-------------------------------- REPORT ON 1993 SNOWMASS WORKSHOP -------------------------------- GEM WORKING GROUP 2: PARTICLE ENTRY, BOUNDARY STRUCTURE, AND TRANSPORT co-chairs: P. T. Newell and L. C. Lee Working Group 2 debated 8 controversial topics. A review of discussions on topics 1,3,4 and 6 is given Pat Newell. A review of the remaining four topics (2,5,7 and 8) is given by L. C. Lee. Also given below are problems identified by Working Group 2. Topic 1: The Relationship Between Dayside Auroral Transients and Merging Our featured speakers were W. Denig and J. Minow, both of whom argued that at least a large subclass of poleward moving auroral forms such as first reported by Horwitz and Akasofu [1977] (and whose phenomenology was investigated and illuminated by Sandholt et al. [1986]) do indeed represent FTEs. These two speakers represent the views of most of the small community of researchers in the field of Poleward Moving Auroral Forms (PMAFs). In particular, both researchers cited the occurrence of these events on open field lines (as established by coincident satellite particle measurements from DMSP), their azimuthal velocity dependence on By, and their occurrence at times when Bz was southward (as measured directly, or more commonly, indirectly by an equatorward expansion of the auroral oval). Minow also reported new results by Fasel et al. [1993, not yet published], indicating that the distribution of inter-PFAF intervals -- the time between observing consecutive PFAFs -- had a mode of 3 minutes and a mean of 6 minutes. This was thought to be highly significant given the recent report of Lockwood and Wild [1993] that the inter-FTE distribution had a mode of 3 minutes and a median of 8 minutes. However Le and Russell presented results of an FTE survey using the same data set (ISEE magnetometers) which showed the inter-FTE distribution to have a mode of 8 minutes and a mean of 9.6 minutes. At least some of the apparent discrepancy owes to the threshold perturbation required to qualify as an FTE, with Le and Russell using the stricter requirements (10 nT peak in the bipolar signature). Although Le suggested that some of Lockwood and Wild's "little" FTEs might be wave activity or noise, no one in our working group could suggest a method for definitively addressing this issue. Denig also reported on simultaneous measurements of ion drifts (from DMSP) and all-sky images. He reported that the electric field structure seemed compatible with the Southwood [1987] model of FTEs. Denig's measurements showed that the largest of three events investigated had a total potential of no more than 4 kV associated with it (and possibly much less). Topic 2: Is the Reconnection Externally Driven or Spontaneous? Le reported that the FTEs observed in the dayside magnetopause by ISEE 1 and 2 spacecraft are mostly spontaneous and are not driven by the variations in solar wind and IMF. Variations in IMF do not occur before FTEs. However, Jacob and Cattell reported a few cases in which the northward turning of IMF triggers FTEs. Lockwood and Wild's observational data showed that FTEs have a wide spectrum of time scales. The time interval between two consecutive FTEs ranges from 3 minutes to 20 minutes. Fasel et al.'s ground observations of the poleward moving auroral forms (PMAFs) indicated multiple time scales, similar to those of FTEs. According to the 2-D MHD simulations by Fu and Lee, the 2-D reconnection, with formation of magnetic islands, has a time scale ~ 8 minutes at the dayside magnetopause. This may correspond to the large FTE's observed by Le which have a time scale ~ 8 minutes. The studies Nishida, Otto, and Lee et al. indicated that reconnections at multiple patches or in 3-D processes may provide a wide spectrum of FTE occurrence frequencies. The observations by Lockwood and Wild and by Fasel et al. may be related to the 3-D reconnection processes. Topic 3. Is Dayside Merging and Convection Primarily Continuous or Dominated by Bursts? Our featured speakers on this topic were O. de la Beaujardiere and A. Rodger. Both agreed that it was premature to give a final answer to the question of the relative importance of bursty and continuous merging. However both agreed that some component of continuous merging was required by simultaneous radar and satellite observations. Incoherent radar and DMSP observations have been used to infer the continual presence of the cusp over a several hour period [Waterman et al., 1993]. Rodger reported on unpublished results using DMSP satellites and coherent HF radar observations which lead to the same conclusion. Thus both speakers agree that continuous merging with embedded transients is occurring. However, although bursty merging alone as the sole form of merging can be ruled out, neither speaker felt that the relative significance of bursty and continuous merging can yet be determined. de la Beaujardiere also pointed out observations indicating that the local time extent of the merging gaps -- especially the nightside merging gap -- is much wider than often supposed in theoretical models. Rodger also pointed out that a surge in the merging rate can be accomplished in a number of ways, from the viewpoint of an ionospheric researcher. The merging line can become elongated; or the flow of flux through a constant merging line can increase; or the merging line can move equatorward (thus encompassing more flux in the open polar cap). Topic 4. Is Dayside Merging Primarily Antiparallel or Subsolar The featured speakers on this topic were P. Newell and C. Russell. Newell argued that observations on the penetration of magnetosheath plasma and fields into the magnetosphere -- and especially observations from the low-altitude particle cusp -- demonstrate that merging is primarily at high latitude (as would be implied by the antiparallel model). The following reasons were advanced: (1) The altitude of the merging site inferred from mid-altitude cusp ion pitch angle dispersion observations [Menietti and Burch, 1988] is 8-12 Re. (2) Observationlly, the high-energy cutoff of ions in the particle cusp drops promptly poleward of the equatorward edge. Quantitative modelling by Onsager et al. [1993] shows the implication: the ions are swimming upstream against the sheath flow shortly after merging, indicating a high-latitude merging site. (3) Low energy ions are less able to enter the winter cusp than the summer cusp, again implying a high-latitude merging site. (4) The local time behavior of the cusp as a function of By and Bz corroborates long-standing predictions of the antiparallel merging model. Newell also argued that the antiparallel merging model could explain the observed partial penetration of IMF By into the closed dayside magnetospheric field lines (which requires the closed field lines to shift in a sense opposite to that in which the open field lines move by magnetic tension) but that a subsolar merging model could not account for this effective By penetration. Chris Russell, while taking no definitive position, made several points. First, he pointed out that the question to be debated was not well posed, since merging is well established to occur at both the subsolar point and higher latitude. Instead one might ask how important are magnetic shear and magnetopause site in determining whether merging occurs. Secondly, he pointed out that FTEs (often regarded as a particular type of merging signature) appear to originate at the subsolar point. Russell presented numerous independent sets of measurements which indicate that there is something of a half-wave rectifier effect, with the magnetosphere much more strongly coupled to the IMF when Bz is southward than northward. Russell stated that any successful theory should explain this effect. He also argued that observational evidence existed which could most naturally be interpreted as paired flows away from a tilted merging line passing through the subsolar point. In the ensuing general discussion, most present and speaking thought that the half-wave rectifier effect did not help in distinguishing the two models, although there was some uncertainty on this point, and it was hoped that someone would follow up on addressing that issue. Topic 5: Formation of LLBL: Nishida talked about the formation of LLBL by the re-reconnection process. He also discussed the ionospheric signatures associated with this process. Richard presented simulations of particle entry into the magnetosphere for the northward IMF case. He calculated the particle trajectories based on the electromagnetic fields obtained from MHD simulations. A dawn-dusk asymmetry in the plasma density of the LLBL is found. Lyons presented a poster which showed that the LLBL can be formed on open field lines. Lee et al. proposed that the plasma can be transported through the magnetopause to form the boundary layer by kinetic Alfven waves. Discrete bundles of plasmas can move across the magnetic field into the magnetosphere due to the presence of field-aligned electric fields. In addition, they found that two counter-streaming electron beams can be present in the boundary layer. Topic 6. The Ionosphere as a Source for the LLBL Our featured speakers were J. Horwitz and D. Klumpar (Klumpar's viewgraphs were also presented by Horwitz). Horwitz presented observations of the Cleft Ion Fountain (CIF). Simulations and models indicate that the CIF flows naturally to the tail lobes and plasma sheet. A pathway to the LLBL may require large energization near the ionosphere to stay on LBLL field lines. Alternatively, ions might flow upward through a large portion of a "convection cycle" or even multiple cycles to arrive at the LLBL. Horwitz also presented some order of magnitude estimates showing that if a pathway existed, the plasmasphere could apparently supply almost any of the species observed in the LLBL. D. Klumpar presented (through Horwitz) observations that he and Fuselier have published in the literature. These high-altitude plasma composition observations of the LLBL show that electron beams apparently of ionospheric origin are common in the LLBL, and that at least for northward IMF significant amounts of moderate energy ions (~1 keV), especially O+, populate the LLBL after streaming upwards from the ionosphere. Topic 7: Impulsive Penetration: Ma, Lee, and Otto's 3-D simulations of the magnetic reconnection showed the generation of field-aligned currents and Alfven waves. They suggested that field-aligned currents and Alfven waves can also be generated by the impinging of the magnetosheath plasma with an extra momentum on the magnetopause and that the plasma penetration may be facilitated by the kinetic Alfven waves. Y. Song and Lysak presented a new concept about Alfvenon and discussed the plasma transport through meso-scale effects. Topic 8: Turbulence at the Magnetopause: P. Song showed that the plasma and fields are more turbulent at the magnetopause for the southward IMF cases than for northward cases. The whistler mode waves were observed at the magnetopause boundary layer. Drake showed simulations in which the current gradient drives whistler mode waves. Small-scale (~ 1-2 km) structures are present in the simulation. He also showed that the electron pressure term in the generalized Ohm's law can drastically change the nature of magnetic reconnection. Problems Identified by Working Group 2 -------------------------------------- Working Group 2 also suggested problems for future study. Four problem areas are briefly listed below (by L. C. Lee). (1) Turbulence at the Magnetopause Transition Layer: Identification of whistler waves and other waves from very small scales to the FTE scale by both satellite observations and computer simulations. (2) Boundary Layer Formation and Structure: (a) Carry out the simulations of re-reconnection to check if it can indeed occur. (b) Perform simulation and theoretical studies of plasma transport by kinetic Alfven waves; (c) Examine the transport process by meso-scale effects. (3) Impulsive Penetration: (a) Carry out 3-D MHD simulations of the impulsive penetration process and examine the generation of field-aligned currents and Alfven waves; (b) Examine the effects of kinetic Alfven waves and meso-scale structure on the plasma transport associated with impulsive penetration process. (4) Non-MHD Effects on Reconnection: (a) Examine the role of pressure term in generalized Ohm's law for magnetic reconnection; (b) Examine the whistler driven reconnection; (c) Carry out particle simulations to study the generalized Ohm's law near X line.