This week I cover whether primordial black holes and WIMPs can provide important contributions to the dark matter density at the same time, constraints on extensions to ΛCDM from KiDS-1000, and a more detailed analysis of the so-called “lensing is low” problem.
#1 2011.01930: Black Holes and WIMPs: All or Nothing or Something Else by Bernard Carr, Florian Kühnel, and Luca Visinelli (alphabetical)
Primordial black holes (PBHs) and weakly interacting massive particles (WIMPs) are among two of the leading theoretical dark matter (DM) candidates, though ultra-light scalars have recently been taking over. An interesting question is: can PBHs and WIMPs co-exist? This question has been addressed over the years in a number of important papers: 1003.3466 by Lacki and Beacom, 1405.5999 by Kohri et al., 1607.00612 by Eroshenko, 1712.06383 by Boucenna et al. (including Kühnel and Visinelli), 1901.08528 by Adamek et al., and 1905.01238 by Bertone et al. (there are certainly many other important papers I am missing - please feel free to send me a “cite me” email if that is the case :) ). The nuances of the conclusions vary from paper to paper, however the overall vibe is: it looks like hybrid models composed of PBHs and WIMPs might overall be problematic, because the clumping of WIMPs around PBHs would lead to enhanced annihilations, and consequently γ-ray signals which we should have seen. Most of these papers explored these hybrid PBHs-WIMPs scenarios for a rather restricted range of masses, either on the PBHs side, or on the WIMPs side. Given the recent renewed interest in PBHs, it is important to revisit the question of whether PBHs and WIMPs can co-exist, focusing on as wide a range of masses for the two components as possible.
This is the question Carr, Kühnel, and Visinelli try to address in this week’s paper, considering PBHs with masses in the range between 10^-18 and 10^15 M☉ and WIMPs with masses in the range between 10 GeV and 1 TeV, and also distinguishing between Galactic and extragalactic γ-ray backgrounds. The main results are given in their Fig. 2 for the case where WIMPs are the dominant component, and Fig. 3 for the reverse case (both figures are reproduced below). For most of the parameter space, one component would completely dominate over the other. In other words, if we were sure that the DM were one of the two, the other would be almost excluded. There are a few holes in parameter space where one can have an appreciable but <1 fraction of the DM in both WIMPs and PBHs: in this case, one needs at least a third DM candidate (the “something else” in the title). I strongly encourage interested readers to read Sections IVC, IVD, and V more carefully for further discussions which go beyond what I can cover here.
Overall, this was a very interesting paper which as far as I know for the first time explores the implications of mixed PBH-WIMP scenarios over such a wide range of parameter space for both. It is also somewhat of a follow-up on the earlier 2008.08077 by the same authors, which received a huge amount of media attention (see for instance here). One important conclusion of the present paper is that if PBHs with mass above 10^-9 M☉ comprise at least 1% of the DM, then the canonical WIMP scenario would be excluded. Now, I know there are a few people out there that are convinced deep down (the same way Trump is convinced the elections are being rigged) that LIGO has seen PBHs and that PBHs make up at least 1% of the DM. While I am not convinced (and in any case Nature couldn’t care less about my, or anybody else’s, opinion), it is the case that if this were true, then the standard WIMP scenario would be ruled out, and one might need the “something else” in the title if PBHs don’t make up 100% of the DM. In this sense, Section V is rather interesting as it reviews a number of hints for the existence of PBHs, which would lead to this conclusion. I stress however that before jumping to this conclusion one needs to be 101% sure that LIGO has indeed seen PBHs, something of which at present we have perhaps some hints, but absolutely no certainty about.
#2 2010.16416: KiDS-1000 Cosmology: constraints beyond flat ΛCDM by Tilman Tröster et al. (KiDS collaboration)
KiDS is the Kilo-Degree Survey, a cosmic shear survey which gets the “1000” bit of its name because of its imaging area being 1006 deg^2 (though the effective area after masking is 777 deg^2). In a series of recent papers (see more below), the KiDS collaboration provided constraints on ΛCDM and validated the robustness of their results against many observational systematics. However, as with results from any survey, an important question one should always ask is: “How stable are these results against an enlarged parameter space?” The choice of what additional parameters to vary is of course to some degree arbitrary, as it is impossible to check every possible conceivable extension, but in practice there is a subset of parameters one usually picks from, depending on the particular survey and whether it mostly probes early- or late-time physics (examples are neutrino masses, curvature, dark energy equation of state, effective number of neutrinos, lensing amplitude, and so on). In this week’s paper, Tröster and the rest of the KiDS team address this question and check how stable their previous results are against extension to ΛCDM.
This paper considers four extensions, allowing the curvature parameter, the sum of the neutrino masses, the dark energy equation of state, and the Hu-Sawicki parameter f_R0 to vary. For those who aren’t familiar with the latter, the Hu-Sawicki model is a modified gravity model of the f(R) type, rather popular against observers due to the fact that it has been coded up in a number of Boltzmann codes. The fifth force associated to the scalaron df(R)/dR is characterized by its Compton wavelength, whose value is controlled by f_R0. The analysis performed is the “usual” 3×2pt one using cosmic shear, galaxy clustering, and galaxy-galaxy lensing from KiDS-1000, BOSS, and 2dFLenS. In all four extensions, no significant deviations from ΛCDM are found. Of particular interest are the curvature parameter being constrained to Ωk~0.011±0.055 and the dark energy equation of state being constrained to w~-0.99±0.12. None of the extensions in questions were found to be preferred over ΛCDM in a statistical sense (although the converse is also true), and none of these was found to be able to solve the cosmic shear/S8 tension with Planck data, even when additional games were played such as fixing the amplitude of the primordial scalar perturbations As.
A few more useful info: this paper was the final paper from the KiDS-1000 data release, with previous papers being 2007.15362, 2007.15633, 2007.15635, 2007.01844, and 2007.01845. The KiDS online talks can be viewed here, while Shaun Hotchkiss’ Cosmology Talks Youtube channel has two excellent KiDS talks from Ben Giblin and Marika Asgari which I really recommend watching (and regardless you should subscribe to the channel as it is awesome!).
#3 2011.02377: On the halo-mass and radial scale dependence of the lensing is low effect by Johannes Lange et al.
I incidentally wrote about the “lensing is low” (LIL) tension/effect/problem (it doesn’t have a proper qualifier yet, unlike the Hubble tension) very recently in my Week 42 entry. The background here is that once galaxy-galaxy clustering properties are known, one should be able to predict the galaxy-galaxy lensing properties, armed with a galaxy-halo connection model (Halo Occupation Distribution models). The trouble is that, as shown in 1611.08606, current BOSS galaxy clustering data (and perhaps galaxy clustering data from other surveys as well?) appears to overpredict the galaxy-galaxy lensing signal compared to observations - thus, lensing is low (compared to expectations). As far as I know no convincing solution to the LIL problem exists in the literature, although various works have claimed that assembly bias (again, see my Week 42 entry) and baryonic feedback might help. To try and better understand what classes of solutions can address LIL, it is important to explore the mass- and radial scale-dependence of the LIL effect. This is precisely what Lange and collaborators do in this week’s paper.
Their results are very interesting in that they find no strong halo mass-, stellar mass-, and radial scale-dependence of the LIL effect. These results are interesting as they place constraints on what possible solutions to the LIL problem might be. For instance, the lack of scale-dependence severely limits the extent to which baryonic feedback or a different HOD model can solve the LIL problem. The paper also speculates on possible cosmological solutions to the problem. While the discussion is very generic and not tied to any specific model (though generic models that lower sigma8/Omegam, such as decaying dark matter, are mentioned), it is noted that it appears on general grounds very hard to solve this problem with a new cosmological model while a) keeping a good fit to CMB, BAO, and SNeIa data (a general problem even with the Hubble tension) and b) explaining the scale-dependence of the LIL effect, or more precisely lack thereof. It is certainly interesting to see more people getting interested in the LIL problem, and it is certainly possible that it might have connections to other cosmological tensions (such as the sigma8/Omegam tensions between Planck and cosmic shear, SZ clusters, and Ly-a data).