g (0.51 ± ±
18 photons cm − sr − s − ˚A − in the NUV band using our model, which is in good agreement with the derivedlimit of 49 ±
13 photons cm − sr − s − ˚A − for the band using the Spitzer
FLS sources.This gives strong evidence that most of the diffuse background derived from the
GALEX observations have a Galactic origin specifically at galactic latitudes, | b | < ◦ .We have begun a massive program to look at the small scale structure of diffuse back-ground in all GALEX data of greater than 5000 seconds. These include data throughoutthe sky and sample a variety of different environments, although avoiding bright UV regionssuch as the Coalsack or Orion. In parallel, we are developing more sophisticated models tobetter match the high quality data obtained here. We believe the
GALEX data will allow usto place the study of the diffuse UV radiation on the same level as
IRAS did for the infraredcirrus.We thank our anonymous referee for the helpful comments and suggestions which havesignificantly improved this paper. This research is based on data from the NASA’s
GALEX program.
GALEX is operated for NASA by the California Institute of Technology underNASA contract NAS5-98034. We have also made use of NASA’s Astrophysics Data Systemand the SIMBAD database operated at CDS, Strasbourg, France. We acknowledge the useof NASA’s SkyView facility (http://skyview.gsfc.nasa.gov) located at NASA Goddard SpaceFlight Center.NVS is supported by a DST Young Scientist award. Support for RCH was provided byNASA
GALEX grant NNGO5GF19G to the Johns Hopkins University.As always, Patrick Morrissey has greatly helped with our understanding of the GALEXdata.
Facilities: GALEX . 11 –
REFERENCES
Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393Boffi, F. R., et al. 2007, ACS Instrumental Handbook, Version 8.0 (Baltimore:STScI)Bowyer, S. 1991, ARA&A, 29, 59Brune, W. H., Feldman, P. D., Anderson, R. C., Fastie, W. G., & Henry, R. C. 1978, Geophy.Res. Letters, 5, 383Burrows, D. N., & Meadenhall, J. A. 1991, Nature, 351, 629Chatterjee, T. N., & Das, T. K. 1995, MNRAS, 274, 858Draine, B. T. 2003, ARA&A, 41, 241Fazio, G. G., et al. 2004, ApJS, 154, 10Gordon, K. D. 2004, Astrophysics of Dust, eds. A. N. Witt, G. C. Clayton & B. T. Draine,ASP Conf. ProceedingsHammer, D., Hornschemeier, A. E., Mobasher, B., Miller, N., Smith, R., Arnouts, S., Mil-liard, B., & Jenkins, L. 2010, ApJS, (in press)Henry, R. C. 1991, ARA&A, 29, 89Henry, R. C. 2002, ApJ, 570, 697Henry, R. C., Murthy, J., & Sujatha, N. V. 2010, AAS, 215, 319.05Henyey, L. C., & Greenstein, J. L. 1941, ApJ, 93, 70Kurucz, R. L. 1992, in The Stellar Populations of Galaxies, IAU Symp., 149, ed. B. Barbuy& A. Renzini (Dordrecht: Kluwer), 225Lacy, M., et al. 2005, ApJS, 161, 41Leinert, C., et al. 1998, A&AS, 127, 1Lilienthal, D., Wennmacher, A., Herbstmeier, U., & Mebold, U. 1991, A&A, 250, 150Lockman, F. J., & Condon, J. J. 2005, AJ, 129, 1968Low, F. J., et al. 1984, ApJ, 278, L19 12 –Magnani, L., Blitz, L., & Mundy, L. 1985, ApJ, 295, 402Martin, C., Hurwitz, M., & Bowyer, S. 1990, ApJ, 354, 220Martin, D. C., et al. 2005, ApJL, 619, L1Miville-Deschenes, M. A., Boulanger, F., Reach, W. T., & Crespo, A. N. 2005, ApJ, 631,L57Morrissey, P., et al. 2007, ApJS, 173, 682Murthy, J. 2009, Ap&SS, 320, 21Murthy, J., Henry, R. C., Paxton, L. J., & Price, S. D. 2001, BASI, 29, 563Park, S. J., Min, K. W., Seon, K. I., Han, W., Lee, D. H., Edelstein, J., Korpela, E., &Sankrit, R. 2009, ApJ, 700, 155Perryman, M. A. C., et al. 1997, A&A, 323, 49Rieke, G. H., et al. 2004, ApJS, 154, 25Sandage, A. 1976, AJ, 81, 954Sasseen, T. P., & Deharveng, J. M. 1996, ApJ, 469, 691Sasseen, T. P., Lampton, M., Bowyer, S., & Wu, X. 1995, ApJ, 447, 630Sujatha, N. V., Chakraborty, P., Murthy, J., & Henry, R. C. 2004, BASI, 32, 151Sujatha, N. V., Murthy, J., Karnataki, A., Henry, R. C., & Bianchi, L. 2009, ApJ, 692Sujatha, N. V., Shalima, P., Murthy, J., & Henry, R. C. 2005, ApJ, 633, 257Wells, D. C., Greisen, E. W., & Harten, R. H. 1981, A&AS, 44, 363Werner, M. W., et al. 2004, ApJS, 154, 1Xu, C. K., et al. 2005, ApJ, 619, L11
This preprint was prepared with the AAS L A TEX macros v5.2.
Table 1. Observation Log
Tile Name RA Dec l b
NUV Exposure FUV Exposure a Observation period NUV Visits FUV Visits a (deg) (deg) (deg) (deg) (sec) (sec) (yyyy/mm/dd)SIRTFFL-00 259.11 59.91 88.84 35.05 52917.15 52016.95 2003/07/03 – 2008/08/25 41 39SIRTFFL-01 260.41 59.34 88.08 34.44 26006.10 30922.1 2003/07/04 – 2004/07/26 20 24SIRTFFL-02 260.09 58.5 87.08 34.66 39037.05 26859.35 2003/08/18 – 2007/09/02 30 25SIRTFFL-03 258.33 58.86 87.61 35.55 39830.40 29570.9 2003/08/19 – 2007/09/01 30 28SIRTFFL-04 256.98 59.72 88.76 36.13 3874.45 3874.45 2004/05/01 – 2004/05/01 3 3SIRTFFL-05 260.68 60.7 89.71 34.2 5305 5305 2004/05/01 – 2004/05/03 4 4SIRTFFL-06 257.58 60.45 89.6 35.74 27658.75 2540.55 2005/07/25 – 2008/04/07 22 2SIRTFFL-07 260.54 60.81 89.85 34.26 34376.55 21276.1 2005/07/27 – 2007/09/02 25 20SIRTFFL-08 262.61 59.15 87.78 33.33 40639.6 22540 2005/06/19 – 2007/09/01 28 20SIRTFFL-09 257.2 59.72 88.74 36.02 15737.4 2755.45 2005/07/29 – 2008/04/07 12 2SIRTFFL-10 256.99 58.8 87.63 36.24 27383.75 10757.7 2005/07/25 – 2007/08/28 21 15Region I b a There are often fewer visits in the FUV because of intermittent failures in the FUV power supply. b Tile name: GI1-005007-J092810p702308.
14 –Table 2. H I Components in the FieldCloud l b
Peak N(H I ) V LSR (deg) (deg) (cm − ) (km s − )Ridge (LVC) 87.44 35.93 1.8 × -2IVC1 89.47 34.25 5.6 × -41IVC2 88.82 34.17 5.0 × -41IVC3 86.53 33.73 4.4 × -34IVC4 (Draco) 89.85 35.60 7.8 × -23HVC (Complex C) 89.15 35.20 6.9 × -190 15 –Table 3. Airglow and Zodiacal Contribution in Each FieldTile Name Average Airglow Zodiacal light Total Foreground Emission a FUV NUV NUV FUV NUV(photons cm − sr − s − ˚A − )SIRTFFL-00 391 394 367 396 766SIRTFFL-01 387 338 407 392 750SIRTFFL-02 332 314 373 337 692SIRTFFL-03 318 308 382 323 695SIRTFFL-04 320 378 342 325 725SIRTFFL-05 362 440 342 367 787SIRTFFL-06 306 304 358 311 667SIRTFFL-07 356 313 381 361 699SIRTFFL-08 327 304 365 332 674SIRTFFL-09 333 333 367 338 705SIRTFFL-10 368 369 365 373 739Region I b
349 355 440 354 800 a Includes 5 photons cm − sr − s − ˚A − dark count. b Tile name: GI1-005007-J092810p702308. 16 –Table 4. Correlation details of UV emission with different components of N(H I )Data LVC IVC HVC LVC+IVC LVC+IVC+HVCFUV Ridge 0.88 0 0 0.84 0.82FUV Total 0.75 0.3 -0.09 0.70 0.51NUV Ridge 0.63 0 0 0.63 0.63NUV Total 0.63 0.27 -0.05 0.63 0.52 17 –Table 5. Number Counts of Extragalactic Objects present in the NUV diffuse map of Spitzer fieldAB Mag N objects
Log (N objects /deg /mag)20.25 14 0.9420.75 130 1.9121.25 584 2.5621.75 1941 3.0822.25 3288 3.3122.75 4211 3.4223.25 4307 3.4323.75 3904 3.3924.25 3364 3.32 18 – Fig. 1.—
IRAS map of the region in galactic coordinates. The GALEX field of view of the 11DIS targets are overplotted as circles with diameter 1.25 ◦ and marked as 0 to 10. The brightarc extending through the fields 3 & 10 is the low velocity cloud (LVC 88+36-2) discussedin the text and the brightest feature on the left top is the Draco Nebula. 19 – −4 −3 −2 −1 0 1 2 3 4−1000100200300400500600 Hours from Local Midnight (t) F U V T E C ( ph c m − s r − s − A o − ) AG(t) = 24.5 t + 11.6 t −4 −3 −2 −1 0 1 2 3 4−50050100150200250300350 Hours from Local Midnight (t) NU V T E C ( ph c m − s r − s − A o − ) AG(t) = 16.1 t + 5.9 t Fig. 2.— Total count rate (TEC, in photons cm − sr − s − ˚A − ) in the FUV (top) and NUV(bottom) is plotted against the local time from midnight. A baseline has been subtractedfrom each visit so that the count rate is zero at local midnight. The solid line represents thebest fit curve to the data whose quadratic equation is given in the left top of the plot. 20 – Solar Flux at Earth (x 10 Jy) F U V T E C ( ph c m − s r − s − A o − ) SIRTFFL−00SIRTFFL−01SIRTFFL−02SIRTFFL−03SIRTFFL−04, 09SIRTFFL−05, 07SIRTFFL−06SIRTFFL−08SIRTFFL−10Region I r = 0.91
Solar Flux at Earth (x 10 Jy) NU V T E C ( ph c m − s r − s − A o − ) SIRTFFL−00SIRTFFL−01SIRTFFL−02SIRTFFL−03SIRTFFL−04, 09SIRTFFL−05, 07SIRTFFL−06SIRTFFL−08SIRTFFL−10Region Ir = 0.88
Fig. 3.— Minimum TEC level in each visit at the local midnight is plotted against the10.7 cm solar flux at the Earth for FUV (top) and NUV (bottom) channel. An offset wassubtracted from each observation (one set of visits). The strong correlation observed hereindicates that the variation in the TEC level within an observation is due to the solar activity. 21 – -2 sr -1 s -1 A % -1 )020406080 O b se r ve d S ca tt e r ( ph c m - s r - s - A % - ) . x * + FUV VisitsNUV VisitsFUV OverlapNUV Overlap
Fig. 4.— Comparison of observed scatter and intrinsic photon noise in the data. The symbols‘filled square’ and ‘x’ represent the scatter when a single observation is broken up into twosets of visits for each band, respectively, while the symbols ‘asterisk’ and ‘+’ represent thescatter in the regions of overlap between different observations. In general, the observedscatter is consistent with photon noise alone with the high points being due to the smallerarea of overlap. 22 –
91 90 89 88 87 86 Galactic longitude, l323334353637 G a l ac t i c l a t i t ud e , b
200 400 600 80010001200
91 90 89 88 87 86 Galactic longitude, l323334353637 G a l ac t i c l a t i t ud e , b
200 300 400 500 600 700 800
Fig. 5.— Diffuse FUV (top) & NUV (bottom) images (in photons cm − sr − s − ˚A − ) ofthe region derived from the central 1.2 ◦ field of view of each GALEX observation. Theforeground emission has already been subtracted from each band. 23 –
200 400 600 800 1000 1200FUV Surface Brightness (ph cm -2 sr -1 s -1 A % -1 )2004006008001000 NU V S u r f ace B r i gh t n ess ( ph c m - s r - s - A % - ) + Dracox Region I Fig. 6.— Correlation between FUV and NUV intensity. The blue points (‘x’) representRegion I and the black points ‘+’ represent the Draco region. Good correlation between theFUV and NUV bands indicating that the dominant contributor of the diffuse background inthe field is the scattered starlight from the interstellar dust grains. 24 – -1 )40060080010001200 F U V S u r f ace B r i gh t n ess ( ph c m - s r - s - A % - ) + . x RidgeOutside RidgeRegion I -1 )300400500600700800900 NU V S u r f ace B r i gh t n ess ( ph c m - s r - s - A % - ) + . x RidgeOutside RidgeRegion I
Fig. 7.— Correlation between
IRAS
100 micron intensity and diffuse FUV (top) and NUV(bottom) background radiation. In each plot, the blue points (‘+’) represent the ridge ofdust, the ‘dots’ represent the region outside the ridge and the ‘x’ points represent Region I.The background radiation is strongly correlated with IR in Draco region but is saturated inRegion I because of the high optical depth in the UV. 25 – -1 )0100200300400500 U V /I R R a t i o ( ph c m - s r - s - A % - ( M Jy s r - ) - ) xx ++ ** FUV/IR outside ridgeNUV/IR outside ridgeRidge FUV/IRRidge NUV/IRRegion I FUV/IRRegion I NUV/IR
Fig. 8.— UV/IR ratio (in photons cm − sr − s − ˚A − (MJy sr − ) − ) as a function of IR 100 µ m intensity. The ratio exponentially drops off with IR due to the rapid increase of opticaldepth in UV. 26 – -2 sr -1 s -1 A % -1 )0.60.81.01.21.41.61.8 F U V / NU V R a t i o + . x RidgeOutside RidgeRegion I Fig. 9.— Ratio between the UV bands (after subtracting the foreground emissions) is plottedagainst the FUV surface brightness. The increase in the ratio with FUV radiation indicatesthe presence of excess emission in the FUV band. 27 – LVC Cloud Column Density (cm -2 )0100200300400500600700 E xcess F U V E m i ss i on ( ph c m - s r - s - A % - ) x Outside Ridge+ Ridge Fig. 10.— Excess FUV emission in the observations is plotted against N(H I ) in the LVC(Lockman & Condon 2005). There is a strong correlation inside the dust ridge where theexcess emission is due to molecular hydrogen fluorescence but a poorer correlation outsidewhere the excess emission may be due to line emission from C IV or Si II . 28 – Emission (ph cm -2 sr -1 s -1 A % -1 )0100200300400500600 M od e l P r e d i c t i on ( ph c m - s r - s - A % - ) Error bar+ Outside Ridge (r=0.61)x Ridge (r=0.88)
Fig. 11.— Predicted levels of H emission with a formation rate (R) of 1 x 10 − cm − s − areplotted against the excess emission in the field. There is reasonable agreement everywherebut particularly in the nearby cloud LVC 88+36-2. 29 –
600 800 1000 1200 1400 1600Median FUV background around source (CU)50010001500200025003000 A ve r a g e F U V I n t e n s i t y a t s ou r ce po s i t i on ( CU ) A ve r a g e NU V I n t e n s i t y a t s ou r ce po s i t i on ( CU ) Fig. 12.— Comparision of average UV intensity (in continum unit; 1 CU = 1photons cm − sr − s − ˚A − ) from 9 ′′ bin and median background from 2 ′ bin centered at each IRAC objectposition in our diffuse maps. The enhancement in some of the
IRAC source position indicatethe presence of undetected faint galaxies by the SExtractor in our diffuse maps. 30 –
20 21 22 23 24NUV Magnitude01234
Log ( N / d e g / m a g ) Fig. 13.— Number counts of extragalactic objects present in the diffuse NUV map of the