ELECTRON-DENSITY POWER SPECTRUM IN THE LOCAL INTERSTELLAR-MEDIUM

Interstellar scintillation (ISS), fluctuations in the amplitude and phase of radio waves caused by scattering in the interstellar medium, is important as a diagnostic of interstellar plasma turbulence. ISS is also of interest because it is noise for other radio astronomical observations. The unifying concern is the power spectrum of the interstellar electron density. Here we use ISS observations through the nearby (less than or similar to 1 kpc) ISM to estimate the spectrum. From measurements of angular broadening of pulsars and extragalactic sources, decorrelation bandwidth of pulsars, refractive steering of features in pulsar dynamic spectra, dispersion measure fluctuations of pulsars, and refractive scintillation index measurements, we construct a composite structure function that is approximately power law over 2 x 10(6) m < scale < 10(13) m. The data are consistent with the structure function having a logarithmic slope versus baseline less than 2; thus there is a meaningful connection between scales in the radiowave fluctuation field and the scales in the electron density field causing the scattering. The data give an upper limit to the inner scale, l(0), less than or similar to 10(8) m and are consistent with much smaller values. We construct a composite electron density spectrum that is approximately power law over at least the approximate to 5 decade wavenumber range 10(-13) m(-1) < wavenumber < 10(-8) m(-1) and that may extend to higher wavenumbers. The average spectral index of the electron density over this wavenumber range is approximate to 3.7, very close to the value expected for a Kolmogorov process. The outer scale size, L(0), must be greater than or similar to 10(13) m (determined from dispersion measure fluctuations). When the ISS data are combined with measurements of differential Faraday rotation angle, and gradients in the average electron density, constraints can be put on the spectrum at much smaller wave numbers. The composite spectrum is consistent with a Kolmogorov-like power law over a huge range (10 or more decades) of spatial wavenumber with an inferred outer scale, L(0) greater than or similar to 10(18) m. This power-law subrange-expressed as ratio of outer to inner scales-is comparable to or larger than that of other naturally occurring turbulent fluids, such as the oceans or the solar wind. We outline some of the theories for generating and maintaining such a spectrum over this huge wavenumber range.