We use a new mass modelling method, GravSphere, to measure the central dark matter density profile of the Draco dwarf spheroidal galaxy. Draco's star formation shut down long ago, making it a prime candidate for hosting a 'pristine' dark matter cusp, unaffected by stellar feedback during galaxy formation. We first test GravSphere on a suite of tidally stripped mock 'Draco'-like dwarfs. We show that we are able to correctly infer the dark matter density profile of both cusped and cored mocks within our 95% confidence intervals. While we obtain only a weak inference on the logarithmic slope of these density profiles, we are able to obtain a robust inference of the amplitude of the inner dark matter density at 150pc, $\rho_{\rm DM}(150\,{\rm pc})$. We show that, combined with constraints on the density profile at larger radii, this is sufficient to distinguish a $\Lambda$ Cold Dark Matter ($\Lambda$CDM) cusp $-$ that has $\rho_{\rm DM}(150\,{\rm pc}) > 1.8 \times 10^8\,{\rm M}_\odot \,{\rm kpc}^{-3}$ $-$ from alternative dark matter models that have lower inner densities. We then apply GravSphere to the real Draco data. We find that Draco has an inner dark matter density of $\rho_{\rm DM}(150\,{\rm pc}) = 2.4_{-0.6}^{+0.5} \times 10^8\,{\rm M}_\odot \,{\rm kpc}^{-3}$, consistent with a $\Lambda$CDM cusp. Using a velocity independent SIDM model, calibrated on $\Lambda$SIDM cosmological simulations, we show that Draco's high central density gives an upper bound on the SIDM cross section of $\sigma/m < 0.57\,{\rm cm}^2\,{\rm g}^{-1}$ at 99% confidence. We conclude that the inner density of nearby dwarf galaxies like Draco provides a new and competitive probe of dark matter models.

Gravitational potential fluctuations driven by bursty star formation can kinematically 'heat up' dark matter at the centres of dwarf galaxies. A key prediction of such models is that, at a fixed dark matter halo mass, dwarfs with a higher stellar mass will have a lower central dark matter density. We use stellar kinematics and HI gas rotation curves to infer the inner dark matter densities of eight dwarf spheroidal and eight dwarf irregular galaxies with a wide range of star formation histories. For all galaxies, we estimate the dark matter density at a common radius of 150pc, $\rho_{\rm DM}(150\,\mathrm{pc})$. We find that our sample of dwarfs falls into two distinct classes. Those that stopped forming stars over 6Gyrs ago favour central densities $\rho_{\rm DM}(150\,\mathrm{pc})>10^8\,{\rm M}_\odot\,{\rm kpc}^{-3}$, consistent with cold dark matter cusps, while those with more extended star formation favour $\rho_{\rm DM}(150\,\mathrm{pc})<10^8\,{\rm M}_{\odot}\,{\rm kpc}^{-3}$, consistent with shallower dark matter cores. Using abundance matching to infer pre-infall halo masses, $M_{200}$, we show that this dichotomy is in excellent agreement with models in which dark matter is heated up by bursty star formation. In particular, we find that $\rho_{\rm DM}(150\,\mathrm{pc})$ steadily decreases with increasing stellar mass-to-halo mass ratio, $M_*/M_{200}$. Our results suggest that, to leading order, dark matter is a cold, collisionless, fluid that can be kinematically 'heated up' and moved around.

We derive the local dark matter density by applying the integrated Jeans equation method from Silverwood et al. (2016) to SDSS-SEGUE G-dwarf data processed and presented by B\"udenbender et al. (2015). We use the MultiNest Bayesian nested sampling software to fit a model for the baryon distribution, dark matter and tracer stars, including a model for the 'tilt term' that couples the vertical and radial motions, to the data. The $\alpha$-young population from B\"udenbender et al. (2015) yields the most reliable result of $\rho_{\rm DM} = 0.46^{+0.07}_{-0.09}\, {{\rm GeV\, cm}^{-3}} = 0.012^{+0.001}_{-0.002}\, {{\rm M}_\odot \, {\rm pc}^{-3}}$. Our analyses yield inconsistent results for the $\alpha$-young and $\alpha$-old data, pointing to problems in the tilt term and its modelling, the data itself, the assumption of a flat rotation curve, or the effects of disequilibria.

We present a new non-parametric Jeans code, GravSphere, that recovers the density $\rho(r)$ and velocity anisotropy $\beta(r)$ of spherical stellar systems, assuming only that they are in a steady-state. Using a large suite of mock data, we confirm that with only line-of-sight velocity data, GravSphere provides a good estimate of the density at the projected stellar half mass radius, $\rho(R_{1/2})$, but is not able to measure $\rho(r)$ or $\beta(r)$, even with 10,000 tracer stars. We then test three popular methods for breaking this $\rho-\beta$ degeneracy: using multiple populations with different $R_{1/2}$; using higher order `Virial Shape Parameters' (VSPs); and including proper motion data. We find that two populations provide an excellent recovery of $\rho(r)$ in-between their respective $R_{1/2}$. However, even with a total of $\sim 7,000$ tracers, we are not able to well-constrain $\beta(r)$ for either population. By contrast, using 1000 tracers with higher order VSPs we are able to measure $\rho(r)$ over the range $0.5 < r/R_{1/2} < 2$ and broadly constrain $\beta(r)$. Including proper motion data for all stars gives an even better performance, with $\rho$ and $\beta$ well-measured over the range $0.25 < r/R_{1/2} < 4$. Finally, we test GravSphere on a triaxial mock galaxy that has axis ratios typical of a merger remnant, $[1:0.8:0.6]$. In this case, GravSphere can become slightly biased. However, we find that when this occurs the data are poorly fit, allowing us to detect when such departures from spherical symmetry become problematic.

We present a new method for determining the local dark matter density using kinematic data for a population of tracer stars. The Jeans equation in the $z$-direction is integrated to yield an equation that gives the velocity dispersion as a function of the total mass density, tracer density, and the tilt term that describes the coupling of vertical and radial motions. We then fit a dark matter mass profile to tracer density and velocity dispersion data to derive credible regions on the vertical dark matter density profile. Our method avoids numerical differentiation, leading to lower numerical noise, and is able to deal with the tilt term while remaining one dimensional. In this study we present the method and perform initial tests on idealised mock data. We also demonstrate the importance of dealing with the tilt term for tracers that sample $\gtrsim 1$ kpc above the disc plane. If ignored, this results in a systematic underestimation of the dark matter density.