Searching for Dark Matter Annihilation in the Smith High

13 Pages·2014·5.66 MB·English

Searching for Dark Matter
Annihilation in the Smith
High-Velocity Cloud
Alex Drlica-Wagner
!
G. Gómez-Vargas, J. Hewitt, T. Linden, L. Tibaldo
on behalf of the Fermi-LAT Collaboration
!
August 29, 2014
COSMO-2014
ApJ 790 24 (2014) [arXiv:1405.1030]
High-Velocity Clouds (HVCs)
• HVCs are coherent over-densities of HI
covering ~40% of the sky.
!
• They are kinematically separable from
Galactic disk gas (v ~ 102 km/s)
!
• The origin of HVCs is unclear and may
differ from object to object
– Ejected from the Galactic disk?
– Stripped from the Magellanic Clouds?
– Accreted during galaxy formation?
Alex Drlica-Wagner | Smith Cloud 2
High-Velocity Clouds (HVCs)
• HVCs are coherent over-densities of HI
covering ~40% of the sky.
!
• They are kinematically separable from
Galactic disk gas (v ~ 102 km/s)
!
• The origin of HVCs is unclear and may
differ from object to object
– Ejected from the Galactic disk?
– Stripped from the Magellanic Clouds?
– Accreted during galaxy formation?
CHAPTER 2. INDIRECT DETECTION 9
• A sub-population of HVCs may inhabit
dark matter halos that failed to form
Simulated Dark Matter Distribution
galaxies.
!
• Potential targets for indirect detection
of dark matter annihilation
– May trace nearby dark matter halos
– Some gamma-ray emission expected
from cosmic-ray interactions
Springel et al. (2008)
Figure 2.2 A high resolution numerical simulation of a Milky-Way-sized dark matter
Alex Drlica-Wagner | Smith Cloud halo from the Aquarius Project [13]. Thousands of dark matter subhalos c2an be
seen superimposed on the smooth primary halo. The brightness of each pixel is
proportional to the logarithm of the squared dark matter density projected along the
line of sight. Adapted from Springel et al. [13].
2.2 Indirect Detection Formalism
Having discussed the distribution of dark matter in the Galactic environment, we
now consider the predicted �-ray signal from dark matter annihilation. The �-ray
flux from dark matter annihilation depends both on the spatial distribution of dark
matter and on the particle physics governing annihilation. The signal flux, ⇧ (�⇤)
�
(phcm 2s 1sr), expected from annihilation in a dark matter density distribution,
� �
⇤(r), can be expressed as
1 ⌅v Emax dN
⇧ (�⇤) = ⇤2(r)dl d⇤ ⇥ ⇤ �dE . (2.6)
� ⇥· 4⇥ 2m2 dE �
��⇤⇥�l.o.s. ⇤ DM �Emin �
J-factor ⇥PP
⌃ ⇧⌅ ⌥ ⌃ ⇧⌅ ⌥
The preceding J-factor represents the line-of-sight integral through the square of the
dark matter density integrated over a solid angle �⇤. The second factor, ⇥ ,
PP
is strictly dependent on the properties of the dark matter particle: the thermally-
averaged annihilation cross section, ⌅v , the particle mass, m , and the di⌅erential
DM
⇥ ⇤
L22 LOCKMAN ET AL. Vol. 679
Fig. 1.—GBT H i image of the Smith Cloud at V p 100 km s!1 showing
LSR
the cometary morphology that strongly suggests that the Cloud is moving to lower
longitude and toward the plane and that it is interacting with the Galactic ISM.
Arrows mark the tracks of the velocity-position slices of Figs. 2 and 3. Fig. 3.—GBT H i velocity-position slice through the major axis of the Cloud
at the location of the arrows in Fig. 1. Marks on the vertical axis are every
157.5!. Along this track, there are H i clumps at low velocity that match the
kinematic bridges between the Cloud and Galactic emission
gaps in the main Cloud. The clumps have likely been stripped from the Cloud.
(several are marked with dotted arrows), as well as clumps of
Two are marked by the solid arrows. Two line wings that form kinematic
H i (two are marked by solid arrows) at velocities "40 km s!1
bridges between the Cloud and Galactic gas are marked by the dotted arrows.
that correspond to gaps in the Cloud. The clumps are likely The main part of the Cloud shows systematic velocity gradients from the
changing projection of its space velocity with respect to the LSR. The tilted
material stripped from the Cloud.
lines show the expected run of V with position for V p 296 km s!1 (upper
LSR tot
part of the Cloud) and V p 271 km s!1 (lower part). The Cloud consists of
tot
4. DISTANCE TO THE CLOUD at least two coherent kinematic pieces.
Portions of the Smith Cloud appear to have been decelerated
the “far” kinematic distance for a flat rotation curve with
by the ambient medium through which it moves, and we use
R p 8.5 kpc and V p 220 km s!1.
The Smith Cloud
0 0
this to estimate a distance to the Cloud. The GBT data show
There are other determinations of the distance. The brightness
disturbances in Galactic H i attributable to the influence of the
of diffuseHa emission fromtheCloudandamodelfortheGalactic
L22 LOCKMAN ET AL. Vol. 679
Smith Cloud at V 35 km s!1 but not at V 0 km s!1.
UV flux give either 1 or 13 kpc (Bland-Hawthorn et al. 1998;
LSR LSR
≥ ≤
If the Smith Cloud is interacting with Galactic gas whose nor-
Putman et al. 2003). Recently, Wakker et al. (2008) have bracketed
mal rotational velocity is in this range, it implies that that the
the distance by looking for the Cloud in absorption against several
Cloud has a distance in the range 11.1 kpc ! d ! 13.7 kpc,
k stars, finding 10.5 kpc ! d 14.5 kpc. The three methods give
≤
• The Smith Cloud is one of the best identical results, and we adopt the kinematic distance d p
12.4 ! 1.3 kpc for the remainder of this Letter.
characterized HVCs (e.g., Lockman et al. 2008)
!
5. PROPERTIES OF THE CLOUD
• Located at (l,b) = (38.67, -13.41) with an HI gas
The Smith Cloud lies in the inner Galaxy below the Perseus
mass of ~106 M (total gas mass >2x106 M )
⊙ ⊙ spiral arm, R p 7.6 kpc from the Galactic center. The prop-
!
erties of the Cloud obtained from the GBT data are presented
in Table 1. The brightest H i emission at l, b p 38.67", !13.41"
• Accurate distance determination from
is near the Cloud tip. ThLeoHckmi mana ests aol.f (1200608M) is a lower limit
,
– Stellar bracketing
because the Cloud appears to consist of multiple fragments
– Interaction with disk gas
TABLE 1
– H-alpha reflection from the Galactic disk
Fig. 1.—GBTHHiiPimraogpeeorfttiheesSomfiththCelouSdmaittVh Cplo1u0d0 km s!1 showing
! LSR
the cometary morphology that strongly suggests that the Cloud is moving tolower
longitude and toward thPerpolapneertaynd that it is interactinVgawluiteh the Galactic ISM.
• The Smith Cloud resides at a heliocentric Arrows mark the tracks of the velocity-position slices of Figs. 2 and 3. Fig. 3.—GBT H i velocity-position slice through themajoraxisoftheCloud
l, b (deg) .................. 38.67, !13.41 at the location of the arrows in Fig. 1. Marks on the vertical axis are every
distance of 12.4 +/- 1.3 kpc (nearest dwarf
Distance (kpc) ............ 12.4 ! 1.3 157.5!. Along this track, there are H i clumps at low velocity that match the
kinematic bridges between the Cloud and Galactic emission
galaxy at 23 kpc). R (kpc) ................... 7.6 ! 0.9 gaps in the main Cloud. The clumps have likely been stripped from the Cloud.
(several are marked with dotted arrows), as well as clumps of
Two are marked by the solid arrows. Two line wings that form kinematic
z (kpc) .................... !2.9 ! 0.3
H i (two are marked by solid arrows) at velocities "40 km s!1
bridges between the Cloud and Galactic gas are marked by the dotted arrows.
! T (K) ..................... 15.5
that correspbond to gaps in the Cloud. The clumps are likely The main part of the Cloud shows systematic velocity gradients from the
Dv (km s!1) .............. 16.0
changing projection of its space velocity with respect to the LSR. The tilted
material stripped from the Cloud.
• The distance, direction of motion, and N (cm!2) ................ 5.2 # 1020 lines show the expected run of V with position for V p 296 km s!1 (upper
Hi LSR tot
V (km s!1) ............. 99 ! 1 part of the Cloud) and V p 271 km s!1 (lower part). The Cloud consists of
Fig. 2.—sGByTsHteimveilocc itvy-epolositciointysli cde ithsrtorugibh uthetiSomnith aClllooudwal otnhgea 3D LSR tot
H i mas4s. (DMIST)A.N.C..E..T.O...T.HE CLOU1D106 at least two coherent kinematic pieces.
track through the minor axis of the Cloud (marked by arrows in Fig. 1). The ,
trajectory of the Smith Cloud to be determined Projected size (kpc) ...... 3 # 1
edges of the Cloud show a sharp gradient in velocity from V 100 km s!1 Portions of the Smith Cloud appear to have been decelerated
LSR ∼ the “far” kinematic distance for a flat rotation curve with
to the lower velocities of Galactic H i. We interpret this as evidence of the by the ambieNnottme.e—diAumll bthutroiungtehgrwalhqicuhanittitmiesovaepsp,lyand we use
R p 8.5 kpc and V p 220 km s!1.
interaction between the Cloud and the gaseous halo of the Milky Way. The this to estitmo athtee dairdeicsttioanncoef gtoretahteestCNlouadt.tThehpeoGsiBtioTndata show 0 0
Hi There are other determinations of the distance. The brightness
arrow marks the decelerated ridge shown in Fig. 4. l, b p 38.67", !13.41".
disturbances in Galactic H i attributable to the influence of the
ofdiffuseHaemissionfromtheCloudandamodelfortheGalactic
Smith Cloud at V 35 km s!1 but not at V 0 km s!1.
UV flux give either 1 or 13 kpc (Bland-Hawthorn et al. 1998;
Alex Drlica-Wagner | Smith Cloud LSR ≥ LSR ≤ 3
If the Smith Cloud is interacting with Galactic gas whose nor-
Putman et al. 2003). Recently, Wakkeretal.(2008)havebracketed
mal rotational velocity is in this range, it implies that that the
the distance by looking for the Cloud in absorption againstseveral
Cloud has a distance in the range 11.1 kpc ! d ! 13.7 kpc,
k stars, finding 10.5 kpc ! d 14.5 kpc. The three methods give
≤
identical results, and we adopt the kinematic distance d p
12.4 ! 1.3 kpc for the remainder of this Letter.
5. PROPERTIES OF THE CLOUD
The Smith Cloud lies in the inner Galaxy below the Perseus
spiral arm, R p 7.6 kpc from the Galactic center. The prop-
erties of the Cloud obtained from the GBT data are presented
in Table 1. The brightest H i emission at l, b p 38.67", !13.41"
is near the Cloud tip. The H i mass of 106 M is a lower limit
,
because the Cloud appears to consist of multiple fragments
TABLE 1
H i Properties of the Smith Cloud
Property Value
l, b (deg) .................. 38.67, !13.41
Distance (kpc) ............ 12.4!1.3
R (kpc) ................... 7.6!0.9
z (kpc) .................... !2.9 ! 0.3
T (K) ..................... 15.5
b
Dv (km s!1) .............. 16.0
N (cm!2) ................ 5.2#1020
Hi
V (km s!1) ............. 99!1
Fig. 2.—GBT H i velocity-position slice through the Smith Cloud along a LSR
H i mass (M ) ........... 1106
track through the minor axis of the Cloud (marked by arrows in Fig. 1). The ,
Projected size (kpc) ...... 3#1
edges of the Cloud show a sharp gradient in velocity from V 100 km s!1
LSR ∼
to the lower velocities of Galactic H i. We interpret this as evidence of the Note.—All but integral quantities apply
interaction between the Cloud and the gaseous halo of the Milky Way. The to the direction of greatest N at the position
Hi
arrow marks the decelerated ridge shown in Fig. 4. l, b p 38.67", !13.41".
Solar 

Location
Approximate
Impact Site
Smith Cloud
Artist’s Concept
4
Smith Cloud Dark Matter
• The 3D trajectory of the Smith16 4C4loud NICHOLS &NBicLhAoNlsD &-H BAlaWnTdH-HOaRwNthorn (2009) Vol. 707
suggests that it passed through the
+150 Myr
Galactic disk ~70 Myr ago.
−150 Myr
!
6
• The gaseous component of the cloud has a
4
c) 20
p 2
k
weak self-gravity and ram pressure forces z ( 0 0
−2
would dissipate the cloud during a passage −20
−30 −20 −10 0 10 20 y (kpc)
through the Galactic disk.
x (kpc)
!
Figure 3. Orbit of the Smith Cloud, calculated using the potential from Wolfire et al. (1995). The current position is represented by an unfilled circle and the Smith
Cloud is travelling in the direction of the arrows, with heights below the disk represented by a dotted line. The Sun’s position is shown as a filled circle on the Solar
See Doug Spolyar’s talk about Gaia…
• This suggests that the Smith CCirclleo. Tuhedth inmdoatteyd l inbe reep resents the projection of the Smith Cloud onto the disk. The disk is represented by a solid line at 30 kpc.
bound by a dark matter halo with tidal
Table 1
The Astrophysical Journal, 790:24 (7pp), 2014 July 20 Drlica-Wagner et al.
NFW, Burkert, and Einasto Profiles
mass ~108 M (Nichols & Bland-Hawthorn,
⊙
NFW TabBluerk1ert Einasto 90
2009).
f (x) x 1(1+x) 2 Summary of Smithf C(xl)oud(1D+axrk) M1(1at+texr2)H1alo Parameters f exp[ 2/α(xα 1)]/4
ρ − − ρ − − ρ
= = = − −
! f (x) 3 ln(1+x) x f (x) 3 ln(1+x2) +ln(1+x) tan 1x f βγ(3/α,2xα/α) 100
• Such a dark matter halo would efmϕx(xt,ev=sn) =!d3 t1o− lna(−1x+nx1)+ x"Profile (krpsc)fϕ(x,mvs()M==322ρk!0p1c+−23x1) tan−1x −(MM1−ti+dax1l) −ln(1"+x)(GeVJ-2fafccϕmt(ox−r,5vss)r)=m β= 21/αα−1/α6γ0(2/α,2xα/α) Kdeg.)
! " ⊙!# +1 $1 1 ln(1#+x⊙2) $ γ!(3/α,2xα/α)/x 1 e(
angular radius of ~5˚ around the cloud. NFW 1.04 3.7 1027 − x 1.1 108 9.6 1019 − g.) 30 − ur
! fgas(x,vs,cg) = e−3(vs/cg)2(1B+urxk)3e(rvst/cg)2/x 1.04 fgas(x,3v.s7,c××g()11=+0[x7e2#−)((11/+21)/(x1$)/xtan1−1.)13]x(3××(/12)+1(v"0xs/)8c(g1)+21/x) 4.2 ×× 1f0ga1s8(x,vs,cg) = exde(dep(−vs2/cg2fϕ(x")) 10 mperat
• To mitigate the impact of tidal disruption, Einasto 1.04 9.2×× 106 −2.0 × 108 1.8 × 1020β = (3/4)8−1/Latituαe2/α0α−1+3/α ssTe
e
c n
wanen cihoinlasteiorvna stiivgenlayl mfroodme lo tnhley dthaerdspNe kenoiestn dietm;ysn.cpfT.raehoSefittrelfert onef1ubgrears˚rq.g uH
eaetnrateilt.iγe(2si0siN(0n∆t2hoeΩ)eta,eclToh∼.awbIcnelo9ertl.eu5i6ngm.c×rnoamta1erpd0ele−tJthe4-efgsadarce)mtn.omsritasyfauprnreocfitciloaenlfc,pxu,lt≡ahteerdd/arroskvismetrahtaetessrcoamllieadsr-saadpnirugosfil,elvecsfomisn,tethhewehdiatalhrokrcamirdcaiututleasrr1pv◦oeltoencittiya,lapnrodGalactificlge−ifs3ϕ0t,heangdasthseougnads 1 edBright
t
a
r
of the halo. central density of a halo theant cvairpiasluizleadteast tzhe s0p.aTtihaisl fdaicsttorributiopnotoenftitahleisdgairvkenmbayttWerolsfiirgenaetl,al. (1995) normalized by a eg
also contributes to other halo properties such a=s the scale radius circular velocity of v 220 km s 1 at th−e6S0olar Circle. In Int
while Φ sets its spectral character. c −
r ∆ 1/3 and the scale velocity v PP∆1/6. Figure 3, we show the p=redicted orbit of the cloud system. In
s − Thesre is significant uncertainty in the dark matter density
∝ ∝
agreement with Lockman et al (2008), we fin9d0 that the Smith 0.1
Alex Drlica-Wagner | Smith Cloud 3. MOpDrEoLfilSeEoTfUPthe Smith Cloud, thus we calculate the J-factor5for − 100 50 0 50 100
Cloud has intersected the disk 70 Myr ago and wil−l pass −
Velocity (km s 1)
∼ −
each of the three dark matter halothprrooufighletshefidtisbkyagNaiBn Hin093.0TMhyer.
We consider two models of evolution, one in which the Smith ∼ Figure1.Latitude–velocitydistributionofGalacticHigasfromtheLABsurvey
parameters of each profile were deFroivreadll sfurbohmalotmheodteolst,awl edianrvkestigate the effects of dynam-
Cloud is infalling for the first time, hereafter the Infalling Orbit (Kalberla et al. 2005) integrated over the longitude range of the Smith Cloud
ical friction on the orbit trajectory. The formalism used is de-
Models, and a second modelmwahtetreerthteidSamlimthaCslso,udMhtaisdaall,reraedqyuiredstcoribceodnbfiynJeianggas&iBninthneeyS(2m00it0h): we po(i3n6t◦o!utlt!ha4t6th◦)e. vTahleuecolor represents the integrated brightness temperature of
been maximally stripped duCelotoudpredvuioriunsgoribtsitsm, hoesrteareftceernthteinteraction with the Galactic disk. the 21 cm Hi line as a function of latitude and velocity with respect to the local
for the circular velocity in their Table 1 should be v 235
Repeated Orbit Models. TheTsehbeostehtsihdaarel mcoamsmseosnaferaetucraeslc: (u1l)ated independently by NBH09 for standard of recst=. Gas associated with the Smith Cloud is enclosed by the black
km s 1 (not 181 km s 1 as quoted) to be consistent with their
they have the same trajectory today, (2) the dark matter halo − − box and is removed from our Galactic foreground model.
each dark matter profile. We descrainbaelyseias.cBhuot fovtehrethdeaprakstmfeawttheurndred million years, dynam-
has been tidally stripped down from some larger initial mass
(A color version of this figure is available in the online journal.)
density profiles in terms of a scale riacdaliufrsi,ctrio,nainsdfoaunscdatloehdaevnesointyly, minimal effect, even in the
(M ) in an identical fashion before our calculations commence s
vir
high mass limit. This is because, once again, the impact of gas
at apogalacticon. The imporρtan,t adsistilniscttieodn iisntidTaalbsltreipp1i.ngTohfe Einasto profile depends on an
0
loss from the subhalo close to the disk is found to dominate the
the gas is possible in the InafadldlinitgioOnrablitpMaroadmelsetbeurt αnowt ihnitchhe is set to a value of 0.17: The distribution of target material is derived from surveys of
evolution of the subhalo. We assume that any drag between the
Repeated Orbit Models; in both cases, ram pressure stripping
the 2.6 mm CO and 21 cm H i lines, supplemented with inter-
model clouds and the Galactic corona is negligible and does not
by the hot halo is important. For each case, the evolution of the
ρ r3 stellar reddening maps from infrared observations of interstellar
affect the orbit.
Smith Cloud is considered for the NFW, Einasto, a0ndsBurkert
ρ(r) NFW, (2)
Each model cloud is considered to be adduarskt.mNaottetarbployte,ntthiaelofficial Fermi-LAT model of Galactic diffuse
models. = r(r + r)2
s
well filled with gas in isothermal hydrostatic equilibrium. We
emission recommended for discrete source analysis includes a
Theevolutionofthemodelcloudswascalculatedasafunction
assume a primordial helium abundance n /n 1/12 and
of three variables: the initial virial mass at the time of fρormra3tion γH-erayHe=mission component associated with the Smith Cloud.10
0 s metallicity of Z/Z 0.1. We also assume that the gas has
(i.e., before the dark matter halo fell iρn(tor)the Galaxy), the dark Burkert, (3)
= (r + r) r2 + r2 a temperature of 1.⊙2 =104 K and adopt aWn eionriezmatioovnefragcatisoncorrelated with the Smith Cloud from our
matter profile at this time, and the initial hydrogen gas mass
s s × analysis for two reasons. First, the intensity and spectrum of
of 50% for the Smith Cloud, slightly below the newly updated
at apogalacticon. For both the Repeated Orbit Model and the
Infalling Orbit Model, the evolution of 7503 model!clo2uds "r Hα+/H0 ratioinHilletal.(2009).Thistempceroastumreicanrdaiyosniazaretiopnoorly constrained at the distance of the Smith
were calculated, corresponding to 6ρ1(lro)garithρmiecxalply spaced fraction1then giEveinaassotuon.d spee(d4o)f cg 1C1lkomuds,−w1.hTihcehglaesaisds to considerable uncertainty in the predicted
0 =
= −α r distr−ibuted in the potential well according to the gas density
virial masses in the range M (5 107)–(5 1010)M s γ-ray flux. Second, removing gas from the Smith Cloud elimi-
vir # $% & ’(
and 41 logarithmically spaced gas=mass×es in the ra×nge M ⊙ profile nH(x,vs,cg) nH,0fgas(x,vs,cg), where x r/rs
gas = nates a pote≡ntially degenerate emission component which may
To avoid peripheral regions=whereistidthael ssctarileppraindigusm, vayisalttheer hthaleo circular velocity given by
(1 106)–(1 108)M . s
result in artificially strong limits on the dark matter annihilation
Equation (3) and f (x,v ,c ) is given in Table 1.
T×he orbit ×of the S⊙mithdCarlokumd awttaesr dcaelncusliattye,dwuesitnrguntchaete our model of the γ-garsay instengsity
For the initial dimension of the modreal tcelowuditsh, itnhethseoucnldoud.
position and velocity datapfrroofimleLo1ckmfraonmet tahl.e(2c0e0n8t)erfoorf the Smith Cloud. To simplify
◦ crossingtimeis200Myr,fallingtoabout30Myratthedisk.We
We create Galactocentric annuli for the H i gas distribution
the tip of the Smith Cloud: (R,z) (7.6, 2.9) kpc and
comparisons with other dark matter annihilation targets (i.e.,
therefore begin each orbit at the apogalacticon, approximately
= −
(v ,v ,v ) (94,270,73) km s 1. The form of the Galactic by transforming 21 cm brightness temperatures into column
R φ z dwarf−spheroidal galaxies), we compute the integrated J-factor
=
densities using the composite LAB survey (Kalberla et al. 2005)
from the Smith Cloud within this 1 radius (Table 1). This radius
◦
and the Galactic rotation curve given by Clemens (1985). We
contains 60% of the total predicted γ-ray flux when cuspy
∼ follow the procedure employed by Ackermann et al. (2012) to
NFW or Einasto profiles are assumed and 10% of the total
∼ excise the gas associated with the Smith Cloud from the Galactic
predicted flux from the cored Burkert model. Thus, this choice
gas distribution. Specifically, we remove gas in the region
of radius yields a conservative estimate for the total J-factor of
36 l 46 and 20 b 10 , which has a velocity with
the Smith Cloud since the dark matter distribution may extend ◦ ! ! ◦ ◦ ! ! ◦
− −
respect to the local standard of rest in the range 70–125 km s 1
to several degrees. −
(Figure 1).11 The primary uncertainty in the conversion from
brightness temperature to column density comes from the
3. GALACTIC FOREGROUND MODELING
assumed spin temperature (T ) used to correct for the opacity
S
The observed foreground γ-ray emission from the region of the 21 cm line. We find that the gas density in the region
surrounding the Smith Cloud is dominated by π0-decay emis- of the Smith Cloud changes by <15% when the assumed spin
sion produced from cosmic rays interacting with the atomic temperature is changed from T 125 K to T 105 K (i.e.,
S S
= =
and molecular hydrogen gas of the Milky Way.8 The GALPROP the gas is optically thin). When analyzing the γ-ray data we
cosmic-ray propagation code can be used to model the dif- set T 125 K and find that this choice has little impact
S
=
fuse Galactic γ-ray foreground from processes such as inelastic on our results. We follow the procedure of Ackermann et al.
hadronic collisions, bremsstrahlung, and inverse-Compton scat- (2012) to trace the CO distribution from the 2 mm composite
tering.9 GALPROP accounts for effects such as diffusion, reac- survey of Dame et al. (2001). Due to the small CO content of
celeration, and energy loss via mechanisms such as synchrotron HVCs (Akeson & Blitz 1999), we do not alter the CO map
radiation (Strong & Moskalenko 1998; Strong et al. 2009). in the region of the Smith Cloud. The Galactic foreground
also contains a contribution from neutral gas that cannot be
8 The γ-ray emission from inelastic hadronic interactions is composed of
many processes, the most important of which being the production of π0, 10 http://fermi.gsfc.nasa.gov/ssc/data/access/lat/BackgroundModels.html
which decays primarily to γγ. 11 Gas with velocity >125kms 1 contributes less than 0.8% of the total
−
9 http://galprop.stanford.edu column density.
3
Smith Cloud J-Factor
1021
Smith (NFW)
Smith Cloud
Smith (Burkert)
Smith (Einasto)
Dwarf Galaxies
1020
)
5
�
m
c
2
v
e
G 1019
(
r
o
t
c
a
Ultra-Faint 

f
–
J
Dwarf Galaxies
1018
1017
101 102
Distance (kpc)
Alex Drlica-Wagner | Smith Cloud 6
5-Year LAT All-Sky Map
Smith Cloud
Alex Drlica-Wagner | Smith Cloud 7
Diffuse Modeling
• The standard LAT Galactic interstellar
emission model includes the Smith
Cloud gas as a cosmic-ray target.
The Astrophysical Journal, 790:24 (7pp), 2014 July 20 HI LAB Survey (36˚ < l D<r l4ic1a˚-)Wagner et al.
!
• RemoTvaebl eth1is component from the
90
Summary of Smith Cloud Dark Matter Halo Parameters
foreground model to search for
100
excess emission. )
Profile r ρ M J-factor 60 g.
s 0 tidal
e
(kpc) (M kpc 3) (M ) (GeV2 cm 5 sr) Smith
 d
– Rem− ove target gas in the veloc−ity K
⊙ ⊙ Cloud (
e
NFW 1.04 3.7 ra1n0g7 e from1 .710–110285 km/s 9fr.6om1 a01 95˚x 5˚ g.) 30 ur
× × × de at
Burkert 1.04 3.7 re1g0i7on sur1ro.3un1d0i8ng the t4h.e2 Sm10i1t8h ( er
e 10 p
× × × d m
Einasto 1.04 9.2 Cl1o0u6 d. 2.0 108 1.8 1020 u e
t T
× × × ti 0
a s
L s
– Correct for a dark gas contribution e
c n
i t
Note. Integrated J-factors are calculated over a solid-angle cone with radius 1 t h
using SFD dust map. ◦ ac ig
(∆Ω ∼ 9.6 × 10−4 sr). – Run GALPROP to produce templates Gal −30 1 dBr
e
t
a
r
encapsulates the spatial dfiostrr itbhuet ihoandorfonthice, dbarrekmmsastttrearhsliugnngal, ,and g
e
60 t
n
inverse-Compton gamma-ray emission.
 − I
while Φ sets its spectral character.
PP
There is significant uncertainty in the dark matter density
• Similar to the procedure used to 90 0.1
profile of the Smith Cloud, thus we calculate the J-factor for − 100 50 0 50 100
− −
analyze M31 and the Magellanic Velocity (km s−1)
each of the three dark matter halo profiles fit by NBH09. The
Clouds (Ackermann et al. 2012) Figure1.Latitude–velocitydistributionofGalacticHigasfromtheLABsurvey
parameters of each profile were derived from the total dark
(Kalberla et al. 2005) integrated over the longitude range of the Smith Cloud
matter tidal mass, M , required to confine gas in the Smith
(36 l 46 ). The color represents the integrated brightness temperature of
tidal
◦ ! ! ◦
Cloud during its most recent interaction with the Galactic disk. the 21 cm Hi line as a function of latitude and velocity with respect to the local
standard of rest. Gas associated with the Smith Cloud is enclosed by the black
These tidal masses are calculated independently by NBH09 for
box and is removed from our Galactic foreground model.
each dark matter pArolefix lDer.licWa-Weadgenesrc r |i b Semeitahc Chlouodf the dark matter 8
(A color version of this figure is available in the online journal.)
density profiles in terms of a scale radius, r , and a scale density,
s
ρ , as listed in Table 1. The Einasto profile depends on an
0
additional parameter α which is set to a value of 0.17: The distribution of target material is derived from surveys of
the 2.6 mm CO and 21 cm H i lines, supplemented with inter-
ρ r3 stellar reddening maps from infrared observations of interstellar
0 s
ρ(r) NFW, (2)
dust. Notably, the official Fermi-LAT model of Galactic diffuse
= r(r + r)2
s
emission recommended for discrete source analysis includes a
ρ r3 γ -ray emission component associated with the Smith Cloud.10
0 s
ρ(r) Burkert, (3)
We remove gas correlated with the Smith Cloud from our
= (r + r) r2 + r2
s s analysis for two reasons. First, the intensity and spectrum of
α cosmic rays are poorly constrained at the distance of the Smith
! 2 "r
ρ(r) ρ exp 1 Einasto. (4) Cloud, which leads to considerable uncertainty in the predicted
0
= −α r −
s γ -ray flux. Second, removing gas from the Smith Cloud elimi-
# $% & ’(
nates a potentially degenerate emission component which may
To avoid peripheral regions where tidal stripping may alter the
result in artificially strong limits on the dark matter annihilation
dark matter density, we truncate our model of the γ -ray intensity
rate within the cloud.
profile 1 from the center of the Smith Cloud. To simplify
◦
We create Galactocentric annuli for the H i gas distribution
comparisons with other dark matter annihilation targets (i.e.,
by transforming 21 cm brightness temperatures into column
dwarf spheroidal galaxies), we compute the integrated J-factor
densities using the composite LAB survey (Kalberla et al. 2005)
from the Smith Cloud within this 1 radius (Table 1). This radius
◦
and the Galactic rotation curve given by Clemens (1985). We
contains 60% of the total predicted γ -ray flux when cuspy
∼ follow the procedure employed by Ackermann et al. (2012) to
NFW or Einasto profiles are assumed and 10% of the total
∼ excise the gas associated with the Smith Cloud from the Galactic
predicted flux from the cored Burkert model. Thus, this choice
gas distribution. Specifically, we remove gas in the region
of radius yields a conservative estimate for the total J-factor of
36 l 46 and 20 b 10 , which has a velocity with
! ! ! !
the Smith Cloud since the dark matter distribution may extend ◦ ◦ ◦ ◦
− −
respect to the local standard of rest in the range 70–125 km s 1
to several degrees. −
(Figure 1).11 The primary uncertainty in the conversion from
brightness temperature to column density comes from the
3. GALACTIC FOREGROUND MODELING
assumed spin temperature (T ) used to correct for the opacity
S
The observed foreground γ -ray emission from the region of the 21 cm line. We find that the gas density in the region
surrounding the Smith Cloud is dominated by π0-decay emis- of the Smith Cloud changes by <15% when the assumed spin
sion produced from cosmic rays interacting with the atomic temperature is changed from T 125 K to T 105 K (i.e.,
S S
= =
and molecular hydrogen gas of the Milky Way.8 The GALPROP the gas is optically thin). When analyzing the γ -ray data we
cosmic-ray propagation code can be used to model the dif- set T 125 K and find that this choice has little impact
S
=
fuse Galactic γ -ray foreground from processes such as inelastic on our results. We follow the procedure of Ackermann et al.
hadronic collisions, bremsstrahlung, and inverse-Compton scat- (2012) to trace the CO distribution from the 2 mm composite
tering.9 GALPROP accounts for effects such as diffusion, reac- survey of Dame et al. (2001). Due to the small CO content of
celeration, and energy loss via mechanisms such as synchrotron HVCs (Akeson & Blitz 1999), we do not alter the CO map
radiation (Strong & Moskalenko 1998; Strong et al. 2009). in the region of the Smith Cloud. The Galactic foreground
also contains a contribution from neutral gas that cannot be
8 The γ-ray emission from inelastic hadronic interactions is composed of
many processes, the most important of which being the production of π0, 10 http://fermi.gsfc.nasa.gov/ssc/data/access/lat/BackgroundModels.html
which decays primarily to γγ. 11 Gas with velocity >125kms 1 contributes less than 0.8% of the total
−
9 http://galprop.stanford.edu column density.
3
LAT Analysis
The Astrophysical Journal, 790:24 (7pp), 2014 July 20 Drlica-Wagner et al.
Counts Model Residual
• 5-year binned likelihood analysis from 500 MeV to 500 GeV over a 15˚x15˚ ROI
Figure 2. 15 15 ROI surrounding the Smith Cloud in the energy range from 500MeV to 500GeV. The gray contours represent the Hi column density associated
◦ ◦
×
with the Smith Cloud s(1urr1o02u0ncmd−i2ng< tNhHeI <S2m.7ith1 0C20locmu−d2) ,(Pwh7iRle EthPe o_vCer-LpElotAtedNc_irVcl1e 5sh)o.w s the 1◦ truncation radius for the dark matter profile. Left:
× ×
observed γ-ray counts map smoothed by a Gaussian kernel with standard deviation 0.1. Center: map of the background γ-ray emission model fit to the Fermi-LAT
◦
data including diffu•se aLnidkpeoliniht-loikoedba cmkgorodunedls .iRnicghlut: dthee sPo 2issFoGn pLro bsaobiulitrycoefsfin, dtihnget hceuobssteorvmed ndumifbfuersoef c Gouantlsaincetaicch pixel given the model prediction
expressed as a Gaussian significance.
foregrounds, and a local isotropic component modeled with a broken power-law.
(A color version of this figure is available in the online journal.)
• Set bin-by-bin limits on the gamma-ray flux from the Smith Cloud using a spatially-
traced by the comebixntaetinodneodf Hmioadnedl CoOf t(hsoe- cdaallrekd mdaarkttgears )a.nnihiAlautgiousnt s4)ig. nWael.s elect events from the P7REP CLEAN class in
We follow the procedure of Ackermann et al. (2012) to trace the energy range from 500 MeV to 500 GeV and within a 15
• No significant excess found for any of the spatial or spectral models tested ◦
the dark gas using the E(B V ) reddening maps of Schlegel radius of the Smith Cloud (l, b 38.67, 13.41). Extending
◦ ◦
(maximum TS = 4.7)
− = −
et al. (1998). We incorporate a dark gas correction into the this analysis to lower energies would translate to a minor
H i map after the Smith Cloud has been removed (Ackermann improvement in the sensitivity to low-mass dark matter models;
Alex Drlica-Wagner | Smith Cloud 9
et al. 2012).12 We note that our procedure for removing the gas however, below 500 MeV the rapidly changing effective area
content of the Smith Cloud is very similar to the method used to results in a stronger dependence on the spectral model assumed
remove gas associated with the Magellanic Clouds and M31 (see for the Smith Cloud. To reduce γ -ray contamination from the
Appendix B of Ackermann et al. 2012). Earth’s limb, we reject events with zenith angles larger than 100
◦
These observations of the Milky Way gas profile supple- and events collected during time periods when the magnitude of
mented by infrared observations of Galactic dust are input the rocking angle of the Fermi-LAT was greater than 52 .
◦
into the GALPROP code to model the diffuse γ -ray emission We perform a binned maximum likelihood analysis of the
corresponding to hadronic collisions, inverse-Compton scat- 15 15 region-of-interest (ROI) surrounding the Smith Cloud
◦ ◦
×
tering and bremsstrahlung radiation. To provide an accurate (Figure 2). We bin the Fermi-LAT data in this ROI into 0.1 pixels
◦
model for diffuse emission in the region of the Smith Cloud, and 24 logarithmically spaced bins of energy from 500 MeV to
we adopt the best-fit propagation parameters given by Trotta 500 GeV. We model the diffuse emission in this region using
et al. (2011), specifically a convectionless diffusion constant of the templates for the hadronic, bremsstrahlung, and inverse-
8.32 1028 cm2 s 1 at a momentum of 4 GeV, with a power- Compton emission derived in the previous section. Because the
−
×
law momentum scaling D(p) p0.31, a height for the diffusion hadronic and bremsstrahlung components are morphologically
∝
region of 5.4 kpc, and an Alfve´n velocity of 38.4 km s 1. These similar (both trace the interstellar gas in the Milky Way), we
−
parameters were inferred from a Bayesian analysis including the tie their relative normalizations in the γ -ray fit. In addition to
Fermi-LAT data as an input, and the resulting model is well-fit the diffuse Galactic foregrounds, the γ -ray data includes an
to the Galactic diffuse γ -ray emission at intermediate latitudes isotropic contribution from extragalactic γ -rays and charged
corresponding to the Smith Cloud. From this model we produce particle contamination. The spectrum of the isotropic γ -ray
energy-dependent maps for the γ -ray emission from hadronic background is routinely derived from a high-latitude ( b 10 )
! ◦
| |
emission, bremsstrahlung, and inverse-Compton scattering. In fit to the Fermi-LAT data, and is therefore dependent on the
principle, we would consider any alterations to the propaga- data selection and on the modeling of other γ -ray emission
tion parameters which are consistent with the local cosmic-ray components (i.e., the Galactic diffuse emission). It is difficult to
primary-to-secondary ratios measured by satellite and balloon derive the detailed spectrum of this component locally in the ROI
experiments. However, we find that this first attempt yields an of the Smith Cloud due to limited statistics at high energies and a
accurate model of the observed diffuse γ -ray emission in the morphological degeneracy with the inverse-Compton emission.
region of the Smith Cloud and no additional parameter-space Thus, we model the spectrum of the isotropic component with
scan is necessary. a broken power-law model which is simultaneously fit to the
Fermi-LAT data in the Smith Cloud ROI. While a broken power-
4. DATA ANALYSIS law model offers a reasonable fit to the Smith Cloud ROI, it
does not capture the detailed energy dependence of the residual
To search for excess γ -ray emission coincident with the
background. To quantify the impact of this simplification we
Smith Cloud, we select a data sample corresponding to the
also perform the analysis with the standard isotropic background
first five years of Fermi-LAT operation (2008 August 4 to 2013
model, iso_clean_v05.txt,13 and find that the results change by
12 The E(B V ) correction excludes the Smith Cloud due to its low
metallicity. − 13 http://fermi.gsfc.nasa.gov/ssc/data/access/lat/BackgroundModels.html
4