Long‐term evolution in the global distribution of solar wind speed and density fluctuations during 1997–2009

Interplanetary scintillation (IPS) observations made with the 327‐MHz multistation system of the Solar‐Terrestrial Environment Laboratory (STEL) are analyzed to investigate the global distribution of solar wind speed and density fluctuations (Δ N e ) and their evolution during 1997–2009. This study aims at elucidating the evolution of Δ N e distribution during the cycle 23 and subsequent extended minimum, which is useful for improving understanding of the heliospheric response to the peculiar solar activity. The computer‐assisted tomography (CAT) method is used in the present study to deconvolve the line‐of‐sight integration of STEL IPS observations. This CAT method enables retrieval of the quasi‐stationary large‐scale structure of the background solar wind. The results show that the high (low)‐latitude region is dominated by reduced (enhanced) Δ N e plasma, being closely associated with the fast (slow) solar wind. The solar wind speed data show a distinct change with solar activity, and an excellent positive (negative) correlation is revealed between the fast (slow) wind area and the polar field strength of the Sun. In contrast, the Δ N e data do not show such a solar cycle variation, but instead reveal a significant increase in the fractional area of low‐Δ N e region in 2004 preceded by a constant value with a small amount of fluctuation. This change is observed for all latitudes, distinctly after 2007 for low latitudes. Our finding is consistent with the long‐term variation of the solar wind density revealed from in situ measurements at the Earth orbit, if Δ N e  ∝  N e (where N e is the solar wind electron density), and also consistent with the coronal hole distribution during the last solar cycle. It is found that Δ N e is inversely correlated with the solar wind speed V . We obtain the best fit power law function Δ N e  ∝  V −0.36±0.14 for V  > 350 km/s, which is basically consistent with our earlier result. This fact suggests that the fractional density fluctuations Δ N e / N e are greater in the fast wind than in the slow wind. There is no systematic variation in the power law index or slope during 1997–2009 except for a null slope in 2000, which may be ascribed to an insufficient resolution of the CAT analysis. Thus, the inverse relation between Δ N e and V is regarded as a general rule for solar wind turbulence. The important point to note is that a marked drop in Δ N e occurs for the slow speed wind, V  < 350 km/s, particularly for 2004 and 2009. This fact may be attributed to different source conditions of the very‐low‐speed solar wind.