Gamma-Ray Sensitivity to Dark Matter Subhalo Modelling at High Latitudes

Round 1

Reviewer 1 Report

Well written article. Some comments for your consideration.

118 Maybe explain in one line why is this choice suitable for Fermi LAT catalogues

Fig. 1,2,3 Authors should probably add the label for each subhalo model at the top left corner of each of the 4 plots to make it easier for the reader

129 (e.g. blazars is too generic here) BL Lacs are hard, FSRQs soft.
Maybe define what is meant by harder in
spectral index numerical terms.

162 Remove - sign before E > 0.1 GeV

189 J notation is a bit confusing here, is this 4 models in sequence?. 7-8 (4-5) maybe 2 brackets with 4 values is better?.

181-208 Maybe break into 2 paragraphs

198 Please include reference for minimum subhalo mass to form a galaxy.

197-198 But authors cannot trace baryons with DMonly, right?. Dark matter mass does not imply dwarf galaxy formation. Without baryons there is no dwarf galaxy. I guess this is a chicken-egg question, can there be supermassive subhalos without baryons/stars that never become dwarf galaxies?.

Figure 4. There is a lot of information here, but color contrast makes it very hard to see all the subhalo. Maybe go with a lighter background and dark points. It is a pity to lose the information in these figures.

229 Should N_c be zero or is it one?.

295-296 "...predict many subhaloes to have significant angular extension". Maybe this need to be elaborated more, include a figure?.

Author Response

We thank the referee for his/her careful reading of the paper and comments. We will address them one-by-one below.

118 Maybe explain in one line why is this choice suitable for Fermi LAT catalogues

To clarify, we have added the following sentence: “The choice of this cut guarantees that the flux from DM annihilation (for e.g. cross-section values $\sim 10^{26} - 10^{-23} \, \rm  cm^{3}/s$ and masses $\sim$ 100 GeV is well below the \Fermi-LAT catalogues threshold, and therefore that we do not miss any detectable subhalo. As we highlight below, this cut also allows us to study what is the role, if any, of low-mass subhaloes.”

Fig. 1,2,3 Authors should probably add the label for each subhalo model at the top left corner of each of the 4 plots to make it easier for the reader

We thank the referee for the suggestion. We have included the model labelling in the plots.

129 (e.g. blazars is too generic here) BL Lacs are hard, FSRQs soft.
Maybe define what is meant by harder in
spectral index numerical terms.

The referee is correct. We specified that this is for BL Lacs and added the value of their typically hard spectral index. See added text: “(e.g.~BL Lacertae objects, spectral index $\sim$ 2.2)”.

162 Remove - sign before E > 0.1 GeV

Done. We put the text in parenthesis.

189 J notation is a bit confusing here, is this 4 models in sequence?. 7-8 (4-5) maybe 2 brackets with 4 values is better?.

We clarified the notation here. The two intervals 7-8 and 4-5 refer to the threshold for 3FGL and 2FHL and encompass all the 4 models. We modified as: “is about $7 \times 10^{17} \, (4 \times 10^{17}) \rm GeV^2/cm^{5}$ for all four models”, in order to explicitly show only the minimal J factor value. We hope this improve clarity.

181-208 Maybe break into 2 paragraphs

Done.

198 Please include reference for minimum subhalo mass to form a galaxy.

We have clarified, by changing the sentence “there is a minimum subhalo mass to form a galaxy, which is about $10^{7-8}$ \Msun.” Into “galaxy formation models agree that DM haloes with mass > 10^8 Msun are massive enough to systematically host galaxies”.

197-198 But authors cannot trace baryons with DMonly, right?. Dark matter mass does not imply dwarf galaxy formation. Without baryons there is no dwarf galaxy. I guess this is a chicken-egg question, can there be supermassive subhalos without baryons/stars that never become dwarf galaxies?.

The referee is correct: with DM only we cannot trace stars nor, strictly speaking, form dwarfs. Nevertheless, numerical simulations and galaxy formation models agree that DM halos with mass > 10^8 Msun are massive enough to systematically host galaxies (the minimal virial mass is likely below but the value is debated, see previous remark). It is in that sense that we refer to the most massive subhaloes (> 10^8) in our simulations as dwarf galaxies, even in the DMonly case.  

Figure 4. There is a lot of information here, but color contrast makes it very hard to see all the subhalo. Maybe go with a lighter background and dark points. It is a pity to lose the information in these figures.

We now use an inverted color scheme. We hope this improve the readability of the figure.

229 Should N_c be zero or is it one?

For the sensitivity reach calculation, N_c (the number of candidate for DM subhalos in Fermi-LAT data data) is indeed 0. However, the subhalo that effectively set the bound is the one with the highest J-factor (so we should consider 1 halo in the simulation to set the bound). For the sake of clarity, we added the following sentence: “(and therefore the subhalo to be used to set the bound is the one with the largest \Jf)”.

295-296 "...predict many subhaloes to have significant angular extension". Maybe this need to be elaborated more, include a figure?.

We quantified a little more in the text: “depending on the model , from a few up to tens of subhaloes with significant angular extension ($> 1^\deg$).”. However, we prefer not to provide any plot here since this will be the object of a dedicated analysis that we are currently performing.

Reviewer 2 Report

Dear authors,

I found this paper very interesting to read. It is very well written and the results and method used clearly explained. I report below minor comments and suggestions for your consideration.

Line 22-24. Not clear what the authors refer to when they write ‘However, deviations from cold DM predictions on small scales tantalise this paradigm and cast serious doubts on the weakly interacting massive particle (WIMP) hypothesis, the most scrutinised model for cold DM so far [2]. ’. I would ask the authors to specify this.

Line 36. ‘Have been proved’ is a bit strong in the sense that this does not come from a direct measurement but from simulations. So I would rephrase into ‘have been predicted’.

Line 77 several? I though only four.

Line 86 and 87: 'The total number of clumps is set to 300…' Why this assumption? I think the authors should justify this.

Line 92 and 93. ‘Through gravitational shocking, the disc is very efficient at disrupting most subhaloes  in the inner 30 kpc of the host galaxy. ‘ Does this apply only on the disk or also above and below the disk?

Line 118-120: ‘The choice of this cut originates from the specific scope of this paper: To show how many subhaloes would be detectable in the Fermi-LAT catalogues, and what is the role, if any, of low-mass subhaloes. ’ It is not clear from the text why sun halos with J>1e17 is appropriate for the scope of the paper.

Fig 1 and 2. This is a suggestion to make the comparison between plots and results faster. What about putting inside each plot the name of the simulation considered?

Author Response

We thank the referee for his/her careful reading of the paper and comments. We will address them one-by-one below.

Line 22-24. Not clear what the authors refer to when they write ‘However, deviations from cold DM predictions on small scales tantalise this paradigm and cast serious doubts on the weakly interacting massive particle (WIMP) hypothesis, the most scrutinised model for cold DM so far [2]. ’. I would ask the authors to specify this.

We realized an oversight in the sentence. We clarified and provided additional references. ‘However, null outcomes of weakly interacting massive particle (WIMP, [1]) searches in direct, indirect and collider experiments, together with deviations from cold DM predictions on small scales [1] challenge this paradigm and feed the interest for alternative DM scenarios.’

Line 36. ‘Have been proved’ is a bit strong in the sense that this does not come from a direct measurement but from simulations. So I would rephrase into ‘have been predicted’.

We respectfully disagree with the referee:  the fact that dwarfs are dynamically dominated objects is an observational evidence, based on the Jeans analysis of the dwarfs stellar content (although some modelling assumptions have to be adopted). We therefore prefer not to change “proved” into “predicted”.

Line 77 several? I though only four.

We corrected the text (several changed into four).

Line 86 and 87: 'The total number of clumps is set to 300…' Why this assumption? I think the authors should justify this.

We now are more explicit and added a reference by now saying “The total number of subhaloes is fixed by assuming 300 high-mass clumps with masses larger than $10^{8}\,\rm M_{\odot}$, as an upper bound to the Aquarius findings \cite{Springel:2008b}.”

Line 92 and 93. ‘Through gravitational shocking, the disc is very efficient at disrupting most subhaloes  in the inner 30 kpc of the host galaxy. ‘ Does this apply only on the disk or also above and below the disk?

This also applies to subhaloes above and below the disc because these also are on orbits which do cross the disc at some point, therefore they do experience gravitational shocking as well.

Line 118-120: ‘The choice of this cut originates from the specific scope of this paper: To show how many subhaloes would be detectable in the Fermi-LAT catalogues, and what is the role, if any, of low-mass subhaloes. ’ It is not clear from the text why sun halos with J>1e17 is appropriate for the scope of the paper.

To clarify, we have added the following sentence: “The choice of this cut guarantees that the flux from DM annihilation (for e.g. cross-section values $\sim 10^{26} - 10^{-23} \, \rm  cm^{3}/s$ and masses $\sim$ 100 GeV is well below the \Fermi-LAT catalogues thresholds, and therefore that we do not miss any possibly detectable subhalo. As we highlight below, this cut also allow us to study what is the role, if any, of low-mass subhaloes.”

Fig 1 and 2. This is a suggestion to make the comparison between plots and results faster. What about putting inside each plot the name of the simulation considered?

Thanks, we have followed this suggestion also provided by the other referee (also for fig. 3).