© Nature Publishing Group 1977 514 3 Stein_ W. Astrophysl. 148, 689(1967) 4 Colvm, J.D. Astrophys J. 190,515 (1974). 5 C_harugin, V. M. Astrophys. Space Sci. 37,449 (1975). Smgh, R. P. Planet Space Sci. 20,2073 (1972) V. V., Somayajulu & B. A. P. Planet Space Sci. 21, : Horita, R. E. Planet Space Sci. 20,409 (1972). 10 Bell, T. F. & Buneman, D. Phys. Rev. 133A, 1300 (1964) Kulsrud, R. & Pearce, W. Astrophys. J. 156,445 (1969). · 1\t!easuring solar wind velocity With spacecraft phase scintillations TH: discovery of interplanetary scintillations (IPS) of natural radw sources 1 _ has been followed by extensive multiple-station ?r IPS measurements of the solar wind velocity m regwns not yet probed by direct spacecraft"-"· Few measure- have, h?wever, been obtained within 40 R 0 because amphtude. or mtensity scintillations saturate when they are strong as IS the case near the Sun. This is unfortunate because the scarcity of wind velocity measurements in the acceleration close the Sun is one of the reasons why the construc- tion of a .umque p!cture of solar wind dynamics has not yet been poss1ble, desp1te numerous theoretical studies and direct spacecraft observations farther out 9 • 10 • The demonstration of the measurement of spacecraft phase scintillations with a cohe.rent radio system 11 paves the way for solar _velocity measurements with multiple- statiOn scmttllatwns. Although similar in principle to the IPS techmque, several factors make this new tool a more powerful and promising one. Because phase scintillations do not saturate when they are strong, observations can be carried out closer to the Sun than with amplitude scintillations. Spectral broadening has this same property 12 and we have obtained measurements at 2.3 GHz as_ cl?se 1.7 R0 with the Helios spacecraft. Extensive phase sc_mtiilatwn observations in the accleration region of the solar wmd are, therefore, possible using a dual-frequency S- and X-band coherent radio link 11 • For multiple-station observations the baselines are restricted to distances shorter than the correlation distance of the observed scintillations. For amplitude scintillations the correlation distance is of the order of the size of the fi;st Fresnel zone 1 a which is roughly 140 km at 2.3 GHz. On the other hand the correlation distance for phase scintillations is at least 10 6 km (refs 11, 13), a much wider range of baselines is possible. Particularly suited for these measurements are the three 64-m S- and X-band tracking antennae of NASA's Deep Space Net located at Goldstone (California), Canberra and Madrid. These antennae can be used for two-station observations with base- lines of the order of 10 4 km. Because the slowly varying phase scintillations are consider- ably stronger than the faster varying amplitude scintillations 1 a the signal-noise ratio (SNR) needed for long baseline measurements is much lower than that needed for short base- line amplitude measurements. The frequency spectrum of the phase scintillations 11 • 13 indiates that if the baseline distance is increased by a factor of 10 the SNR can be reduced by at least 20 dB. The choice of observing antennae for phase scintillations is, therefore, flexible, since a wide range of antenna sizes and recei':er noise temperatures can be considered, in addition to range of baseline distances. This is especially important smce It would be desirable to add observing antennae to the group of NASA 64-m antennae to permit three-station measure- ments. Also, for a given SNR, slowly varying phase scintillations yield measurements at larger solar elongations than faster amplitude scintillations. We have discussed some of the important advantages of measuring solar wind velocity with multiple-station phase scintillations. Current and future NASA deep space missions e'!-uipped with the coherent S/X radio system include Viking, Pwneer Venus and Mariner Jupiter/Saturn. Extensive obser- vati?ns this new technique during the numerous superior conjunctions of these spacecraft will provide much needed Nature Vol. 266 7 April 1977 measurements in the acceleration region of the solar wmd. These proposed observations will complement the large number _of measuremen!s that have been and are still being made With natural radto sources and will lead to a better of the dynamics and driving mechanisms of the solar wmd. This work was supported by NASA. Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91103 Received 13 December 1976; accepted 10 February 1977. RICHARD Woo Hewish, A., Scott, P. F. & Wills. D. Nature 203,1214-1217 (1964). Hew1sh, A., Denmson, P. A. & Pilkington, J.D. H. Nature 209, 1188 1189 (1966). 3 Denmson, P. A. & Hew ish, A. Nature 21.l, 343-346 ( 1967). R. D. & Little, L. T. Astr. Astroph_vs. IO, 310-316 ( 1971). V•tkevJtch, V. V. & Vlasov, V.I. Sol'iet Astr. 16,480-489 ( 1972). T., ShJbasak1, K. & Kakmuma, T. J. geophys. Res. 79, 7 Coles, W. A., Rickett! & ':'·H. in 5lo/ar Wind Three (ed. Russell, 8 C. T.) 351-367 ot Cahlorma, Los Angeles, 1974). 9 Coles, W. A. & R1ckett, H. J. J. geophys. Rev. _81, 4797-4799 (1976). Hundhausen, A. J. Coronal Expmts/011 and .)o/ar Wuul (Sprmgcr-Verhg New York, 1972). ' ' A. Rev. Geophys. Space Phys. 13,1049-105.1 (1975). 12 Woo, R., Yang, F. C., Yip, K. W. & Kendall, W. B. Astroph_v1. J. 208 (1976). Woo, R., Yang, F. C. & JsiHmaru, A. Astrophys . .!. 208 (1976). 13 Woo, R. Astrophys. J. 20I, 2.18-248 (1975). Tidal acceleration of the moon deduced from observations of artificial satellites DoUGLAS et a!. 1 demonstrated the existence of an apparent latitude dependence of tidal friction by determining disparate values of the second degree Love number (k e) from perturba- tions of the inclinations of the G EOS-1 and G EOS-2 satellites. Lambeck et a!. 2 correctly explained this phenomenon as being due to neglect of ocean tide perturbations. Parameter values for some ocean tide components have been obtained from several satellites 3 , but parameter values for the M 2 tide, the dominant (85 %) effect of the oceans on the tidal acceleration of the Moon, have not been published. Using an improved method for computing mean elements, we 4 obtained an observa- tion equation for the Me tide from the satellite 1967-92A. Applying this technique to the satellite GEOS-3, we now obtain an additional observation equation for the Me tide. As shown in ref. 2, solid and fluid tide effects on satellites cannot be separated, requiring assumption of the solid tide amplitude and phase parameters for a fluid tide solution. Assuming k 2 0.30, 8 2 = 0', and using the values of Lambeck 5 for the minor 0 1 and Ne contributions to 1i,, our fluid tide para- meters for theM 2 ocean tide yield the value of the tidal accelera- tion ti, -- 27.4±3 arc s (100 yr)- 2 , in excellent agreement with the value il, = --27.2+ 1.7 arcs (100 yr)- 2 obtained by Muller 6 from a combination of ancient and modern observations. These two values are lower than the value n, = -35±4 arcs (100 yr) 2 obtained from numerical ocean tide models 5 • Our assumption of a negligible solid tide phase angle is supported by a recent determination by J. T. Kuo (personal communication) that the phase angle obtained from a transcontinental network of tidal gravimetric stations is < I 0 • Changes 0.5° in solid tide phase angle change our result for the combined solid/fluid il, by no more than I arc s per (IOOyr) (ref. 2). C. C. GOAD B. C. DoUGLAS National Oceanic and Atmospheric Administration, National Ocean Survey, Geodetic Research and Development Laboratory, Rockville, Maryland 20852 Received 18 January; accepted 16' February 1977. 1 Douglas, B. C .• Klosko. S. M., Marsh. J. G. & Williamson, R. G. Ce/esfi(/1 Mechanics 10, 165-168 (1974). z Lambeck, K., Cazcnave, A. & Balmino, Rev. Geophys. Space Phys. 12,421--434 (1974).