Datasets:

SouthernCrossAI
/

Arxiv

Dataset card Viewer Files Files and versions Community

Split (1)

train · 132k rows

Filename stringlengths 22 64	Paragraph stringlengths 8 5.57k
Processed_Probing_the_cool_ISM_in_galaxies_via_21cm_HI_absor.txt	Molecular Gas, Dust, and Star Formation in Galaxies Proceedings IAU Symposium No. 292, 2012 T. Wong & J. Ott, eds.
Processed_Probing_the_cool_ISM_in_galaxies_via_21cm_HI_absor.txt	Abstract. Recent targeted studies of associated H i absorption in radio galaxies are starting to map out the location, and potential cosmological evolution, of the cold gas in the host galaxies of Active Galactic Nuclei (AGN). The observed 21 cm absorption proﬁles often show two distinct spectral-line components: narrow, deep lines arising from cold gas in the extended disc of the galaxy, and broad, shallow lines from cold gas close to the AGN (e.g. Morganti et al. 2011). Here, we present results from a targeted search for associated H i absorption in the youngest and most recently-triggered radio AGN in the local universe (Allison et al. 2012b). So far, by using the recently commissioned Australia Telescope Compact Array Broadband Backend (CABB; Wilson et al. 2011), we have detected two new absorbers and one previously-known system. While two of these show both a broad, shallow component and a narrow, deep component (see Fig. 1), one of the new detections has only a single broad, shallow component. Interestingly, the host galaxies of the ﬁrst two detections are classiﬁed as gas-rich spirals, while the latter is an early-type galaxy. These detections were obtained using a spectral-line ﬁnding method, based on Bayesian inference, developed for future large-scale absorption surveys (Allison et al. 2012a).
Processed_Probing_the_cool_ISM_in_galaxies_via_21cm_HI_absor.txt	Figure 1. 21 cm absorption in the late-type host galaxies of two compact radio sources.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	A semi-empirical calculation of the 6s - 7s Stark amplitude α in Cs has been performed using the most accurate measurements and calculations of the electromagnetic amplitudes available. This is then used to extract the pa- rameters of the electroweak theory from experimental data. The results are: α = 269.0(1.3)a3 0, weak charge of Cs QW = −72.41(25)exp(80)theor, deviation from the Standard model S = −1.0(.3)exp(1.0)theor and limit on the mass of the extra Z-boson in SO(10) model MZx > 550GeV .
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	Here QW is the weak charge of the cesium nucleus and N is the number of neutrons.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	The method for ab initio calculations of EP N C that we used in [5] was based on an all- orders summation of the dominating diagrams of the many-body perturbation theory in the residual Coulomb interaction using a relativistic Hartree-Fock basis set and Green’s functions. This technique has been described in [5,8].
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	The error in the theoretical value was tested in many diﬀerent ways: by estimating the contribution of the unaccounted higher-order diagrams, by comparing the calculated and mea- sured values of the energy levels, the ﬁne and hyperﬁne structure intervals, the probabilities of electromagnetic transitions, etc. (see Ref. [5]). The result for the PNC amplitude practically did not change when we introduced factors into the correlation potential to ﬁt the energy lev- els (in imitation of the unaccounted higher-order diagrams). Important tests of our method included predictions of the spectrum [10] and electromagnetic transition amplitudes for the Fr atom [11], which is an analogue of Cs. Recently the positions of many energy levels [12] and some transition rates [13] of Fr were measured and found to be in excellent agreement with our predictions.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	PNC eﬀects were recently measured to an accuracy of about 1% [3] and found to be in good agreement with our predictions. This means that our estimates for the theoretical accuracy were correct and probably even too pessimistic. For example, in our ﬁrst calculation of the Fr energy levels [10] we claimed the accuracy of our predictions to be about 0.5% while the actual agreement with latter measurements was found to be 0.1%. The situation was similar for the electromagnetic transitions 6s-6p1/2 and 6s-6p3/2 in Cs (see below). These numerous tests give us ﬁrm ground to believe that the theoretical error in EP N C (2) indeed does not exceed 1%.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	As can be seen from (1) an accurate value of the vector transition polarizability β is also required for the interpretation of the PNC measurements. There are no direct experimen- tal measurements of β and so the value β = 27.0(2)a3 0 calculated in [16] was used for the interpretation of the PNC measurements. The theoretical ratio of the scalar transition polar- izability α = −268(3)a3 0 to β (α/β)theory = −9.93(14) [16] was in good agreement with the corresponding experimental value (α/β)exper = −9.9(1) [17] available at that time. Since then the ratio (α/β) was remeasured to a very high accuracy: (α/β) = −9.905(11) [18]. There have also been new, very precise measurements of the lifetimes of the 6p1/2 and 6p3/2 states of Cs [19]. This allows us to improve the accuracy in the determination of β, and thus in the interpretation of the PNC measurements, by incorporating the new experimental results into our calculations.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	It easy to see that β vanishes in the absence of the spin-orbit interaction, which splits energy levels and radial integrals. Thus, it is practically impossible to do accurate calculations of β using experimental results due to the strong cancelation between diﬀerent terms, which causes the relative statistical error to be larger. Therefore, we calculated the scalar transition polarizability α instead and used the measured ratio α/β to ﬁnd β. Note however that the calculation of α/β using theoretical radial integrals and experimental energies reproduces the experimental value for this ratio to an accuracy of about 1%.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	practically does not contribute to the error in the ﬁnal result. Our calculations of the 6s - 6p electromagnetic amplitudes were recently conﬁrmed to an accuracy of about 0.1% by very accurate experimental measurements [19]. The 6p - 7s amplitudes are also known from [20] with an accuracy of 0.5% and they agree with the theory.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	The values of the energy levels and spin-orbit splitting can be reproduced almost exactly in the numerical calculations by introducing factors into the correlation potential Σ (since the accuracy of the ab initio calculations is high, these factors are close to 1 anyway). The calculated matrix element h6s\|r\|6pi practically coincides with the value obtained from the accurate measurements of Ref. [19]. Therefore, we believe that the absolute accuracy in the calculation of the diﬀerence between the doublet radial integrals is always higher than the experimental accuracy (to avoid confusion we should note that we use Dirac wave functions, i.e. we do not expand in ξ while doing calculations). Thus we can take the experimental value of h6s\|r\|7p1/2i which is measured more accurately, and ﬁnd h6s\|r\|7p3/2i using the calculated diﬀerence h6s\|r\|7p3/2i − h6s\|r\|7p1/2i. Surprisingly, the result of this procedure gives precisely the result of the measurement of the h6s\|r\|7p3/2i amplitude, which formally has a larger error (1.8%). The ratio h6s\|r\|7p3/2i/h6s\|r\|7p1/2i also has a smaller experimental error (0.4%) than the error in h6s\|r\|7p3/2i [21]. Therefore, we may assume that the actual relative error in the h6s\|r\|7p3/2i is 0.7%, similar to that in h6s\|r\|7p1/2i. We use theoretical values of the h7s\|r\|7pi transition amplitudes since we believe that the expected theoretical error here ( 0.3%) is smaller than the experimental error. All higher transitions, including continuum and core electron transitions, were also calculated theoretically, even though their contribution was small (see below).
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	We used experimental energy levels from [23] and radial integrals from Table I to calculate the contributions of the 6p and 7p states. We used both experimental and theoretical data to select the “best values” of these integrals. Note that the errors in α(7p1/2) and α(7p3/2) are proportional and so we added them. When new data for electromagnetic amplitudes are available it will be easy to reﬁne this result by multiplying the corresponding term by the ratio of the new amplitude to the old one.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	Here we separated the error coming from the 6p and 7p E1-amplitudes from the error coming from all other sources, including the weak matrix elements and the amplitudes for transitions to the states above 7p. The error in the weak matrix elements can be roughly estimated using the deviation of the calculated hyperﬁne intervals from the experimental values since both the weak and hyperﬁne interactions are approximately proportional to the density of the electron wave function near the nucleus. Note that, the error from the E1-amplitudes exceeds the error in the theoretical values for the EP N C (2). To avoid confusion we should stress that the calculation in Ref. [5] was based on the Green’s function technique and does not contain partial cancelations of the diﬀerent terms which increase the error in the direct sum-over-state approach.
Processed_Polarizabilities_and_parity_non-conservation_in_th.txt	In conclusion we would like to stress that accurate measurements of the E1-amplitudes (Table I) are very desirable for an improvement of the interpretation of the PNC measurements in Cs. For α the most important improvement would be a more accurate value of the 6s - 7p amplitude. An improvement for the 7s - 7p amplitude is also very important because of the disagreement between theory and existing data.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	The lattice regularization of QCD provides us with the most systematic way of com- puting non-perturbative properties of hadrons directly from the ﬁrst principles of QCD. The recent rapid development of parallel computers has enabled us to start realistic and systematic simulations with dynamical quarks. In this paper, I report on the ﬁrst results from recent systematic studies on the lattice with dynamical quarks.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	We formulate QCD on a 4-dimensional hyper-cubic lattice with a ﬁnite lattice spacing a. When the lattice volume is also ﬁnite, the theory is completely ﬁnite and well-deﬁned, so that supercomputers can be used to calculate non-perturbative quantities. Numer- ous developments and ideas in the last two decades, both in algorithms for numerical computations and in the computer technology, have made lattice ﬁeld theory one of the most powerful tools in computing the non-perturbative properties of QCD. In particular, the development of parallel computers in the late 90’s enabled us to start realistic and systematic simulations of QCD.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	To obtain a reliable result for continuum physics, it is essential to have a good control of the systematic errors due to these extrapolations. A systematic approach is required. Because of the huge amount of computations required, major calculations have been made in the quenched approximation, in which the eﬀects of dynamical quark loops are ignored. A recent extensive study of quenched QCD by the CP-PACS Collaboration [1], however, has shown a limitation of the quenched approximation. As a next logical step, systematic “full QCD” simulations removing the quenched approximation have started. In this paper, I summarize the latest status of lattice QCD simulations, choosing “dynamical quark eﬀects” as the keyword.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	In Secs. 2 and 3, the latest results for the light hadron spectrum from lattice QCD are summarized. This provides us with the most direct test of QCD in the low-energy non- perturbative domain. In Sec. 4, lattice results for the mass of light quarks are discussed. Sec. 5 is devoted to the status of the U(1) problem on the lattice. In Sec. 6, lattice results for B meson decay constants are summarized. We ﬁnd that, in all of these physics, dynamical quarks have signiﬁcant eﬀects. A short conclusion is given in Sec. 7.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	The computation of the hadronic mass spectrum directly from the ﬁrst principles of If the lattice spectrum QCD is one of the most important tasks of lattice QCD [2,3]. reproduces the experimental data, this will provide us with the most convincing demon- stration of the validity of QCD as the fundamental theory for strong interactions in the low-energy non-perturbative domain. Furthermore, through these calculations, we can see how the lattice QCD describes the real physics and what is required to achieve high accu- racy suppressing lattice artifacts. Another motivation is a determination of fundamental parameters in QCD; see Sec. 4.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Because huge computer power is required to overcome the systematic errors discussed in Sec 1, major simulations have been done in the quenched approximation. With this approximation, the amount of computations is reduced by a factor of several hundreds.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Recently, the CP-PACS Collaboration performed a large-scale calculation of this with a precision signiﬁcantly better than previous studies [1]. Using a dedicated massively parallel computer CP-PACS developed at the University of Tsukuba [4], they studied four lattices 323 × 56 to 643 × 112 with lattice spacing in the range a ≈ 0.1–0.05 fm. The spacial lattice size was ﬁxed to be about 3 fm with which the ﬁnite size eﬀects are estimated to be maximally 0.5% in the spectrum. On each lattice, ﬁve quark masses, corresponding to the mass ratio of pseudo-scalar and vector mesons mPS/mV ≈ 0.75–0.4, were studied. The light u and d quarks were treated as degenerate. Their ﬁnal results in the continuum limit are summarized in Fig. 1. The light quark mass mud and the lattice scale a were ﬁxed using the experimental values for mπ and mρ as inputs, while the s quark mass was ﬁxed either by mK (K-input) or mφ (φ-input). See [1] for details of the simulation and extrapolations.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	From Fig. 1, we see that, although the global pattern of the experimental spectrum is reproduced, there remain systematic discrepancies of up to about 10% (7 standard deviations) between quenched QCD and experiment, for both choices of the input for the strange quark mass. Because other systematic errors are well under control, these discrepancies might be due to the quenched approximation.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	In the previous section, a limitation of the quenched approximation was made clear. The next logical step is to perform a “full QCD” calculation removing the quenched approximation, to see if these discrepancies disappear with dynamical quarks. However, a naive extension of quenched simulations is diﬃcult as the computational power required increases several hundred fold. Recently, there was much progress in improving lattice actions, with which lattice artifacts due to ﬁnite lattice spacing are reduced. Therefore, a coarser lattice may be used for continuum extrapolations with correspondingly less computer time.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	The CP-PACS Collaboration adopted the combination of an RG-improved gauge action and a “clover”-type improved Wilson quark action, and carried out the ﬁrst systematic investigation of full QCD. Two ﬂavors of dynamical quarks, to be identiﬁed with degen- erate u and d quarks, are simulated [5]. The heavier s quark was treated in the quenched approximation.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Figure 2. Coulomb coeﬃcient in full (ﬁlled circle) and quenched QCD (open circle). Solid line represents an estimate expected for the full QCD value in the chiral limit, and dashed lines its error.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Figure 3. Continuum extrapolation of vec- tor meson masses mφ and mK ∗ in full and quenched QCD, using the K meson mass as input [5].
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	coupling is diﬀerent between quenched and full QCD, we should see a diﬀerence in the Coulomb coeﬃcient α which represents the physics at µ ∼ 1–2 GeV [6]. In Fig. 2, results for α obtained on a 243 × 48 lattice are shown for full and quenched QCD at the same lattice string tension [5]. As expected, the full QCD results are larger than the quenched value at mPS/mV = ∞, and increase as the sea quark becomes lighter. The horizontal line in Fig. 2 is an expected value for α at mPS/mV = 0 in full QCD, estimated from the quenched value using the two-loop beta function with two ﬂavors of massless quarks. The full QCD values for light sea quarks are consistent with this estimate. Therefore, we do have dynamical quark eﬀects with the expected magnitude.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Now, we study implications of dynamical quarks in non-perturbative physics. Figure 3 shows the lattice spacing dependence of hadron masses from NF = 2 full QCD, compared with quenched masses. Plotted together are the results of a quenched calculation using the same improved action as in the full QCD calculation, with which one-to-one comparisons between quenched and full QCD results from calculations with similar parameters are made possible. We see that both quenched calculations lead to universal values for hadron spectrum in the continuum limit, and the discrepancies from experiment in quenched spectrum are much reduced by two ﬂavors of dynamical quarks. The remaining diﬀerence might be caused by the quenching of the s quark.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Figure 4. Continuum extrapolation of the average u and d quark mass mud and the s quark mass ms in the MS scheme at 2 GeV. ms is from the K-input. Filled symbols are for full QCD. Quenched results with the standard action and the improved action are shown with thin and thick open symbols, respectively.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	servables using the functional relation between these hadronic quantities and QCD pa- rameters. A lattice QCD determination of the hadron spectrum provides us with such functional relations without resorting to a phenomenological model.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	On the lattice, there exist several alternative deﬁnitions for quark mass. They should converge to a universal value in the continuum limit. The CP-PACS Collaboration ob- served that light quark masses determined from either the vector Ward identity (VWI) or the axial-vector Ward identity (AWI), while diﬀering at ﬁnite lattice spacings, actu- ally converge to a common value in the continuum limit [1]. See Fig. 4. However, in the quenched QCD, the resulting value for ms diﬀers by about 20% between K-input and φ-input, due to the discrepancy of quenched spectrum from experiment discussed in Sec. 2.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	the phenomenological value ≈ 150 MeV often used in hadron phenomenology, and almost saturating an estimate of the lower bound 90–100 MeV from QCD sum rules [8].
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Another important issue in QCD is the U(1) problem, that is to clarify the mechanism for a large η′ meson mass. Propagators of ﬂavor non-singlet meson consist of a loop of valence quark propagator, while propagators of ﬂavor singlet η′ meson have an additional contribution with two disconnected valence quark loops. The fact that η′ is much heavier than the corresponding non-singlet meson π means that the two-loop contribution should exactly cancel the π pole of the one-loop contribution, leaving the heavy η′ pole. This phenomenon is considered to be related with the anomalous violation of the ﬂavor singlet axial U(1) symmetry and topological structure of gauge ﬁeld conﬁgurations in QCD. This is a critical test of QCD and can be answered using lattice QCD.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Calculation of the two-loop contribution requires a large amount of computations. For this reason only limited results have been available. The ﬁrst systematic investigation in full QCD including continuum extrapolations was recently done by the CP-PACS Col- laboration [5]. In an approximation ignoring with the s¯s state, the ﬂavor-singlet u¯u + d ¯d meson mass was estimated as mu¯u+d ¯d = 863 ± 86 MeV. In the real world, the u¯u + d ¯d state mixes with the s¯s state to lead to η(547) and η′(958) mesons. Therefore, the result 863 ± 86 MeV which is slightly smaller than the experimental value 958 MeV is quite encouraging. Further studies, including an inspection of the mixing with the s¯s state as well as the study of topological structures, are under way.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	The study of the non-perturbative properties of QCD is important also to investigate the Weinberg-Salam theory for the electro-weak interactions [9]. In order to determine CKM parameters for quarks, we need to know QCD corrections (hadronic matrix elements, form factors, etc.) in weak decays and mixings. In particular, properties of heavy mesons consisting of a b quark are being studied intensively by the B factories. The precise determination of the decay constants and B-parameters for B(= Bd) and Bs mesons will lead to a strong constraint on the value of CKM parameters. The lattice calculation of these quantities are beginning to contribute in reducing the ambiguities in these studies [10].
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Major calculations have been done in the quenched approximation. Recently, full QCD calculations of the decay constants fB and fBs have been made by several groups. In this section, I concentrate on the dynamical quark eﬀects in these decay constants.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	On the lattice, simulation of the heavy b quark is not a trivial extension of light quark simulations, because mb ∼ 4 GeV is larger than the lattice cutoﬀ ∼ 1–4 GeV to date. Two methods have been developed to simulate heavy quarks directly. One is based on a non-relativistic eﬀective theory of QCD (NRQCD) deﬁned through an expansion in the inverse heavy quark mass [11]. Another employs a relativistic action and reinterpret it in terms of a non-relativistic Hamiltonian [12].
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Figure 5. Recent lattice results for the B meson decay constant from quenched simu- lations (NF = 0) and two-ﬂavor full QCD simulations (NF = 2), with various methods.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Figure 6. Constraints on the CKM param- eters ¯ρ and ¯η using lattice results. Allowed region from all the constraints is shaded by thick lines. See text for details.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	region, and “NRQCD” and “relativistic” are the results of the lattice NRQCD method and the relativistic method. Estimated statistical and systematic errors are included. Although the errors are still large, we can see that both methods are approximately con- sistent with each other. Comparing the quenched results (Nf = 0) and the full QCD Nf = 2 results, the latters seem to be about 10% larger than the former values. From this ﬁgure, I estimate fB = 200 ± 20 MeV. Also for the case of fBs, increase of a similar magnitude is observed.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	In Fig. 6, the implication of the lattice results for fB and fBs on the unitarity triangle of CKM parameters is summarized for ¯ρ = ρ(1 − λ2/2) and ¯η = η(1 − λ2/2) using the Wolfenstein parameterization (see [9] for details). Regions surrounded by two arcs are the allowed regions from each constraints, except for the two thin dashed lines denoted as ∆Ms which show the range of lower bounds for ¯ρ. In this ﬁgure, the quenched lattice results for BK, BB, etc. are used in the constraints from ǫ and \|Vub/Vcb\| [23], while two-ﬂavor full QCD results for fB and fBs are used for the constraints from ∆Md and ∆Ms. Allowed region from all the constraints is shaded by thick lines. The increase of fB due to the dynamical light quarks causes a shift of the allowed region from ∆Md towards positive ¯ρ, leading to a more severe constraint. The region shaded by thin lines is the allowed region from the conventional phenomenological estimates for hadronic matrix elements [9,23]. Hence, lattice results are contributing to reduce errors in the determination of CKM parameters.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Due to the large computer power required, major lattice simulations have been made in the quenched approximation, in which the dynamical sea quark eﬀects are ignored.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	Recently, it became possible to start realistic and systematic simulations of QCD with dynamical quarks, through the development of parallel computers. From the ﬁrst simu- lations, it turned out that dynamical quark eﬀects are quite important. The eﬀect of two ﬂavors of dynamical light u and d quarks is as large as about 20% in the values of light quark mass and about 10% in B meson decay constants. Both of the shifts has signiﬁcant implications to phenomenological studies of the standard model.
Processed_Hadronic_Properties_from_Lattice_QCD_with_Dynamica.txt	I thank the members of the CP-PACS Collaboration [24] for discussions. In particular, I am grateful to Hugh Shanahan for useful comments and preparation of several ﬁgures. I also thank Tony Thomas and other participants of the Conference for stimulative dis- cussions at Adelaide. This paper is in part supported by the Grants-in-Aid of Ministry of Education, Science and Culture (No. 10640248) and JSPS Research for Future Program.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	man, Sternberg & McKee 2017), and the strength of the Lyα absorption increases towards the centre of the galaxy (q.v. Borthakur et al 2015). But are we seeing halo gas, a disk- halo transition region, or cooling material settling to an outer ‘proto-disk’? The answer is unclear although statistical evi- dence is emerging of a rotating torus-like region in the outer- most parts of disk galaxies (q.v. Ho et al 2017).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	The most direct probe of the atomic hydrogen that typi- cally dominates the outer disks of z = 0 spiral galaxies is the 21-cm H I emission line. Until recently, the deepest H I ob- servations have rarely dipped below NH ≈ 1019.3 cm−2 at 5σ in a narrow velocity channel (e.g. Westmeier, Koribalski & Braun 2013). In an important unpublished experiment (see van Gorkom 1991), van Albada et al observed the isolated spiral galaxy NGC 3198 for 100 hours with the VLA in order to reach a column density sensitivity of NH ≈ 4 × 1018 cm−2, with the expectation of probing the H I emission to beyond R ∼ 60 kpc. To their surprise2, the 5 × 1019 cm−2 and 4 × 1018 cm−2 contours were coincident within their 1(cid:48) beam (2.7 kpc for an assumed distance of 9.4 Mpc, but see below). The impact of a cosmic ionizing background. Our interpreta- tion of “cool” gas in the extreme outer regions of galaxies is complicated by the highly uncertain (and controversial) cos- mic ionizing background (CIB). Following earlier discussions of the inﬂuence of the CIB on neutral hydrogen in galaxy disks (Sunyaev 1969; Felten & Bergeron 1969; Bochkarev & Sunyaev 1977), Maloney (1993) showed that there should be a critical column density Nc in H I disks at which the gas would go from largely neutral to largely ionized, that Nc would not depend strongly on either the galaxy parameters or the inten- sity of the ionizing background, and that this transition could plausibly explain the H I truncation observed in NGC 3198, provided the CIB is of order φi ∼ 104 phot cm−2 s−1.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	tions for understanding how gas gets into disk galaxies. Inferring the cosmic ionizing intensity. We brieﬂy review the broad range of estimates for the CIB intensity to emphasize that its value is highly uncertain. The present-day background ionizing intensity depends on environment and cosmic evolu- tion ( ´Cirkovi´c et al 1999; Maloney & Bland-Hawthorn 1999, 2001; Haardt & Madau 2012). Both of these contributions are only crudely constrained at the present time. Weymann et al (2001) set a 2σ upper limit of 8 mR3 for the Hα intensity of an opaque cloud illuminated by the CIB, although the uncertain- ties are dominated by the sky subtraction and the uncertain geometry of the cloud. Compared to Maloney’s (1993) value Jo = 4 × 10−23 erg cm−2 s−1 Hz−1 sr−1, and the J = 0.7Jo value adopted by Dove & Shull (1994), this is equivalent to an inten- sity at the Lyman limit of J = 0.4Jo. The theoretical estimate of the present-day background of Haardt & Madau (2012) is about a factor of two below this upper limit (J = 0.2Jo).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	In terms of the H ionization rate, the CIB intensity favored by Maloney (1993) corresponds to ΓH ≈ 10−13 s−1, while the z = 0 prediction of Madau & Haardt (2015) is ΓH = 2.3 × 10−14 s−1 (cf. Fig. 11 of Fumagalli et al 2017). However, a num- ber of papers on the low-redshift Lyα forest argued that their number was too small by a factor of a few (Kollmeier et al 2014; Shull et al 2015; Khaire & Srianand 2015; Viel et al 2017), and the revised model of Madau & Haardt (2015), up- dated with new quasar emissivities, predicts a z = 0 ionization rate of ΓH (cid:39) 6 × 10−14 s−1.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Very recently, Fumagalli et al (2017) have claimed detec- tion of CIB-induced Hα ﬂuorescence from the edge of the HI disk of the edge-on spiral NGC 7321. Performing photoion- ization modeling of the disk4 they arrive at an H ionization rate ΓH ∼ (6 − 8) × 10−14 s−1, which is in good agreement with the theoretical and low-z Lyα forest-derived numbers discussed above. It is, however, in considerable disagreement with the upper limit (and subsequent claimed 7σ detection5) of Adams et al (2011), which they report as ΓH = 2.0 × 10−14 s−1 for the same galaxy, even though the Adams et al (2011) upper limit to the Hα surface brightness is fully consistent with the detection of Fumagalli et al (2017). This should be taken as an indication of the importance of careful photoion- ization modeling in extracting φi from the Hα observations, and that caution should be exercised in accepting any par- ticular value at present; as with Weymann et al (2001), the uncertainty in the Fumagalli detection is largely systematic. The main goals of our study. So how are we to understand the well established existence of gas clouds − even down to NH ∼ 1017 cm−2 (Wolfe et al 2016) − far beyond the conﬁnes of the opaque disk? Can we detect these clouds directly in future deep H I and Hα studies? Do H I truncations even exist at all? These are the main questions addressed by this study.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Figure 1. (Left) Our computed curve of the dependence of column density NH with galactic radius (solid red line) for comparison with Dove & Shull (1994; dashed red line); we show the same comparison with Maloney (1993; dashed black line) and our model (solid black line). The adopted values of Jo (in units of erg cm−2 s−1 Hz−1 sr−1) are chosen to match the respective values used by the above authors. The dots indicate an exponential proﬁle for the neutral hydrogen emission in NGC 3198 prior to external photoionization. The proﬁle is shown over the limited radial range of 20 kpc to 40 kpc, as ﬁrst presented by the above authors. The gas ﬁlling fraction and covering fraction are everywhere f = 1, c = 1 respectively. (Right) Our computed Hα surface brightness Em as a function of galactic radius from a face-on disk (solid line) for comparison with Maloney (1993; dotted); this was not computed by Dove & Shull (1994).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Figure 2. The impact of different ﬁlling factors ( f ) and covering fractions (c) on the survival of neutral gas in the presence of external photoionization. The external ionizing ﬁeld is the same as in Fig. 1 with Jo taken from Maloney (1993). The red asterisks are from Pisano (2014); the black crosses indicate where the total H I column is 50 percent ionized. The solid and dashed curves (black, blue, green) correspond to f = 0.01, 0.1, 1 respectively as indicated. The dashed lines have matching ﬁll factors but show the trend in decreasing covering fraction, from c = 1 at 20 kpc to c = 0.1 at 80 kpc. Higher values of f increase the local gas density and make it more robust against being fully ionized. The black dots and green solid curve are the same as in Fig. 1, as shown by the shaded region; note that the radial scale is increased threefold. For guidance, H I column densities down to ∼ 1016 cm−2 can be detected in emission using radio receivers; QSO absorption lines reach down to much lower columns. The grey box shows the domain of the original study by Maloney (1993) in Fig. 1.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Figure 3. The predicted Hα surface brightness for the models presented in Fig. 2 where the disk is seen face-on. The calculation is for two-sided ionization given that the cosmic ionizing intensity illuminates both sides of the H I disk; the internal dust extinction is assumed to be negligible as expected in the outer parts of galaxies. For a disk inclined with angle i (where i = 0 is face on), the predicted values should be increased by sec i for i < 90◦; see text for a full discussion on this correction. For guidance, an Hα surface brightness below 1 mR is currently undetectable.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	in §5. In §6, we look forward to what will be possible in the next few years. At many scalelengths beyond the optical edge, disk stars are rarely observed and this may conceivably be the domain of the coherent cold ﬂow. But to map these struc- tures over large areas will require a different observational approach and large amounts of dedicated telescope time, as we discuss. In §7, we examine the broader implications of our work.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Simple arguments can be made that an exponential H I disk is a consequence of cooling in a spinning hot halo (Fall & Ef- stathiou 1980; Mo, Mao & White 2010). In support of this picture, the hot coronae (∼ 106 K) of several nearby high- mass disk galaxies have been detected through either stack- ing or radial binning of a direct x-ray image (Anderson & Bregman 2011; Dai et al 2012; Miller & Bregman 2013). A disk condensing out of the hot halo can extend to a signiﬁcant fraction of a virial radius if the hot halo has sufﬁcient angu- lar momentum. In these models, the start time for accretion moves outwards from the centre, and the accretion timescale increases as a function of the galactic radius. Using this ap- proach, Tepper-Garcia & Bland-Hawthorn (2017) produce a dynamically stable model of the Galaxy in line with the ob- served halo emissivity constraints. The atomic hydrogen disk has a scalelength of rd ≈ 7 kpc and NH ≈ 3 × 1017 cm−2 at a radius of r = 80 kpc. This is the non-ionized hydrogen proﬁle we adopt in this work which is broadly consistent with the ob- served H I proﬁle in the Galaxy (see below; Kalberla & Dedes 2008).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	H I observed in face-on disk galaxies and independent of ra- dius in the outer disks; this is also comparable to the expected magnitude of σg for gas in the warm neutral phase (but see the discussion below).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	What does the existence of such a critical column density imply for galactic H I disks? Notice that this is predicted to be a generic feature of H I disks, provided the assumptions about the gas distribution hold, and that Nc is not strongly dependent on either φi or the galaxy parameters. Note also that even if the (lower) Haardt & Madau (2012) CIB intensity is correct, Nc will drop by only about a factor of two. For NGC 3198, Maloney (1993) argued that the observed trunca- tion could be explained by photoionization, provided that the total hydrogen column continued to fall as extrapolated from the inner part of the disk (where the total and neutral hydrogen columns are nearly identical). Although reportedly the trun- cation in NGC 3198 is independent of azimuth (ruling out a warp, for example, as an explanation), only the major axis cut was available for analysis. There are noticeable differences in H I extent between the NE and SW sides of the disk, with the SW being more extended. However, such asymmetries are not really surprising due to the long dynamical timescales at these radii: the orbital period for gas at 30 kpc in NGC 3198 is more than 109 years. The Milky Way’s H I distribution, for exam- ple, is noticeably asymmetric, with more gas in the south than in the north (Kalberla & Dedes 2008). The observed neutral hydrogen distribution along the two sides of the major axis was reproduced with a model in which the deviations on the two sides from the same exponential fall-off in total H were less than 50%.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Hence the moderate asymmetry between the NE and SW major axis cuts is not really an issue for the photoionization model. However, it does illustrate an important point: even if all of the model assumptions are satisﬁed, whether photoion- ization by the CIB results in a sharp truncation of the disk depends on the shape of the underlying total hydrogen distri- bution. If this exhibits asymmetries or a change in behavior with increasing radius, this will be reﬂected in the observed H I distribution. A gas disk with a shallow radial proﬁle can exhibit an extended H I disk even if the column density has dropped below the critical column.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	lar processes; the very extended H I disk of NGC 3198 (more than 11 disk scalelengths) is precisely why it was chosen by van Albada et al for their deep imaging experiment.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Similarly, Kalberla & Dedes (2008) in their study of the Milky Way’s H I emission found that gas extended to R ∼ 60 kpc at low column densities, and argued that this was incon- sistent with the results of Maloney (1993). However, the ﬁt to their H I data shows an abrupt shallowing of the radial dis- tribution beyond 35 kpc and the vertical velocity dispersion required to explain the vertical extent is 74 km s−1. Whatever the physical mechanism responsible for this distribution, it is unavoidable that this gas is highly clumped, as the velocity dispersion cannot possibly represent thermal motion.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	The distribution and structure of the gas in the outer disks of galaxies is a complicated subject in which many uncertainties remain. The 21-cm H I emission-absorption study of the outer Milky Way — the galaxy for which we have the best data — of Dickey et al (2009) showed that the mean spin tempera- ture of the atomic hydrogen is nearly constant out to 25 kpc in radius, implying that the mix of cool and warm gas is sim- ilarly independent of both radius and height above the disk.7 Roughly 15-20% of the gas is in the cool phase, while the rest is warm. This implies that changes in the interstellar environ- ment (in particular, the drop in average ISM pressure) with R and z-height do not result in the disappearance of the cool phase. Both the cool and warm gas have scale heights that increase with radius. Strasser et al (2007) showed that the distribution of the cool gas in the outer Galaxy is not random, but is preferentially found in large, coherent structures that may be connected to the Galaxy’s spiral arms, suggesting that it may be gravitational perturbations that trigger the cooling of the cool atomic gas out of the warm neutral medium.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	To study the detailed structure of the Milky Way’s H I, Kalberla et al (2016) combined data from the Galactic Effelsberg-Bonn HI survey with the third release of the Parkes Galactic All Sky Survey and, using an unsharp masking tech- nique, found an abundance of small-scale cold ﬁlaments that they argue show the CNM is largely organized in sheets which are themselves embedded within a WNM with approximately ten times the column density of the CNM sheets (NH ∼ 1019 cm−2); sheets, rather than clumps, are what is expected when pressure effects dominate over self-gravity. To what extent the outer H I disk of the Milky Way resembles the outer disk of a galaxy such as NGC 3198 is an open question.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Even in an otherwise smooth distribution, the introduction of clumpiness into the gas distribution can have dramatic im- pact on the resulting radial H I distribution in the CIB pho- toionization model. In what follows we include a very simple prescription for clumping of the gas into the photoionization model and show how this can lead to extended disks at much lower column densities than in the unclumped models.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	same distance (Tully et al 2016). The Cosmicﬂows-3 cata- logue gives a distance modulus for NGC 3198 of 30.61±0.17 using both Cepheid variables and the Tully-Fisher relation; for NGC 2997, the Tully-Fisher distance modulus is for group member ESO 434-34 of 30.74±0.40 in close proximity. From their respective rotation curves, NGC 3198 is less massive than NGC 2997 by a small factor although their different in- clinations (edge on vs. face on) makes the factor uncertain. We ignore this difference below as the corrected scaleheight will involve an even smaller factor. For continuity with Mal- oney (1993) and Dove & Shull (1994), we assume both galax- ies have the same radial scaling at a distance of 9.4 Mpc. This scaling is 40% smaller than implied by the new distance mod- ulus from Cosmicﬂows-3. Thus there is a case for increasing all quoted radii throughout the paper by the same amount. But we refrain from such a large correction to maintain conti- nuity with the earlier work.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Both galaxies show gas disks that extend far beyond the op- tical emission. In NGC 3198, as discussed above, based on VLA observations, after a fairly smooth decline the disk ap- pears to truncate (somewhat unevenly between the two sides of the major axis) at radii between 33 and 40 kpc, where the projected column density NH ∼ 5 × 1019 cm−2; note that cor- rected for the galaxy’s substantial inclination (i ≈ 72◦), this is a face-on column density NH ∼ 2 × 1019 cm−2. In NGC 2997, which is at low inclination and hence geometric corrections are small, the disk extends much further out, to approximately 53 kpc at the NH ∼ 1.2 × 1018 cm−2 level, and any steepening in the slope of NH versus radius occurs at or beyond the limits of the data.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Our model is based on the original observations of NGC 3198 (Maloney 1993) although we extend the work to a dis- cussion of both NGC 2997 and the Galaxy. The galactic disk is modelled as having a spherical dark matter halo and ex- ponential stellar + gas disk. Azimuthal symmetry is assumed, thus restricting the problem to two dimensions in (R, Z) where R denotes galactocentric radius and Z vertical height from the midplane. Due to symmetry about the midplane, the prob- lem can be further restricted to Z ≥ 0. We scale up the pro- jected measures (e.g. Hα surface brightness) for two-sided photoionization. For the gas column densities relevant to the outermost H I disk, the disk opacity is negligible for the ex- pected metallicities.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	where a is the core radius (=7.92 kpc) and ρdm,0 (= 6.7 × 10−3 M(cid:12) pc−3) is the density at the galaxy’s centre (Dove & Shull 1994).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	The galactic model is also explored for a disk with dif- ferent ﬁlling factors f to account for a non-uniform den- sity within each column; this increases the local density and makes the gas more difﬁcult to ionise. We concentrate on f = (0.01, 0.1, 1) which covers the range of reasonable atomic hydrogen densities8 ((cid:104)nH(cid:105) = 0.1 − 10 cm−3). Because the pos- sible ways of implementing such a volume ﬁlling factor are practically unconstrained, and because our main focus is on illustrating the effects of such clumpiness, we have chosen an extremely simple form for f .
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	We also explore the impact of a non-uniform covering frac- tion of gas c such that c < 1 indicates there are holes in the gas disk when viewed face-on. Clumpiness is seen in simu- lations of the outermost reaches of H I disks (e.g. Stewart et al 2011). The covering fraction is a 2D (projected) quantity; the factors f and c can both be operating simultaneously and are entirely independent in our treatment. Thus, for c < 1, we do not attempt to conserve the gas mass at each radius. The necessary modiﬁcations are given in the next section.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	assuming isothermality of the disk at T = 104 K. Again, this is only a working approximation because other poorly con- strained sources (e.g. magnetic ﬁelds, cosmic rays) may con- tribute (Cox 2005). The density ρ(cid:48) now represents the local mass density in toto (stars, gas, dark matter). For our 1D cal- culation, we include the ﬁlling factor by decreasing the cal- culated gas scale height by f ( f ≤ 1). This compresses the gas distribution, raising the midplane density by a factor of 1/ f . This simpliﬁed treatment is intended only to illustrate how increasing the local gas densities from the smooth case will affect the truncation and extent of the disk.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Figure 4. The impact of different ﬁlling factors ( f ) and covering fractions (c) on the survival of neutral gas in the presence of external photoionization. The external ionizing ﬁeld is now J = 0.1Jo, i.e. 10% of the assumed cosmic intensity in Fig. 2. The solid curves correspond to f = 0.01, 0.1, 1 as indicated. The dashed lines have matching ﬁll factors but show the trend in decreasing covering fraction, from c = 1 at 20 kpc to c = 0.1 at 80 kpc. Higher values of f increase the local gas density and make it more robust against being fully ionized.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Figure 5. The predicted Hα surface brightness for the models presented in Fig. 4 where the disk is seen face-on. The calculation is for two-sided ionization given that the cosmic ionizing intensity illuminates both sides of the H I disk; the internal dust extinction is assumed to be negligible as expected in the outer parts of galaxies. For a disk inclined with angle i (where i = 0 is face on), the predicted values should be increased by sec i for i (cid:46) 90◦. For guidance, an Hα surface brightness below a few mR is currently undetectable.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	A more sophisticated analysis would be to treat radiation transfer in 3D through a porous (fractal) medium (e.g. Bland- Hawthorn et al 2015). Such a framework would be much more realistic, but also introduces a level of poorly-constrained complexity that is not justiﬁed for the presently available data. We return to this point in §6.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	to its minimum value at Rmax. The idea of a clumpy medium may appear to be inconsistent with equations 6 and 7. But here we assume that the clumpy medium is conﬁned by a hot diffuse phase transparent to radiation with negligible mass.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	where νo is the frequency of the Lyman limit, Jo is the speciﬁc intensity of radiation (in ergs cm-2 s-1 Hz-1 sr-1) at the Lyman limit and α ≈ 1 for a spectrum dominated by AGN.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	where, for species i in ionisation state j, σν,i j is the photoioni- sation cross-sectional area and νi j,0 is the minimum frequency of a photon which can ionise the species. Γsec refers to the ionisation rate of secondary ionisations, which are those re- sulting from high-energy electrons ionising neutral hydrogen. However, this component is negligible and can be ignored for the given electron temperature of Te = 104 K (Dove & Shull 1994).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	There are two sources of ionisation in the galactic disk — a direct and a diffuse component. The direct component refers to that from the isotropic cosmic background, and the diffuse component refers to the photons emitted by recombinations, namely those with sufﬁcient energy to ionise hydrogen.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	where τv is the optical depth between the layer and the point at which the radiation enters the disk, and µ is the cosine of the angle between the normal to the layer and the line of propagation into the layer (Maloney 1993). We have assumed that there are no additional sources of radiation in the extreme outer disk.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	The computational method comprises two phases: the ini- tial calculation of the ionisation state from the direct radia- tion component only, using the ‘on-the-spot’ approximation to treat the effect of recombinations (e.g., Osterbrock & Fer- land 2006), followed by a more accurate iterative equilibrium computation.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Case A recombination coefﬁcients are used in this step, as it is no longer assumed that all recombinations produce photons which are absorbed locally, and so ground-state recombina- tions must be considered in calculations (Osterbrock & Fer- land 2006).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	diffuse-radiation producing processes, the following routine is carried out, beginning with initial guesses for species den- sities: the photoionisation cross-sectional area is calculated. Optical depths for the relevant frequency are calculated using dτν = σi jni, j∆l, l denoting layer height, which are then used to ﬁnd the recombination rate in each layer using eqn. 14, using the previously calculated species densities. Then the diffuse radiation resulting from this process incident on each layer, including contributions from all other layers in the disk (in- cluding those on the other side of the midplane), is calculated using equation 13.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	After this has been carried out for each process, the result- ing diffuse radiation intensity values are used in eqn. 16 to compute the ionisation fraction of the species in each layer. H II, He II and He III number and column densities can then be computed. The whole process is repeated using these as new starting values, until the values converge to no more than a 10−3 discrepancy. The ﬁnal electron density is used to cal- culate the emission measure f n2 e∆l where ne is the electron density in each of m layers and ∆l is layer height, and where f is the ﬁlling factor. This process is repeated for each radial block.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	effectively giving the sum of the total H recombinations along the line of sight L where np is the proton number density. For a plasma at temperature 104 K, an emission measure of 1 cm−6 pc is equivalent to an Hα surface brightness µ(Hα) = 330 mR.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	We add a further note of caution: our predicted radial pro- ﬁles for the H I column density (NH) and the expected Hα sur- face brightness (µ(Hα)) assume a smooth, face-on spiral disk. Observed proﬁles are normally ‘boosted’ by the effects of disk inclination. For smooth, diffuse emission, the observed pro- ﬁles must be deprojected before a direct comparison can be made with our models (see below). If the emission is patchy, the situation is more complex and will require a more careful treatment.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Derived column densities. In Figs. 2 and 4, we revisit the cal- culations of Maloney (1993) and Dove & Shull (1994) over a much larger radial range. We also consider the geometric ﬁll- ing and covering factors f and c which allow us to examine the impact of clumpy gas in a consistent and controlled man- ner. As the local density rises through decreasing f , the gas is more able to survive photoionization at a ﬁxed Jo through faster recombination. The smallest ﬁll factor we consider is f = 0.01 as this produces clouds with densities at the upper range expected in outer disks (∼10 H atoms cm−3).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	It is important to note that there is still a critical column density present in all of the models: it has simply moved to lower column densities (and hence to larger radii) as a conse- quence of the ﬁlling factor; there is also a secondary effect of the transition getting shallower with decreasing f .
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	As discussed in Maloney (1993), the gas in NGC 3198 truncates at a radius of about 33 kpc to the NE. This be- haviour follows the c = 1, f = 1 (solid green) track closely in Fig. 2 (J = Jo). But such a cut-off is much less evident on the opposite side of the galaxy, where the gas extends (non- monotonically) to 40 kpc. As discussed above, this may sim- ply be a consequence of modest asymmetries in the total hy- drogen distribution within the disk, which are common in spi- ral disks and are not terribly surprising at large radii. Another possibility is that the asymmetry is the result of differences not just in total column density but in the degree of clumpi- ness of the atomic gas between the two extremes of the major axis.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Alternatively, it is possible the H I disk is shaped at least partly by some other process, such as ram pressure (although NGC 3198 is notably isolated and the asymmetries of the disk are quite modest). The NE side may be conﬁned by an ex- ternal medium; the SW side may represent the outer gas be- coming clumpy. This type of strong to mild asymmetric ram pressure conﬁnement of the outer H I disk is seen in dozens of nearby disks, including NGC 7421 (Ryder et al 1997), NGC 2276 (Gruendl et al 1993), the Large Magellanic Cloud (Kim et al 2003; Salem et al 2015) and in the highly-inclined Sculp- tor galaxy NGC 253 (Koribalski et al 1995).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Table 1 H I column density as a function of galactic radius R (kpc) for different combinations of J, f , c. Column 8: (J, f , c) = (Jo, 1, 1) is the original model ﬁrst presented by Maloney (1993). The columnar ionization fraction for all models increases with radius. The last line indicates the radius (r0.5) which corresponds to 50% ionization of the total hydrogen column. The 50% cut-off radius of Maloney (1993) is emboldened to emphasize the radial extent of all other models.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	jor axis cuts) of the radial H I distribution in NGC 2997 be- cause of the low resolution of the GBT observations. How- ever, his Table 1 provides the extent (area) of the H I disk as a function of column density. We have used these data to derive the red points plotted in Fig. 2; in effect, this is the symmetrized radial H I distribution for the modestly asym- metric disk. The plotted data match very well with the c = 1, f = 0.01 model. We note that at some level this agreement is fortuitous: the mass model used has not been matched to the rotation curve of NGC 2997 (although the effects, assuming a ﬂat rotation curve in this region, will be modest), and we have not attempted to match the H I distribution at smaller radii — we note also that the large beam of the GBT means that the inner disk is not well resolved. However, it does demonstrate that such a model can easily explain the extent of the disk and the slow roll-off of the neutral hydrogen. Now consider Fig. 4 where we present the same model parameters for a weaker cosmic ionizing intensity (J = 0.1Jo) at the low end of the likely CIB intensity. Again fortuitously, the red points fall along the c = 1, f = 0.1 (solid blue) track. This demonstrates the strong degeneracy between J and f . As the models show, there is a degeneracy between J, f and c.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	In Fig. 4, we consider a weak external ﬁeld, J = 0.1Jo. For f = 1, the crossing points are now approximately (29, 41, 46, 51, 60) kpc as a function of the column density. For our lowest f value, these move out to (30, 49, 61, 69, 79) kpc. The 50% ionization radius now occurs at r0.5 = 58 kpc such that we may expect to ﬁnd detectable H I gas twice as far out as indicated by the early NGC 3198 data, at least in principle.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	For all models, the columnar ionization fraction is increas- ing with radius. In the last row of Table 1 we list the radius at which the total column of hydrogen becomes 50% ionized. As indicated, the 50% ionization radius for Maloney (1993) is r0.5 ≈ 30 kpc at a column density of H I above 1019.5 cm−2. For all other models, r0.5 occurs at lower NH and is up to a factor of two further out in the most extreme cases.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	A broad distribution in f is a natural consequence of a tur- bulent medium (Lazarian & Pogosyan 2000). A spread in c can arise from increased clumping of the outer disk gas as the cold-ﬂow accretion channel becomes more dispersed, as observed in numerical simulations (Nuza et al 2014; Scan- napieco et al 2015; Crain et al 2017). Thus we consider the effect of slowly declining covering fraction as shown by the dashed lines in Figs. 2 and 4. As expected, a correction for a decreasing covering fraction shortens the crossing point dis- tances.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Patchy gas makes detection more difﬁcult generally. This can lead to both a lower covering fraction within the telescope beam, and a complex line proﬁle in velocity space. Simply stated, a covering fraction of c requires a telescope with an aperture that is 1/c times larger to observe the same signal under the same observing conditions. The deepest HI ob- servations require the largest apertures if the gas ﬁlls or par- tially ﬁlls the beam. But if the gas is in the form of dense ‘nuggets,’ higher sensitivity can be achieved by resolving out the clumps, e.g. spectroscopy with a very long baseline inter- ferometer (VLBI). Derived emission measures. In Figs. 3 and 5, we also present the predicted Hα surface brightnesses for the strong and weak external ﬁeld, respectively. The boosted signal from clump- ing, if relevant, is included in predictions. For the maximum cosmic intensity, we expect 20 mR at best, declining to a few mR at the limit of the data; for the weak external ﬁeld, the ex- pected value is an order of magnitude less. As already men- tioned, an Hα surface brightness of 1 mR is exceedingly faint, corresponding to 5.7 × 10−21 erg cm−2 s−1 arcsec−2 in more conventional units (Bland-Hawthorn et al 1995). This level of sensitivity is an order of magnitude fainter than what is possi- ble in conventional spectroscopy and needs some explanation. There are two approaches, the ﬁrst which average over most or all of the detector to achieve the detection limit, and the second from stacking many independent spectra.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	Bland-Hawthorn et al (1997) use Fabry-Perot ‘staring’ where the detector is azimuthally binned to produce a nar- row spectroscopic band (∼ 10Å). Apertures are typically in the range 5 arcminutes to 1 degree (e.g. Madsen et al 2001). The smaller apertures are well matched to the resolution of the biggest single-dish radio telescopes. Zhang et al (2016) use a very different method. They identify 0.5 million galax- ies with 7 million sight-line spectra which are co-added once shifted to a laboratory rest frame. They detect the presence of emission line gas out to 100 kpc and identify this as being pri- marily from the hot halo. In reality, a substantial contribution may come from an extended thick disk.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	where the cosmic ionizing spectrum Jν is deﬁned in eq. 8. This reduces to Jo = αhφi/π which establishes a simple rela- tionship between φi and Jo, both of which can be related easily to the Hα surface brightness µ, i.e.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	For individual galaxies, the deepest outer disk detection is from Bland-Hawthorn et al (1997). These signals at 50−100 mR are suprisingly strong, even after considering geometric corrections, and reﬂect either the presence of star formation activity at exceedingly low levels, or the warp of the outer disk causing the gas to be exposed to the inner radiation ﬁeld, i.e. self-ionization (Bland-Hawthorn et al 1998). Interest- ingly, there is new evidence of very weak star formation in the outer warped disk of one disk galaxy which makes it pos- sible to age-date the onset of the warp (Radburn-Smith et al 2016). Such activity may be affecting outer disk measure- ments in other galaxies (Christlein et al 2010; Fumagalli et al 2017).
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	The recent claim of a roughly 20 mR indirect detection of the CIB in the outer disk of UGC 3721 is not obviously con- sistent with the non-detections mentioned above, especially the Weymann et al (2001) result, even with due consideration for patchy emission or de-projection (Fumagalli et al 2017). The deepest non-detection experiments were undertaken with Fabry-Perots in ‘staring’ mode (Bland-Hawthorn et al 1994, 1995). In principle, the much larger RΩ product of a Fabry- Perot interferometer, where R is the spectral resolving power and Ω is the instrument solid angle, makes it far more sensi- tive to diffuse light than conventional spectrographs.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	However, the Fumagalli result is consistent with estimates for the CIB intensity based on the low-redshift Ly alpha forest and the most recent theoretical estimates of the background, as discussed in the Introduction. Given these disparate results, caution in interpreting current measurements is mandatory, but as we discuss in the next section, there is some prospect of a more deﬁnitive detection.
Processed_In_search_of_cool_flow_accretion_onto_galaxies_$-$.txt	binning in observations with our most advanced telescopes. Zhang et al (2016) detect Hα+[NII] emission at exceedingly faint levels (∼ 3 mR) in the 50 − 100 kpc radius bin in 7 mil- lion stacked spectra from the SDSS survey. This is an or- der of magnitude fainter than the levels observed by Bland- Hawthorn et al (1997) and Christlein et al (2010) in their survey of Hα emission at or beyond the H I edge of nearby galaxies, although the detection limit of the former method is of order 1 mR at 1σ (Bland-Hawthorn et al 1994; Madsen et al 2001).