JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 93, NO. A12, PAGES 14,465-14,477, DECEMBER 1, 1988 A MULTI-INSTRUMENT STUDY OF FLUX TRANSFER EVENT STRUCTURE ] C. J. Farrugia, R. P. Rijnbeek, ]'2 M. A. Saunders, ] D. J. Southwood D. J. Rodgers, 3 M. F. Smith,3, 4 C. P. Chaloner, 5 D. S. Hall, 5 P. J. Christiansen,6 and L. J. C. Woolliscroft 7 Abstract. We present simultaneous measure- ments at high temporal resolution of flux transfer events (FTEs) earthward of the magnetopause using four instruments on the Active Magnetospheric Particle Tracer Explorers (AMPTE) UKS spacecraft. The seven events in our study are distributed in local time from early dawn to midafternoon over a GSM latitude range of ~40 ø, from 24øN to ]7øS. A multilayered structure is revealed with systematic behavior in fields and charged particle populations. The plasma pressure varies in anticorrelation with the field pressure, the total pressure in general exceeding that of the surrounding medium. While it is not the prime concern of this paper to discuss the validity of models proposed for FTEs, the new, detailed experimental results it contains make it clear that the Russell-Elphic []978] model needs, at the very least, more detailed development. 7. Introduction Magnetic reconnection is a process important in astrophysics, solar physics, and laboratory plasma devices as well as in the Earth's magnetosphere. However, the magnetosphere affords the only opportunity to make detailed measurements of plasma and fields in situ. At the dayside magnetopause, reconnection appears to occur in both quasi-steady and time- dependent forms. Here we report on measurements made in flux transfer events (FTEs), which are thought to be the result of sporadic and localized reconnection [Russell and Elphic, ]978; Haerendel et al., ]978]. Over the last decade, ISEE 7 and 2 and Heos 2 spacecraft data have , , ,, 7 The Blackett Laboratory, Imperial College, London SW7 2BZ, England. 2Now at Institut f•r Weltraumforschung, Inffeldgasse ]2, A-8070 Graz, Austria. 3Mullard Space Science Laboratory, University College London, Holmbury St. Mary, Dorking, Surrey, RH5 6NT, England. 4Nowat Instrumentation and Space Research Division, Southwest Research Institute, San Antonio, TX 78284. 5 Rutherford Appleton Laboratory, Chilton, Didcot, Oxon, OX77 OQX, England. 6School of Mathematical and Physical Sciences, University of Sussex, Brighton, Sussex, BN7 9QH, England. 7physics Department,University of Sheffield, Sheffield S3 7RH, England. Copyright 7988 by the American Geophysical Union. Paper number 7A9272. 0748-0227/88/007A-9272505.00 been the source of many studies. New data on flux transfer events are now available from the spacecraft of the Active Magnetospheric Particle Tracer Explorers (AMPTE) mission, launched in August 7984. The new studies emerging [e.g., Southwood et al., ]986; LaBelle et al., ]987; Rijnbeek et al., 7987] suggest there remains much to be learnt. This report presents a multi-instrument analysis of FTEs detected when the AMPTE UKS spacecraft was situated earthward of the magnetopause. Seven events were examined, from three orbits (listed in Table ]). Our study extends most previous work on FTEs through the use of a multi-instrument data set and also through the high time resolution used. A brief account of instrumentation is given in the next section. In section 3, we commence our analysis with a detailed description of one event seen on October 28, ]984. The remainder of the events are discussed and compared with the October 28 event in section 4. Our concluding section summarizes our findings and discusses the implications for our understanding of the structure of magnetospheric FTEs. There is already a body of literature on FTEs. Theoretical works and interpretations are often based on the simple original "connected flux tube" model of Russell and Elphic [7978], but other proposals have been made, e.g., the multiple X line reconnection model of Lee and Fu [7985]. Conceptually, it is useful to have a framework to organize our observations, and we shall use the Russell-Elphic model thus, but the final conclusion from our data is that a much more elaborate model is needed. 2. Instrumentation The three-axis orthogonal flux gate magneto- meter on board the AMPTE UKS spacecraft [Southwood et al., ]985] transmits in science mode the components of the magnetic field at a rate of ]6 or 8 samples/s. We shall use 0.5-s averages in the magnetic field displays. At present, calibration yields the absolute magnetic field to an accuracy of 2 nT (W. A. C. Mier-Jedrzejowicz, private communication, 7986). The AMPTE UKS ion spectrometer [Coates et al., ]985] uses a spin- synchronized energy sweep to yield a complete three-dimensional energy distribution of positive ions (assumed to be protons) in the range ]0 eV/q to 20 keV/q once every spin period (-5 s). The three-dimensional electron spectro- meter [Shah et al.• 7985] measures in the energy range 72 eV to ]8 keV, an energy sweep being made once or twice per second. Here we use 2-s resolution. Neither charged particle instrument samples the full energy range of the ambient plasma, and at times discrepancies may appear in the derived ion and electron density signifying 14,465