arXiv:2011.11351v1 [astro-ph.GA] 23 Nov 2020 MNRAS 000, 000–000 (0000) Preprint 24 November 2020 Compiled using MNRAS L A T E X style file v3.0 NIHAO XXV: Convergence in the cusp-core transformation of cold dark matter haloes at high star formation thresholds Aaron A. Dutton 1⋆ , Tobias Buck 2 , Andrea V. Macci` o 1,3,4 , Keri L. Dixon 1,4 , Marvin Blank 1,4,5 , Aura Obreja 6 1 New York University Abu Dhabi, PO Box 129188, Saadiyat Island, Abu Dhabi, United Arab Emirates 2 Leibniz-Institut f¨ ur Astrophysik Potsdam (AIP), An der Sternwarte 16, D-14482 Potsdam, Germany 3 Max Planck Institut f¨ ur Astronomie, K¨ onigstuhl 17, 69117 Heidelberg, Germany 4 Center for Astro, Particle and Planetary Physics (CAP 3 ), New York University Abu Dhabi, United Arab Emirates 5 Institut f¨ ur Theoretische Physik und Astrophysik, Christian-Albrechts-Universit¨ at zu Kiel, Leibnizstr. 15, D-24118 Kiel, Germany 6 University Observatory Munich, Scheinerstraße 1, D-81679 Munich, Germany Accepted 2020 September 28. Received 2020 September 28; in original form 2020 May 10 ABSTRACT We use cosmological hydrodynamical galaxy formation simulations from the NIHAO project to investigate the response of cold dark matter (CDM) haloes to baryonic processes. Previous work has shown that the halo response is primarily a function of the ratio between galaxy stellar mass and total virial mass, and the density threshold above which gas is eligible to form stars, n[cm -3 ]. At low n all simulations in the literature agree that dwarf galaxy haloes are cuspy, but at high n ∼ > 100 there is no consensus. We trace halo contraction in dwarf galaxies with n ∼ > 100 reported in some previous simulations to insufficient spatial resolution. Provided the adopted star formation threshold is appropriate for the resolution of the simulation, we show that the halo response is remarkably stable for n ∼ > 5, up to the highest star formation threshold that we test, n = 500. This free parameter can be calibrated using the observed clustering of young stars. Simulations with low thresholds n  1 predict clustering that is too weak, while simulations with high star formation thresholds n ∼ > 5, are consistent with the observed clustering. Finally, we test the CDM predictions against the circular velocities of nearby dwarf galaxies. Low thresholds predict velocities that are too high, while simulations with n ∼ 10 provide a good match to the observations. We thus conclude that the CDM model provides a good description of the structure of galaxies on kpc scales provided the effects of baryons are properly captured. Key words: cosmology: theory – dark matter – galaxies: formation – galaxies: kine- matics and dynamics – galaxies: structure – methods: numerical 1 INTRODUCTION The structure of dark matter haloes on kiloparsec-scales provides a sensitive astrophysical test of the standard cold dark matter (CDM) paradigm, and more generally the na- ture of dark matter (e.g., Bullock & Boylan-Kolchin 2017). Through the use of dissipationless simulations, the structure of CDM haloes in the absence of baryons is well determined (e.g., Stadel et al. 2009; Dutton & Macci` o 2014). The dis- sipation of gas to the center of haloes is thought to only make the dark matter halo contract (Blumenthal et al. 1986; ⋆ dutton@nyu.edu Gnedin et al. 2004). However, other baryonic processes can cause the dark matter halo to expand: dynamical friction from infalling baryonic clumps (El-Zant et al. 2001), res- onances with galactic bars (Weinberg & Katz 2002), and multiple episodes of gas outflows (Read & Gilmore 2005; Pontzen & Governato 2012; Dutton et al. 2016b). Using 10 cosmological galaxy formation simula- tions from the MAGICC project (Stinson et al. 2013), Di Cintio et al. (2014a) found that the structure of CDM haloes, and hence the trade off between gas inflows and outflows, depends on the ratio between the galaxy stel- lar mass and the halo mass Mstar /M halo (which is pro- portional to the integrated star formation efficiency). At © 0000 The Authors