Serendipity observations of far infrared cirrus emission in the Spitzer Infrared Nearby Galaxies Survey: Analysis of far-infrared correlations
Caroline Bot, George Helou, Francois Boulanger, Guilaine Lagache, Marc-Antoine Miville-Deschenes, Bruce Draine, Peter Martin
aa r X i v : . [ a s t r o - ph . GA ] J a n Serendipity observations of far infrared cirrus emission in theSpitzer Infrared Nearby Galaxies Survey:Analysis of far-infrared correlations ∗ Caroline Bot , and George Helou and Fran¸cois Boulanger and Guilaine Lagache andMarc-Antoine Miville-Deschenes and Bruce Draine and Peter Martin [email protected] ABSTRACT
We present an analysis of far-infrared dust emission from diffuse cirrus clouds.This study is based on serendipitous observations at 160 µ m at high galactic lati-tude with the Multiband Imaging Photometer (MIPS) onboard the Spitzer SpaceTelescope by the Spitzer Infrared Nearby Galaxies Survey (SINGS). These obser-vations are complemented with IRIS data at 100 and 60 µ m and constitute one ofthe most sensitive and unbiased samples of far infrared observations at small scaleof diffuse interstellar clouds. Outside regions dominated by the cosmic infraredbackground fluctuations, we observe a substantial scatter in the 160/100 colorsfrom cirrus emission. We compared the 160/100 color variations to 60/100 colorsin the same fields and find a trend of decreasing 60/100 with increasing 160/100.This trend can not be accounted for by current dust models by changing solelythe interstellar radiation field. It requires a significant change of dust propertiessuch as grain size distribution or emissivity or a mixing of clouds in differentphysical conditions along the line of sight. These variations are important as apotential confusing foreground for extragalactic studies. Subject headings:
ISM:clouds — infrared:ISM * This work is based on observations made with the
Spitzer Space Telescope , which is operated by the JetPropulsion Laboratory, California Institute of Technology, under a contract with NASA. California Institute of Technology, Pasadena CA 91125, USA Institut d’Astrophyisque Spatiale, 91405 Orsay, FRANCE Princeton University Observatory, Princeton, NJ08544, USA Canadian Institute for Theoretical Astrophysics, Toronto, Ontario, M5S 3H8, Canada Observatoire Astronomique de Strasbourg, 67000 Strasbourg, FRANCE
1. Introduction
The InfraRed Astronomical Satellite (IRAS) showed for the first time that extended in-frared emission was present at high galactic latitude, far from star forming regions (Low et al.1984). In these diffuse regions, clouds are optically thin to stellar radiation and the radi-ation field is relatively uniform which results in very limited variations of dust equilibriumtemperature (Boulanger et al. 1996; Arendt et al. 1998; Lagache et al. 1998; Schlegel et al.1998). These high latitude cirrus also show a tight correlation between their infraredemission (100 µ m to 1mm observed by DIRBE and FIRAS) and the HI column density(Boulanger et al. 1996) and the dust emission is well characterized with a constant dustemissivity per hydrogen atom ( τ /N H = 10 − ( λ/ − cm ) close to the value expectedfrom models of interstellar dust grains (Draine & Lee 1984). At shorter wavelength, thesmaller dust grains emission is characterized by a ratio of I µm /I µm ∼ . o ).Smaller scale analysis of infrared colors have been done on individual regions and showclear variations of dust properties. Laureijs et al. (1996) and Abergel et al. (1994) observeda decrease of I µm /I µm toward dense clouds. Bernard et al. (1999) studied the far infraredemission at the arcminute scale in the Polaris flare with IRAS, ISOPHOT and PRONAOS(200 to 600 µ m) in a region where extended emission from cirrus is detected as well as a denserstructure. The spectrum of the extended cirrus indicates a low dust temperature associatedwith a low 60/100 µ m ratio. This was also observed in the Polaris flare toward moderatelydense regions ( A V ∼
1) and in a denser filament in the Taurus complex (Cambr´esy et al.2001; Stepnik et al. 2003). It might be explained by the formation of large dust aggregatesthrough the adhesion of small dust particles onto the surface of larger grains, leading to achange of dust emissivity properties. In the dense regions the very small grains seem to havedisappeared almost completely. However, all these observations were restricted to individualregions, most of which are much denser than the diffuse local interstellar medium seen athigh galactic latitudes.By comparing near-infrared extinction and extinction deduced from far-infrared dustemission in the whole anticenter hemisphere, Cambr´esy et al. (2005) observed a discrepancybetween the two quantities in regions above 1 mag. This effect is also interpreted by achange of dust emissivity due to the presence of fluffy grains and the grain-grain coagulationscenario was therefore extended to larger regions. 3 –Kiss et al. (2006) analyzed the far-infrared emission properties in a large sample ofinterstellar clouds observed with ISOPHOT with respect to extinction in regions of the orderof 100 arcmin . They find variations of the far infrared dust emissivities in the coldest (12K 2. The data The Spitzer Infrared Nearby Galaxies Survey (SINGS Kennicutt et al. 2003) observedin imagery with IRAC (Fazio et al. 2004) and MIPS (Rieke et al. 2004) onboard Spitzer asample of 75 nearby galaxies. While the IRAC images only observed the galaxy itself, asignificant part of the MIPS observations (strips) encompass the surrounding sky. Sincethe SINGS observations were chosen to be at high galactic latitude to limit the galacticforeground contamination, the MIPS observations at 160 µ m provide a good opportunity tostudy the low surface brightness diffuse infrared emission from high galactic cirrus in a largenumber of fields at a resolution of ∼ µ m. The position of the fields on the skyare shown in Fig. 1 while their characteristics are summarized in Tab. 1.Although SINGS observations were also done at 70 µ m with MIPS, the regions observedare offset with respect to the galaxy targeted and only a small fraction of the 70 and 160 µ mobservations overlap outside the galaxy itself, making them inappropriate for our galactic cir-rus emission study. 24 µ m observations were also available, but they are dominated by pointsources emission as well as stronger zodiacal light. Once the point sources removed and theregions at low ecliptic latitude are discarded, the 24 µ m brightness have a low dynamic rangein each field and no meaningful correlation can be done with longer wavelength observations.This study was therefore restricted to the comparison of 60, 100 and 160 µ m brightnesses.The IRIS 25 µ m observations were however used together with longer wavelength in order toremove point sources (like galaxies) more efficiently in the observations. The observations at 160 µ m were reduced using the GeRT software on the raw MIPSobservations. Standard parameters were used for the reduction, but data where flashes ofthe internal source led to a significant number of saturated pixels which were removed. Theremoved data are most often positioned on the bright center of the galaxy. Other saturatedpixels removed from processing were due to cosmic ray hits. These saturated flashes when notremoved can bias the sensitivity of the diffuse extended emission. This step in the reductionmay be not appropriate for the photometry of the galaxy, but significantly reduces latents(stripes) in the outer regions we are interested in. Each region targeted was observed twice.Discrepant fluxes at the same position between the two observations are removed and thedata are combined into a mosaic for each region.The MIPS 160 µ m maps and IRIS 60 µ m are convolved to the IRIS 100 µ m resolutionassuming gaussian beams with FWHM of 4.0’, 4.3’ and 37” for IRIS 60, 100 µ m and MIPS160 µ m observations respectively.For each MIPS strip, the galaxy and other point sources are detected in the 25, 60and 100 µ m maps using the method described in Miville-Deschˆenes et al. (2002). Thesepoint sources at the IRIS resolution (but with the MIPS sampling) are then smoothed by agaussian kernel with a full width half maximum of 3 × http://ssc.spitzer.caltech.edu/mips/gert/index.html at a resolution of 4.3’ observed at 60, 100 and 160 µ m. Dueto uncertainties in the zodiacal light subtraction at 60 µ m that can dominate the flux at thelow surface brightnesses we sample, we limited the sample at 60 µ m to the 9 observations athigh ecliptic latitude ( | β | > o ).A constant brightness of 0.78 MJy/sr is removed from the IRIS 100 µ m maps to accountfor the cosmic infrared background (Lagache et al. 2000), i.e. the emission from the distantunresolved galaxies (called hereafter CIB). The exact level of CIB emission has not yet beenestablished at 60 µ m and the MIPS observations can have offsets in the calibration of thebrightness that are not physical. To overcome the uncertainties (physical or instrumental)on the zero levels in the different maps, we hereafter perform the analysis of the data throughthe use of correlations (c.f. § µ m respectively (Miville-Deschˆenes & Lagache 2005). At 160 µ m, we take a quadraticcombination of a constant sensitivity limit of 0.12 MJy/sr and a 2% uncertainty on thebrightness (due to the uncertainty on the calibration factor from instrumental units to surfacebrightness (Stansberry et al. 2007)). In low surface brightness regions, the variations of the infrared emission in the obser-vations can come from cirrus emission, fluctuations in the cosmic infrared background or Although there are 75 galaxies in the SINGS sample, some galaxies are in the same field of view:NGC3031 is in the same observation as M81 dwarf B and NGC5195 was observed simultaneously withNGC5194 the sensitivity of the observations is computed for a 16s integration time per pixel using the SENS-PETtool, http://ssc.spitzer.caltech.edu/tools/senspet/ and is divided by √ N where N = 49 is the number ofMIPS 160 µ m PSF inside an IRIS PSF at 100 µ m σ cirrus , ina region and depends on the size of the region (Miville-Deschˆenes et al. 2007). For each ofthe observed field of view, we computed the standard deviation at 100 µ m and the meanbrightness at 100 µ m (minus the average CIB contribution at this wavelength) and then plotthe σ – < B > relationship observed at 100 µ m in our sample. To model σ cirrus , we use therelationship derived by Miville-Deschˆenes et al. (2007) below 10 MJy/sr, for a maximumscale length of 50’ (dotted line in Fig 2). We observe that our observations are consistentwith the model, with a large scatter as in the original relationship. This dispersion is likelyenhanced due to the fact that our fields of view are elongated and the size of the regionmapped varies between field.The contribution from the CIB fluctuations can be describedby two terms: a Poisson noise that represents the galaxies distributed homogeneously withrespect to the resolution and a component with correlated spatial variations correspondingto the clustering of galaxies on large scales. The contribution from the clustering of infraredgalaxies is predicted by using the Lagache et al. (2003) model for galaxy evolution, with abias parameter from Lagache et al. (2007). The contribution from the Poisson noise to the σ observed at 100 µ m is taken to be that measured by Miville-Deschˆenes et al. (2002) sincewe used the same point source detection method. However, compared to their study, weremoved point sources applying the detection scheme at all wavelength (25, 60 and 100 µ m).This enables us to mask faint galaxies at 100 µ m more efficiently and the Poisson noise in ourmeasurements could be lower than their measurement. Because we want to select the fieldswith the least contribution from other sources (CIB) than cirrus to the observed variations,this choice is therefore conservative.Combining all contributions (represented in Fig. 2) to the observed variations, we deter-mine that a cut at 2.5 MJy/sr corresponds to σ cirrus /σ CIB = 1 . σ tot = p σ cirrus + σ CIB are less thatn 20% larger than from cirrus fluctuationsalone). The regions with a mean 100 µ m brightness above this threshold will therefore bedominated by variations of cirrus emission. In each field, we computed the mean brightnessat 100 µ m as well as the standard deviation at 60, 100 and 160 µ m (c.f. Tab.1). By keepingonly the fields above the 2.5 MJy/sr cut, the sub-sample we will study in this paper is com-posed of 15 fields with 100 and 160 µ m brightnesses, among which 9 can be studied as wellat 60 µ m ( | β | > o , see sec. 2.1). 7 –Stellar reddenings obtained from the analysis of the Sloan Digital Sky Survey dataenable us to put an upper limit of 1.2 mag on the extinction in these fields . This confirmsthat the variations in the infrared cirrus emission studied in each region comes from diffuseclouds according to the van Dishoeck & Black (1988) classification. 3. Results3.1. Cirrus emission at 160 and 100 µ m In each field of the selected sample, we plot the point to point correlation between thebrightnesses observed at 100 and 160 µ m (represented in Fig 3 and 4) and apply a linear fittaking into account the errors at both wavelengths. This enables us to obtain for each fielda slope corresponding to the ratio B /B unbiased by variations of the zero point level(residuals from the zodiacal light subtraction, absolute value of the CIB). The correlationcoefficient and the slope derived in each region are summarized in Tab. 2.Large scale observations of high galactic latitude emission of cirrus with COBE were wellcharacterized by a modified black body with a dust temperature of 17.5K and an emissivityindex proportionnal to ν (Boulanger et al. 1996). Using this law, we estimate the large scale160/100 color for cirrus to be of B /B = 2 . B and B are in agreement with the B /B = 2 . σ level with respect to the value of 2.0. The most extreme case is the field of NGC2976, witha fitted slope on the B versus B correlation that is 10 σ away from the 2.0 standardvalue).In Fig. 5, we compare the obtained ratios B /B to the mean surface brightness at100 µ m in each field (black points). We observe a large dispersion in the 160/100 colors thatcan not be explained by the error on the data or the fitting process. At 100 and 160 µ m, theinterstellar emission is dominated by the emission from big dust grains at thermal equilibriumwith the radiation field (D´esert et al. 1990) and the B /B ratio is therefore related tothat characteristic dust temperature. Taking a standard emissivity of dust per hydrogenatom in H i from Boulanger et al. (1996) and an emissivity index of 2, the 160/100 colorvariations we are probing can therefore be related for illustrative purposes to temperatures Using N ( H ) /A V (Bohlin et al. 1978) and B /N ( H I ) ≈ . × − M Jy/srcm , this upper limitimplies B < . M Jy/sr , which is fully consistent with the brightness observed in our sample N H = 3 × to 2 × cm − .We note that these variations are consistent on average with the large scale estimate (bluesolid line), confirming that the fields used in this study are sampling the cirrus emissionobserved on large scale.For a given grain size and composition, this characteristic temperature depends on thelocal radiation field strength and spectrum which depends on the presence and distance ofnearby heating sources and on the extinction. In the framework of this model, the pres-ence of large variations in the 160/100 µ m ratio observed in our sample would suggests thepresence of large variations in the heating of grains at small scales (variations by a factor of3 of the intensity of the incident radiation field). This can be surprising since at low FIRsurface brightness and at high latitude the interstellar radiation field might be expected tobe homogeneous. We looked at the far-infrared color temperature maps derived from DIRBEobservations by Lagache et al. (1998) and Schlegel et al. (1998). The regions we are studyingappear to be reasonably representative of the high latitude cirrus given the small numberstatistics. For the sightlines covered by our sample, the FIR color variations seen in theDIRBE data are compatible with the variations that we observe. Our study is indeed moresensitive than previous observations and therefore able to probe color variations smaller thanthe uncertainties in the previous studies.The shape of the optical spectrum heating the grains could also affect the far-infraredcolors: the radiation field could become gradually harder with position off the galactic plane(Mattila 1980). Using the cirrus model from Efstathiou & Rowan-Robinson (2003) withdifferent stellar populations heating the clouds, we checked that changes in the shape of theoptical radiation field is unlikely to affect the 160/100 and 60/100 colors of cirrus by morethan 20%.The dust equilibrium temperature depends however also on the structure of the grains.Grain aggregates for example cool more efficiently. The temperature variations observedin the diffuse medium could therefore be either due to variations of the intensity of theinterstellar radiation field or to changes in the grain structure.We compared our findings with different studies of far infrared emission from the lit-erature: the quiescent high galactic latitude clouds from del Burgo et al. (2003), the largesample from archival ISOPHOT data by Kiss et al. (2006), the two regions in a high latitudecirrus MCLD 123.5+24.9 observed by Bernard et al. (1999), and the quiescent filament inthe Taurus molecular complex from Stepnik et al. (2003). Because other observations wereobtained with different instruments, we have to interpolate the brightnesses at various farinfrared wavelengths to estimates at 100 and 160 µ m. To do so, we took the dust temper-atures determined in each study with a brightness at 100 or 200 µ m and used a modified 9 –black body law with a spectral index. The power index is either taken from the study itself(if it was computed) or is fixed to a standard value of 2. For each region, we also computethe mean 100 µ m brightness as observed by IRIS and subtract a mean CIB contribution asfor our observations. Despite large scatter, we observe a trend between < B > and the160/100 color that is consistent with the idea that denser regions are colder. However, theeffect of selection biases of these studies remains unclear. The comparison of our results withthat from the literature (Fig. 5) shows that previous studies in the far-infrared have beentargeting higher B /B and < B > , i.e. denser and colder clouds. Due to our bettersensitivity, our observations fill the gap at low B and low B /B .For the first time, we observe interstellar dust emission at low surface brightness in anunbiased way (in the observing strategy) with a high sensitivity. These observations showthat former studies on dust properties at FIR wavelength at small scale, have been biasedtoward colder and denser clouds. Our study shows that the 160/100 brightness ratios ofhigh galactic cirrus clouds at small scales are consistent on average with the observationson large scales. However, these 160/100 colors show a wide dispersion that could arise fromvariations in the heating of the clouds or from change of the dust grain structure. In orderto investigate further the origin of the 160/100 variations, we extend the comparison to the60 µ m data. µ m data To investigate the origin of the 160/100 color variations observed in the diffuse cirrus,we compare the 160 and 100 µ m data to the 60 µ m emission. The sample for this part ofthe study is however reduced to fields with an ecliptic latitude above 15 o in order to avoidartifacts due to the uncertainties in the zodiacal light subtraction in the IRIS data. For eachfield, we determine a 60/100 color by using the same correlation technique as above. Thecorrelations in each region are shown in Fig. 6 and the obtained B /B are summarizedin Tab. 2.Fig. 7 presents the B /B ratio obtained in each field with respect to the B /B ratio. Here again, large variations are observed in the 60/100 colors that can not be ex-plained by the uncertainties in the data or in the analysis. As for the 160/100 colors, the60/100 brightness ratios are consistent on average with the ”reference values” (the pinkcross and the blue star in the figure) obtained for high latitude emission on large scales(Boulanger & Perault 1988; Boulanger et al. 1996; Arendt et al. 1998). Furthermore, thereis a trend of decreasing B /B with increasing B /B . 10 –To try to interpret this trend, we used two models of the dust grain emission at differentinterstellar radiation fields: the Draine & Li (2007) model for the Milky Way and the”DUSTEM” model (updated model based on D´esert et al. (1990)). The models take intoaccount the shape of the IRAS and MIPS/Spitzer filters, the color corrections. For bothmodels, the abundances of different grain types are kept constant. The tracks obtained arecompared to the data in Fig. 7. The comparison shows that, if the variations of colors aredue to variations in heating of the grains, this would imply large changes in the interstellarradiation field at high galactic latitude (from U ≈ . σ deviant from the expected curve), but for B /B < . B /B than predicted by both models, while for B /B > . B /B values than expected from both models.Current dust models might be missing an additional dust grain type. Such an additionmight reproduce all color variations while changing the interstellar radiation field alone.Another way to interpret the observed trend is that the variations in the dust emissionspectrum reflect spatial changes in the grain properties – composition, structure or sizedistribution.The equilibrium temperatures of dust grains is expected to decrease for increasing grainsizes and small grains ( ≤ . µ m) undergo temperature excursions following single-photonheating that enhances the 60 µ m emission. Thus regions with fewer small grains may havelower 60/100 ratios. The observed trend between the 60/100 and 160/100 infrared colorscould be reproduced by changing the dust grains size distribution or composition. For exam-ple, enhancing the amount of small grains in regions with higher interstellar radiation field(i.e. higher temperatures) and reducing it at low equilibrium temperatures could reproducethe observed variations.In the same way, regions with enhanced populations of large grains may have increased160/100 ratios. In that case, reproducing the observations could be obtained with onlymodest variations in the starlight heating rate and shifts in the grain size distribution (fewersmall grains and increased sizes for the larger grains at low temperatures, more small grainsand smaller sizes for the big grains at higher dust temperatures).The 60/100 colors we observe therefore suggest changes of the dust properties (dust We took the model with a PAH fraction q P AH = 4 . 58% but we checked that this parameter does notinfluence significantly the results of this study 11 –size distribution and/or composition) from one region to the other. These changes arerelated to variations in the 160/100 brightness ratio. Whether the 160/100 color variationsrequire a change of the starlight intensity heating the clouds or result from the change ofdust properties alone is unclear. The interpretation of the color variations and of the trendbetween the 160/100 and 60/100 colors is discussed further in the next section. 4. Discussion Despite its rather constant color distribution on large scale, the far infrared emissionfrom diffuse cirrus at high galactic latitude is observed to host large color variations on smallscales. These variations seem related to each other (the 60/100 color decreases as the 160/100color increases). In this section, we will first check that these variations come indeed fromcirrus emission and are not related to the galaxies targeted with the observations. Second,we will discuss possible interpretations for these large color variations and the trend betweenthem. Because the MIPS observations were taken to observe nearby galaxies, it is legitimateto ask whether the infrared emission that we observe could be associated with these targets.In particular, H i observations have shown that gaseous disks can extend much farther thanthe optical diameters. Dust grains could be present in these outer parts of the galaxies andbias our measurements.To avoid this extended emission from the galaxies, we were careful in masking regionslarger than the detected emission (c.f. Sec. 2.1). Some of the galaxies in the observationsused in this study have been observed in H i observations through The HI Nearby GalaxiesSurvey (THINGS, Walter et al. 2005). We checked that the H i diameters reported for thesegalaxies are smaller than the region masked for our study. We are therefore confident thatthe variations observed in the infrared emission between fields do not come from the targetedgalaxies, but rather from diffuse cirrus emission. A possibility to interpret the variations of far-infrared colors at high galactic latitudeis that we are sampling clouds in different physical conditions and/or composition (different 12 –heating of the clouds, different dust size distribution, . . . ). The color changes would then bedue to mixing along the sightline of these different components.An unbiased survey of H absorption in high galactic latitude clouds by FUSE (Gillmon et al.2006) has been interpreted as showing that some clouds have been compressed. The dynam-ical history leading to this compression may involve shock waves or strong turbulence, whichcould also lead to changes in the grain size distribution by shattering in grain-grain colli-sions, possibly explaining the regions of higher than average 60/100 and lower than average160/100 colors.One tantalizing possibility, in terms of mixing, is the presence of Intermediate or highvelocity clouds (IVCs and HVCs) along the sightline. These cloud falling onto our galaxycould have very different dust properties (e.g. Miville-Deschˆenes et al. 2005) and would biasthe measured infrared emission from more local cirrus clouds. We checked for the presenceof intermediate velocity clouds in the LAB H i survey spectra (Kalberla et al. 2005) in thedirection of the fields in our sample. For about half of the sample, there is a intermediatevelocity component seen in H i in the sightline. Two sightlines also have a high velocitycomponent. However, no conclusive trend between the fraction of the H i in the IVCs and/orHVCs and the infrared colors could be seen. This may be due to the lack of resolution of theH i observations ( ∼ . o ), to the difficulty to disentangle IVCs and the Milky Way in someregions or to the small size of our sample. The grain size distribution is the result of processes such as sputtering, shattering andcoagulation, and sightline-to-sightline variations in the wavelength-dependence of optical andultraviolet extinction toward stars have already demonstrated regional variations in the grainproperties. Whether the observed variations in emission can be fully explained by variationsin the size-distribution alone, or whether other properties (e.g., composition or porosity) arealso involved is uncertain.In denser clouds, variations of infrared colors (Stepnik et al. 2003; Kiss et al. 2006) havebeen interpreted with a grain coagulation scenario, combining changes of the size distributionwith changes of the dust emissivity properties. The trend observed between the 60/100 and160/100 colors would be consistent with this idea. In this scenario, most of the 160/100 colorvariations would then be due to the change of emissivity of dust grains (due to changes oftheir structure), while the 60/100 colors would change with the incorporation or release ofsmall grains in large dust aggregates. We could be witnessing variations of dust properties 13 –due to variations in turbulent motions in the diffuse interstellar medium. Alternatively,dust grains in the diffuse medium could retain for some time the aggregate structure theyhad previously acquired in denser regions. So we could be seeing a sequence of regionscorresponding to increasing time since their release from high density environments.Such changes in the dust size distribution and structure of grains would imply relatedvariations of the UV-optical extinction curves at high galactic latitude. An extinction andreddening study of stars at high galactic latitude behind translucent clouds (Larson & Whittet2005) shows variations in the extinction curves obtained with respect to the average curvefor the diffuse interstellar medium. In particular, 48% of their sightlines have R V < . . 05. Such values are indicative of enhancedabundances of small grains, and these regions could have a higher that average 60/100 color.To test if the high 60/100 colors and low R V values are connected, we computed the 60/100colors from IRIS data in a similar fashion to this study for a ∼ × o region around eachsightline of the Larson & Whittet (2005) sample. We do not observe any correlation howeverbetween the R V and 60/100 color, nor between A V and the 60/100 color. Unfortunately, nolonger wavelength observations exist for these regions and it would be important to deter-mine the 160/100 colors in these regions as well and study their dependancy with extinctionproperties. This result is however a concern for the coagulation scenario as an interpretationof the 60/100 color variations we observe.Interpreting the trend observed between the 60/100 and 160/100 colors with a changeof dust optical properties and dust size distribution remains hypothetical without furtherobservations. In particular, H i observations of these diffuse sightlines, with a high resolution(at least similar to the IRIS one), will be needed to determine the emissivity of dust perhydrogen atom and test if variations are observed and correlated with the infrared colors.It is presently not possible to interpret further the far-infrared color variations in termsof physical condition changes, grain size distribution, grain properties, etc. A larger numberof observations of high latitude cirrus could help to probe the spatial variations. H i studiesof these regions at high resolution would also be crucial to probe possible variations of dustgrain emissivities, or to check whether the velocity structure of the cirrus correlates with the60/100/160 colors. Finally, extinction curves on sightlines where cirrus emission propertieshave been determined would be most useful to see whether extinction properties wouldcorrelate with the FIR colors. 14 – 5. Conclusion We performed an unbiased study of dust emission from high galactic latitude cirrus usingserendipitous Spitzer MIPS observations at 160 µ m from the SINGS survey, complemented byIRIS data at 60 and 100 µ m. After an appropriate post-reduction of the data and a removalof the targeted galaxy, a sub-sample is selected so that the variations of the cirrus emissiondominate over the CIB fluctuations in each field.We observe 160/100 colors in our fields that are consistent on average with large scalestudies. However, strong variations are also observed from field to field. This paper extendsformer studies on dust properties at high galactic latitude to more diffuse, fainter and warmerclouds. The 60/100 color is also observed to vary significantly in the sample and there isa trend of decreasing 60/100 with increasing 160/100 ratios. This trend is not completelyreproduced by current models taking only into account variations of the radiation fieldstrength and requires changes in the dust properties, composition or size distribution.The exact origin of these variations remains unknown, but the variations of the 60/100/160colors may reflect variations of the grains size distribution, of grain properties in addition toheating variations. However, we can not completely rule out the possibility that our fieldscontain emission from matter at different heights above the Galactic plane, the juxtapositionof multiple components in the fields could be affecting the infrared color estimates.All in all, we observe unexpected variations of far-infrared colors in the supposed ”ho-mogeneous” cirrus at high galactic latitudes. These variations are not yet understood andfurther studies will be needed to test their origin. In particular, studies on a larger area ofsky is needed to confirm the significance of these variations and their spatial distribution onthe sky could give new clues on their origin.These infrared color variations will most probably be linked with variations of the in-frared colors at longer wavelengths. They therefore represent an important point to studyfor the Planck and Herschel missions. Such longer wavelength observations will enable us todetermine precisely the temperature and spectral index of the dust, and their variations, inhigh latitude regions.We would like to thank our referee M. Rowan-Robinson for his useful and interestingcomments and A. Efstathiou for his help with their cirrus model. We are also grateful to L.Cambr´esy for useful discussions and work on upper limit for extinction in our fields. Facilities: Spitzer. 15 – REFERENCES Abergel, A., Boulanger, F., Delouis, J. M., Dudziak, G., & Steindling, S. 1996, A&A, 309,245Abergel, A., Boulanger, F., Mizuno, A., & Fukui, Y. 1994, ApJ, 423, L59Arendt, R. G., Odegard, N., Weiland, J. L., et al. 1998, ApJ, 508, 74Bernard, J. P., Abergel, A., Ristorcelli, I., et al. 1999, A&A, 347, 640Bohlin, R. C., Savage, B. D., & Drake, J. F. 1978, ApJ, 224, 132Boulanger, F., Abergel, A., Bernard, J.-P., et al. 1996, A&A, 312, 256Boulanger, F., Cox, P., & Jones, A. P. 2000, in Infrared Space Astronomy, Today andTomorrow, 251Boulanger, F. & Perault, M. 1988, ApJ, 330, 964Cambr´esy, L., Boulanger, F., Lagache, G., & Stepnik, B. 2001, A&A, 375, 999Cambr´esy, L., Jarrett, T. H., & Beichman, C. A. 2005, A&A, 435, 131del Burgo, C., Laureijs, R. J., ´Abrah´am, P., & Kiss, C. 2003, MNRAS, 346, 403D´esert, F.-X., Boulanger, F., & Puget, J. L. 1990, A&A, 237, 215Draine, B. T. & Lee, H. M. 1984, ApJ, 285, 89Draine, B. T. & Li, A. 2007, ApJ, 657, 810Efstathiou, A. & Rowan-Robinson, M. 2003, MNRAS, 343, 322Fazio, G. G., Hora, J. L., Allen, L. E., et al. 2004, ApJS, 154, 10Gillmon, K., Shull, J. M., Tumlinson, J., & Danforth, C. 2006, ApJ, 636, 891Kalberla, P. M. W., Burton, W. B., Hartmann, D., et al. 2005, A&A, 440, 775Kennicutt, Jr., R. C., Armus, L., Bendo, G., et al. 2003, PASP, 115, 928Kiss, C., ´Abrah´am, P., Laureijs, R. J., Mo´or, A., & Birkmann, S. M. 2006, MNRAS, 373,1213Lagache, G., Abergel, A., Boulanger, F., & Puget, J.-L. 1998, A&A, 333, 709 16 –Lagache, G., Bavouzet, N., Fernandez-Conde, N., et al. 2007, ApJ, 665, L89Lagache, G., Dole, H., & Puget, J.-L. 2003, MNRAS, 338, 555Lagache, G., Haffner, L. M., Reynolds, R. J., & Tufte, S. L. 2000, A&A, 354, 247Larson, K. A. & Whittet, D. C. B. 2005, ApJ, 623, 897Laureijs, R. J., Clark, F. O., & Prusti, T. 1991, ApJ, 372, 185Laureijs, R. J., Haikala, L., Burgdorf, M., et al. 1996, A&A, 315, L317Low, F. J., Young, E., Beintema, D. A., et al. 1984, ApJ, 278, L19Mattila, K. 1980, A&A, 82, 373Miville-Deschˆenes, M. & Lagache, G. 2005, ApJS, 157, 302Miville-Deschˆenes, M.-A., Boulanger, F., Reach, W. T., & Noriega-Crespo, A. 2005, ApJ,631, L57Miville-Deschˆenes, M.-A., Lagache, G., Boulanger, F., & Puget, J.-L. 2007, A&A, 469, 595Miville-Deschˆenes, M.-A., Lagache, G., & Puget, J.-L. 2002, A&A, 393, 749Rieke, G. H., Young, E. T., Engelbracht, C. W., et al. 2004, ApJS, 154, 25Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525Stansberry, J. A., Gordon, K. D., Bhattacharya, B., et al. 2007, ArXiv e-prints, 707Stepnik, B., Abergel, A., Bernard, J.-P., et al. 2003, A&A, 398, 551van Dishoeck, E. F. & Black, J. H. 1988, ApJ, 334, 771Walter, F., Brinks, E., de Blok, E., et al. 2005, in Bulletin of the American AstronomicalSociety, Vol. 37, Bulletin of the American Astronomical Society, 1258–+ This preprint was prepared with the AAS L A TEX macros v5.2. 17 – 240 270 300-30-30-15-15Powered by Aladin 060090120+30+15+15+30Powered by Aladin Fig. 1.— Position of the SINGS fields (red circles and blue squares) overlaid on the dustcolumn density maps from Schlegel et al. (1998) centered around the north galactic pole (leftpanel) and the south galactic pole (right panel). The blue squares correspond to the fieldsselected for this study (the variation in the infrared emission is dominated by the cirruscomponent). A grid of galactic coordinates is overlayed. 18 –Fig. 2.— Variations of the square of the standard deviation (related to the power spectrum ofthe signal) measured in each field with the mean brightness at 100 µ m. The observed valuesare compared with models for the different contributions: the infrared galaxies clustering(dotted-dashed line), the poisson noise (dashed line) and the cirrus variation (dotted line).This enables us to define a cut in 100 µ m brightness (the vertical black line) above which thecirrus variations dominate over CIB fluctuations. 19 –Fig. 3.— 160–100 scatterplots for all SINGS observations with < B > ≥ . 45 MJy/sr.In each plot, a canonical slope of 2.0 is represented by a dashed line (corresponding toa temperature of 17.5K). A linear fit is performed on the correlation and the best fit isrepresented with a solid line. The value of the slope obtained is written in the legend. 20 –Fig. 4.— idem as Fig. 3 21 –Fig. 5.— Variations of the 160/100 surface brightness ratio with the mean 100 µ m surfacebrightness. The diamonds represent the SINGS observations. The blue line denotes a typicalratio of 2.0. 22 –Fig. 6.— 60–100 scatterplots for all SINGS observations with < B > ≥ . < B > ≥ . µ m cirrus brightnesses in each field as well as standard deviations at 60, 100 and 160 µ m field (l,b) ( λ , β ) Area (deg ) σ < B > (MJy/sr) σ σ NGC0337 ( 126.983 , -70.4576) ( 10.0714 , -12.6317) 0.048 —- 5.75 0.713 1.951NGC0584 ( 148.863 , -68.1957) ( 17.8130 , -15.3840) 0.100 0.057 2.38 0.177 0.387NGC0628 ( 138.043 , -46.3226) ( 27.4637 , 5.05541) 0.115 —- 3.09 0.287 0.809NGC0855 ( 143.962 , -32.1481) ( 39.9931 , 13.3583) 0.056 —- 3.33 0.111 0.399NGC0925 ( 144.527 , -25.8230) ( 44.7625 , 17.7346) 0.124 0.040 3.09 0.101 0.316NGC1097 ( 227.959 , -65.0380) ( 26.0183 , -43.9238) 0.113 0.037 1.08 0.112 0.250NGC1291 ( 247.760 , -57.5207) ( 27.3597 , -55.9323) 0.154 0.018 0.68 0.104 0.212NGC1316 ( 240.631 , -56.8186) ( 32.3476 , -53.2688) 0.197 0.042 1.00 0.093 0.189NGC1377 ( 212.367 , -52.3728) ( 44.7413 , -38.7417) 0.042 0.029 1.65 0.071 0.246NGC0024 ( 40.4333 , -80.1196) ( 351.074 , -23.9783) 0.099 0.043 1.14 0.073 0.130NGC1404 ( 237.002 , -53.9016) ( 38.2431 , -52.8271) 0.069 0.032 0.59 0.059 0.101NGC1482 ( 213.765 , -48.2291) ( 50.1099 , -39.5384) 0.039 0.040 2.44 0.087 0.212NGC1512 ( 248.603 , -48.4042) ( 40.9466 , -61.8125) 0.176 0.042 0.49 0.075 0.132NGC1566 ( 264.199 , -43.5133) ( 32.0129 , -73.2076) 0.116 0.025 0.39 0.081 0.124NGC1705 ( 260.913 , -38.8932) ( 50.3989 , -74.3634) 0.043 0.033 0.42 0.094 0.121NGC2403 ( 150.172 , 28.7424) ( 102.746 , 43.5117) 0.261 0.049 1.62 0.132 0.286HolmbergII ( 143.899 , 32.2721) ( 106.073 , 49.5466) 0.077 0.028 1.34 0.091 0.300M81DwarfA ( 143.536 , 32.5857) ( 106.476 , 49.8996) 0.041 0.043 1.00 0.121 0.248DDO053 ( 148.991 , 34.4963) ( 110.352 , 45.7107) 0.077 0.048 1.57 0.113 0.324NGC2798 ( 178.927 , 43.7838) ( 127.750 , 25.1694) 0.051 0.040 1.04 0.086 0.166NGC2841 ( 166.458 , 43.6210) ( 124.989 , 33.8516) 0.101 0.043 0.74 0.061 0.140NGC2976 ( 143.612 , 40.4772) ( 118.782 , 50.4361) 0.076 0.100 2.59 0.594 1.951HolmbergI ( 140.502 , 38.2525) ( 115.197 , 52.8405) 0.079 0.036 1.18 0.185 0.659NGC3049 ( 226.610 , 44.4312) ( 146.934 , -2.98701) 0.027 —- 2.20 0.223 0.504NGC3190 ( 211.891 , 54.4481) ( 147.733 , 10.7909) 0.090 —- 1.61 0.094 0.195NGC3184 ( 178.537 , 54.9707) ( 139.768 , 28.3651) 0.102 0.031 0.93 0.089 0.149NGC3198 ( 170.656 , 54.2275) ( 138.036 , 32.7413) 0.050 0.039 0.52 0.074 0.157IC2574 ( 139.983 , 43.1538) ( 123.351 , 52.9795) 0.156 0.036 1.38 0.205 0.509NGC3265 ( 200.457 , 58.5603) ( 147.821 , 18.3074) 0.029 0.030 1.30 0.055 0.146MRK33 ( 156.610 , 52.2144) ( 135.166 , 41.0632) 0.032 0.032 0.86 0.212 0.424NGC3351 ( 232.652 , 56.1499) ( 157.328 , 3.64920) 0.074 —- 1.95 0.108 0.210NGC3521 ( 254.316 , 52.8287) ( 166.849 , -5.14601) 0.149 —- 2.81 0.328 0.718NGC3621 ( 280.580 , 25.9208) ( 184.582 , -34.1228) 0.180 0.111 4.12 0.444 0.900NGC3627 ( 240.246 , 64.2844) ( 165.038 , 8.27119) 0.097 —- 1.70 0.179 0.248NGC3773 ( 249.132 , 66.7215) ( 169.474 , 8.73315) 0.034 —- 1.84 0.102 0.125NGC3938 ( 153.689 , 68.7273) ( 156.788 , 39.2954) 0.075 0.065 1.01 0.194 0.328NGC4125 ( 130.164 , 50.9404) ( 139.741 , 57.1897) 0.095 0.037 0.81 0.174 0.285NGC4236 ( 127.310 , 46.9854) ( 134.438 , 60.6054) 0.267 0.035 0.66 0.098 0.170NGC4254 ( 267.627 , 75.3684) ( 177.739 , 15.3315) 0.075 0.054 2.37 0.161 0.370NGC4321 ( 267.852 , 77.0715) ( 178.035 , 17.0118) 0.119 0.053 1.37 0.160 0.214NGC4450 ( 270.395 , 78.8333) ( 178.801 , 18.7003) 0.079 0.032 1.36 0.119 0.338NGC4536 ( 291.454 , 65.1324) ( 186.331 , 5.66591) 0.102 —- 1.64 0.090 0.239NGC4552 ( 285.274 , 74.9059) ( 182.409 , 14.8782) 0.106 —- 2.28 0.210 0.688NGC4559 ( 193.095 , 85.8409) ( 175.296 , 29.2435) 0.124 0.029 1.05 0.082 0.184NGC4569 ( 285.793 , 75.9750) ( 182.385 , 15.9551) 0.101 0.060 2.65 0.174 0.470NGC4579 ( 287.803 , 74.2939) ( 183.194 , 14.3801) 0.071 —- 2.37 0.126 0.311NGC4594 ( 297.305 , 51.1460) ( 193.066 , -6.96614) 0.151 —- 3.12 0.141 0.298NGC4625 ( 131.096 , 75.1008) ( 168.734 , 41.5922) 0.047 0.052 0.88 0.121 0.239NGC4631 ( 145.167 , 83.5710) ( 174.202 , 33.9219) 0.171 0.054 1.02 0.182 0.248NGC4725 ( 271.647 , 88.6791) ( 179.905 , 28.4957) 0.079 0.037 0.70 0.102 0.161NGC4736 ( 124.738 , 75.5220) ( 170.796 , 42.2419) 0.187 0.047 0.67 0.192 0.293DDO154 ( 110.504 , 89.5114) ( 179.918 , 30.2879) 0.042 0.043 0.57 0.085 0.137NGC4826 ( 311.183 , 85.0014) ( 183.181 , 25.6681) 0.125 0.059 1.89 0.182 0.239DDO165 ( 121.301 , 49.1180) ( 142.645 , 62.9864) 0.058 0.051 1.14 0.171 0.301NGC5033 ( 101.051 , 79.6584) ( 178.930 , 40.1141) 0.128 0.033 0.54 0.096 0.149NGC5055 ( 107.903 , 73.9458) ( 175.505 , 45.4687) 0.163 0.052 0.75 0.178 0.267NGC5194 ( 106.114 , 68.6881) ( 174.458 , 50.7106) 0.233 0.087 1.23 0.364 0.712Tololo89 ( 318.732 , 27.7794) ( 219.272 , -19.6125) 0.051 0.186 5.26 0.544 0.978NGC5408 ( 316.804 , 20.0386) ( 223.038 , -26.7581) 0.056 0.163 5.55 0.761 1.702NGC5474 ( 101.503 , 59.8880) ( 174.929 , 59.7255) 0.061 0.027 0.48 0.086 0.185NGC5713 ( 350.094 , 52.5981) ( 217.013 , 14.3062) 0.052 —- 2.08 0.070 0.187NGC5866 ( 92.3464 , 52.7620) ( 186.246 , 66.8587) 0.105 0.029 0.62 0.097 0.210IC4710 ( 327.879 , -22.1180) ( 273.178 , -43.4420) 0.034 0.119 5.69 0.426 1.007NGC6822 ( 24.6969 , -17.9857) ( 294.785 , 6.07295) 0.185 —- 7.77 0.745 1.705NGC6946 ( 95.3945 , 12.0399) ( 356.855 , 72.2046) 0.086 0.125 9.57 0.722 1.948NGC7331 ( 93.0851 , -20.9465) ( 356.050 , 39.1427) 0.126 0.121 3.91 0.429 0.761NGC7552 ( 347.564 , -64.7400) ( 330.605 , -34.6836) 0.046 0.060 0.81 0.128 0.173 25 –Table 1—Continued field (l,b) ( λ , β ) Area (deg ) σ < B > (MJy/sr) σ σ NGC7793 ( 5.28269 , -76.5898) ( 344.625 , -29.1746) 0.119 0.046 0.95 0.126 0.184 Table 2. Infrared colors in the observations, i.e. the 160/100 and 60/100 brightness ratiosobtained from a fit of the correlations in each field field 160/100 color 60/100 colorNGC0337 2 . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . 07 —NGC3621 1 . ± . 04 0 . ± . . ± . . ± . . ± . . ± . . ± . . ± . 04 0 . ± . . ± . . ± . . ± . 04 —NGC6946 2 . ± . . ± . . ± . 07 0 . ± ..