Radial gas flows and star formation in spiral galaxies

We present high-resolution (6″) observations of CO in seven nearby spiral galaxies made with the BIMA interferometer. Complementary single-dish data allow us to produce maps that recover flux on all spatial scales down to the interferometer resolution. The CO data are used in conjunction with previously published H I data to search for evidence of radial inflows in galaxy disks. Utilizing the Fourier decomposition of velocity fields developed by Schoenmakers et al. (1997), we construct simple models illustrating the effects of radial inflow, warps, and bar or spiral-arm streaming on the observed Fourier coefficients.; In all seven galaxies we find kinematic evidence for warps, elliptical streaming motions, or both. These futures lead to apparent inflow or outflow velocities of up to 50 km s-1, masking the effect of any net radial flows. In the inner disks of NGC 5033 and 5055, where the velocity fields show well-ordered rotation, we place an upper limit of 5 km s -1 on any axisymmetric radial flows. We then consider possible indirect evidence for radial flows, based on gas consumption arguments and radial abundance gradients. Applying extinction corrections that vary with gas density, we find gas depletion times of roughly 1 Gyr in the central regions and increasing with radius, similar to the situation in the Milky Way. Combined with estimates of the birthrate parameter, our results are difficult to reconcile with closed-box evolution models. The observed abundance gradients are also steeper than would be expected for a closed-box model, but could be accounted for with fairly modest (<1 km s-1) radial inflows.; Finally we consider the relationship between gas content and the star formation rate as traced by Halpha emission. Depending on the extinction corrections applied, our results support the existence of a Schmidt law with a power-law index of 1.1--1.7. An a1ternative star formation law, in which the star formation timescale is proportional to the orbital timescale, also matches the data if radially varying extinction corrections are used. The gravitational stability parameter Q does not appear to be related to star formation, at least when only the gas is considered; inclusion of the stellar component may be necessary to allow Q ∼ 1 for low gas fractions. We find that the ratio of H I to H 2 increases with galactocentric radius as R 1.5 and propose that it is determined largely by the interstellar pressure. Our results suggest that the interstellar pressure and metallicity control the formation of molecular clouds from H I, whereas star formation in molecular clouds occurs at a roughly constant rate per unit H2 mass.