• A theoretical study of magnetosphere-ionosphere coupling processes

      Cao, Fei; Kan, J. R.; Akasofu, S-I.; Biswas, N.; Shaw, G.; Swift, D. (1991)
      Magnetosphere and ionosphere are coupled electrodynamically by waves, field-aligned currents and parallel electric fields. Several fundamental coupling processes are addressed in my thesis. It is shown that the Alfven wave is the dominant mode in transmitting field-aligned currents. Therefore, dynamic M-I coupling can be modeled by the Alfven wave bouncing between the ionosphere and the magnetospheric boundaries. The open magnetopause, separating the solar wind and the magnetosphere, behaves like a near perfect reflector to the Alfven wave because of the large solar wind inertia. At the plasma sheet, however, the reflection coefficient may extend over a wide range, depending on the location in the plasma sheet. As the Alfven wave propagates back and forth between the magnetosphere and ionosphere, the field-aligned current density increases dramatically at certain locations, especially near the head of the westward traveling surge, causing potential drops to develop along magnetic field lines. It is found that the existence of parallel potential drops can distort the global convection pattern and limit the upward field-aligned current. The magnetic reconnection at the dayside magnetopause is responsible for enhancing the convection in the magnetosphere, which subsequently propagates toward the ionosphere by the Alfven wave. The patchy and intermittent reconnection at the dayside magnetopause can be initiated by the 3-D tearing instability, leading to the isolated magnetic islands and X-line segments. The nonlinear evolution of tearing in terms of the magnetic island coalescence is also studied.
    • A theory of field-aligned current generation from the plasma sheet and the poleward expansion of aurora substorms

      Yamauchi, Masatoshi; Akasofu, Syun-Ichi (1990)
      This dissertation reports a study of the generation of field-aligned currents in the plasma sheet in terms of magnetosphere-ionosphere coupling. For the study, the plasma sheet and the ionosphere are treated as two-dimensional layers by height integration. In the magnetosphere between them, the Alfven wave transition time through this region is assumed to be zero. The ionospheric momentum is allowed to be transferred to the plasma sheet. Both linear analyses and numerical simulation are performed to study the field-aligned current generation. In the linear analysis, evolution from initial perturbations is studied. Zero order configurations are steady state without field-aligned currents. The field-aligned currents are treated as a perturbed quantity and linearly related with the other perturbed quantities. One result for the linear waves is that the magnetohydrodynamics (MHD) fast mode and Alfven mode are coupled through the ionospheric Hall current. The Hall current causes the dawn-dusk asymmetry: a westward-travelling wave is amplified on the region 1 current system, while an eastward-travelling wave is amplified elsewhere. The expansion phase of the magnetospheric substorm after the onset is numerically simulated on the near-earth plasma sheet. The inner edge of the plasma sheet is taken as the outflow boundary. As the initial condition, an enhanced earthward magnetospheric convection is assumed to cause a finite pressure increase at the inner edge of the plasma sheet. The numerical results are as follows. An MHD fast-mode wave is generated. It propagates tailward accompanied by the field-aligned currents. The wave propagation and the field-aligned currents account for the poleward expansion of the aurora and the region 1 field-aligned current during the expansion phase of the substorm. The region 1 field-aligned currents are linked with the dusk to dawn current on this wave, which is driven by the dynamo mechanism of the wave. The ionospheric Hall current causes asymmetry of the wave, and hence, of the field-aligned current distribution. This asymmetry accounts for the stronger field-aligned current in the premidnight sector.
    • An investigation of the dynamics of the mesopause: Fabry-Perot observations of winds and temperatures from nightglow emissions

      Conner, James F.; Smith, R. W. (1994)
      This work is a study of the behavior of tidal and planetary waves in the upper-middle atmosphere near the geographic south pole. This is accomplished with a characterization of the dynamic state of these motions. I used ground-based Fabry-Perot Spectrometer (FPS) measurements of the multiple-line, $P\sb1$(2)$\sb{c,d}$, nightglow emissions from the X$\sp2$II band of the neutral OH* molecule. I developed analytical techniques to determine a space and time distribution of spectral amplitudes and phases for the dynamic parameters of kinetic temperature and neutral wind in the OH* layer. Spectral analysis of the variations in this layer indicate the existence of two groups: a planetary wave group (periods of ${\sim}$1-10 days), with eastward phase progression, and a near semi-diurnal group (periods of ${\sim}$8-13 hours), with westward phase progression. Specific periods vary slightly for different years; this is most likely due to remote propagation conditions. Further separation of each group shows the wind oscillations exhibit wave-number one behavior with associated wave-number zero temperature oscillations, (with a few exceptions). The periodicities in the planetary group neutral wind motions are consistent with the model results of Salby, 1984, for propagation to high latitudes through realistic mean flows. The characterization of the dynamics of this layer has led to the discovery of a basic azimuthal asymmetry in the distribution of spectral amplitude for a given oscillation, that is, preferred azimuths. These preferred azimuths appear to be associated with changes in the direction, not the amplitude, of a cross-polar mean wind. This finding, in conjunction with the evanescence of some features, uncovered two cases of planetary wave dissipation. These occur when oscillations attempt to maintain their preferred alignment with a changing mean wind direction resulting in a decay of wind amplitude and a burst of thermal oscillation. Both cases occur at the same time. Coincident with these decays are enhancements in the wind and thermal energy of other, longer period, oscillations which share the same azimuthal preferences. Also coincident is an acceleration of the mean wind.
    • Characteristics of dayside auroral displays in relation to magnetospheric processes

      Minow, Joseph I.; Smith, Roger W. (1997)
      The use of dayside aurorae as a ground based monitor of magnetopause activity is explored in this thesis. The origin of diffuse (OI) 630.0 nm emissions in the midday auroral oval is considered first. Analysis of low altitude satellite records of precipitating charged particles within the cusp show an unstructured electron component that will produce a 0.5-1 kR 630.0 nm emission throughout the cusp. Distribution of the electrons is controlled by the requirement of charge neutrality in the cusp, predicting a diffuse 630.0 nm background even if the magnetosheath plasma is introduced into the magnetosphere in discrete merging events. Cusp electron fluxes also contain a structured component characterized by enhancements in the electron energy and energy flux over background values in narrow regions a few 10's of kilometers in width. These structured features are identified as the source of the transient midday arcs. An auroral model is developed to study the morphology of (OI) 630.0 nm auroral emissions produced by the transient arcs. The model demonstrates that a diffuse 630.0 nm background emission is produced by transient arcs due to the long lifetime of the O$(\sp1D)$ state. Two sources of diffuse 630.0 nm background emissions exist in the cusp which may originate in discrete merging events. The conclusion is that persistent 630.0 nm emissions cannot be interpreted as prima facie evidence for continuous particle transport from the magnetosheath across the magnetopause boundary and into the polar cusp. The second subject that is considered is the analysis of temporal and spatial variations of the diffuse 557.7 nm pulsating aurora in relation to the 630.0 nm dominated transient aurora. Temporal variations at the poleward boundary of the diffuse 557.7 nm aurora correlate with the formation of the 630.0 nm transient aurorae suggesting that the two events are related. The character of the auroral variations is consistent with the behavior of particle populations reported during satellite observations of flux transfer events near the dayside magnetopause. An interpretation of the events in terms of impulsive magnetic reconnection yields a new observation that relates the poleward moving transient auroral arcs in the midday sector to the flux transfer events.
    • Dependence of the ionospheric convection pattern on the conductivity and the southward IMF

      Shue, Jih-Hong; Kan, J. R.; Akasofu, S.-I.; Biswas, N. N.; Rees, M. H.; Weimer, D. R. (1993)
      Electric field measurements from the DE-2 satellite were used to determine the location of the convection reversal boundary and the potential around this boundary under a combination of interplanetary magnetic field (IMF) and auroral electrojet conditions. The electric potential is obtained by the integration of the electric fields. The convection reversal boundary is defined in this study as where the potential has its absolute maximum and minimum values. The data were sorted into 18 categories according to two levels of the negative IMF $B\sb{z},$ three ranges of IMF $B\sb{y},$ and two substorm phases. The data were fit with both continuous and discontinuous boundaries to get a functional representation of boundary potentials and locations. A simple model is constructed by solving the Laplace's equation in order to illustrate the obtained boundary potentials and locations. The results show that the enhanced electric field in the midnight sector is associated with an intense westward electrojet current. It can also be seen that the convection reversal boundary is found to be discontinuous near midnight. The discontinuous convection reversal boundary on the dayside is related to the merging near dayside cusp region. The discontinuous convection reversal boundary on the nightside is related to the conductivity enhancement. The intrusion of the dawn cell into the dusk cell is due to nonuniformity of the Hall conductivity in the ionosphere. Another model is constructed by solving the current continuity equation with field-aligned current and nonuniform conductivity added. It can be found that a secondary convection reversal, which is detached from the dusk-cell convection reversal, appears in the evening-midnight sector within the polar cap when the IMF $B\sb{y}$ is positive and the conductivity is nonuniform. This convection reversal is attributed to be created by the B $\times$ V dynamo. Also, the inclusion of the region 2 field-aligned current leads to an enhancement of the electric field in the region between the region 1 and region 2 currents.
    • Electron transport and optical emissions in the aurora

      Lummerzheim, Dirk; Rees, M. H.; Lee, L. C.; Olson, J. V.; Royer, T. C.; Stamnes, K. (1987)
      A one-dimensional, steady state auroral model is developed based on a linear electron transport calculation. A set of cross sections for electron neutral collisions describing elastic scattering, energy loss, and photon emission is compiled and used in conjunction with a discrete ordinate transport code. Calculated electron intensities are compared with in situ rocket measurements. Auroral optical emissions that result from direct electron impact on neutrals are calculated for synthetic and observed electron spectra. A systematic dependence of the brightness of auroral features on energy flux, characteristic energy, and atmospheric composition is found and parameterized. A method for interpreting the brightness and the ratio of brightnesses of certain auroral emissions in terms of the energy flux, characteristic energy, and relative oxygen density is described. Application of this method to auroral images acquired by nadir viewing instruments aboard a satellite is discussed and the distribution of energy flux, characteristic energy, and ionospheric conductances over the auroral oval is determined. Emissions that are suitable for analysing auroral spectra in terms of the atomic oxygen abundance in the auroral zone are identified.
    • Equilibrium structure and dynamics of near-earth plasma sheet during magnetospheric substorms

      Zhang, Lin; Otto, Antonius; Sentman, Davis; Smith, Roger; Watkins, Brenton (1997)
      A magnetofrictional method and MHD simulation are used to study MHD equilibria and dynamic evolution of the Earth's magnetosphere during a substorm growth phase. I suggest that the new "entropy anti-diffusion instability" associated with plasma transport across field lines leads to an enhanced entropy gradient and accelerates the formation of a thin current sheet during the final substorm growth phase. Based upon the MHD simulations with a pressure diffusion term, I confirm that entropy anti-diffusion instability can lead to a very thin current sheet with $B\sb{z} < 0.5nT$ and thickness $<$1000km in the near-earth magnetotail ($x \sim -8$ to $-20R\sb{e}$) during the growth phase of substorm. The formation of the thin current sheet can explain the observed explosive growth phase of substorms. In the study of magnetotail equilibrium configurations, it is found that the profile of the magnetic field strength B$\sb{z}$ component in the equatorial plane is mainly determined by the entropy $S(A)\ (S = pV\sp\gamma)$ on magnetic flux tubes. I obtain self-consistent equilibria of the Earth's magnetosphere with very strong lobe fields and a monotonically increasing $B\sb{z}$ component towards the Earth. It is also confirmed that an enhancement of the lobe flux favors the formation of a current sheet during the early substorm growth phase. However, my results do not support the notion that a critical amount of the lobe flux is required for a collapse of the tail current sheet.
    • Longwave Radiative Transfer In The Atmosphere: Model Development And Applications

      Delamere, Jennifer Simmons; Stamnes, Knut H. (2003)
      A FLexible Radiative Transfer Tool (FLRTT) has been developed to facilitate the construction of longwave, correlated k-distribution, radiative transfer models. The correlated k-distribution method is a technique which accelerates calculations of radiances, fluxes, and cooling rates in inhomogeneous atmospheres; therefore, correlated k-distribution models are appropriate for simulations of satellite radiances and inclusion into general circulation models. FLRTT was used to build two new rapid radiative transfer models, RRTM_HIRS and RRTM_v3.0, which maintain accuracy comparable to the line-by-line radiative transfer model LBLRTM. Iacono et al. [2003] evaluated upper tropospheric water vapor (UTWV) simulated by the National Center for Atmospheric Research Community Climate Model, CCM3, by comparing modeled, clear-sky brightness-temperatures to those observed from space by the High-resolution Radiation Sounder (HIRS). CCM3 was modified to utilize the rapid radiative transfer model RRTM and the separate satellite-radiance module, RRTM_HIRS, which calculates brightness temperatures in two HIRS channels. By incorporating these accurate radiative transfer models into CCM3, the longwave radiative transfer calculations have been removed as a significant source of error in the simulations. An important result of this study is that CCM3 exhibits moist and dry discrepancies in UTWV of 50% in particular climatic regions, which may be attributed to errors in the CCM3 dynamical schemes. RRTM_v3.0, an update of RRTM, is a rapid longwave radiative transfer appropriate for use in general circulation models. Fluxes calculated by RRTM_v3.0 agree with those computed by the LBLRTM to within 1.0 W/m2 at all levels, and the computed cooling rates agree to within 0.1 K/day and 0.3 K/day in the troposphere and stratosphere, respectively. This thesis also assessed and improved the modeling of clear-sky, longwave radiative fluxes at the Atmospheric Radiation Measurement Program North Slope of Alaska site by simultaneously addressing the specification of the atmosphere, radiometric measurements, and radiative transfer modeling. Consistent with findings from other field sites, the specification of the atmospheric water vapor is found to be a large source of uncertainty in modeled radiances and fluxes. Improvements in the specification of carbon dioxide optical depths within LBLRTM resulted, in part, from this analysis.
    • Observations Of Metal Concentrations In E-Region Sporadic Thin Layers Using Incoherent-Scatter Radar

      Suzuki, Nobuhiro; Watkins, Brenton (2006)
      This thesis has used incoherent-scatter radar data from the facility at Sondrestrom, Greenland to determine the ion mass values inside thin sporadic-E layers in the lower ionosphere. Metallic positively-charged ions of meteoric origin are deposited in the earth's upper atmosphere over a height range of about 85-120 km. Electric fields and neutral-gas (eg N2, O, O2) winds at high latitudes may produce convergent ion dynamics that results in the re-distribution of the background altitude distribution of the ions to form thin (1-3 km) high-density layers that are detectable with radar. A large database of experimental radar observations has been processed to determine ion mass values inside these thin ion layers. The range resolution of the radar was 600 meters that permitted mass determinations at several altitude steps within the layers. Near the lower edge of the layers the ion mass values were in the range 20-25 amu while at the top portion of the layers the mass values were generally in the range 30-40 amu. The numerical values are consistent with in-situ mass spectrometer data obtained by other researchers that suggest these layers are mainly composed of a mixture or Mg +, Si+, and Fe + ions. The small tendency for heavier ions to reside at the top portion of the layers is consistent with theory. The results have also found new evidence for the existence of complex-shaped multiple layers; the examples studied suggest similar ion mass values in different layers that in some cases are separated in altitude by several km.
    • Plasma dynamics of the Earth magnetopause-boundary layer and its coupling to the polar ionosphere

      Wei, Chang-Quan; Lee, Lou-Chuang; Deehr, C. S.; Swift, D. W.; Tape, W. R.; Watkins, B. J. (1991)
      In this thesis, the plasma dynamics of the Earth magnetopause-boundary layer and its coupling to the polar ionosphere are studied by using computer simulations. First, the plasma dynamics and structure of the magnetopause-boundary layer are studied by a two-dimensional incompressible magnetohydrodynamic simulation code. It is found that the Kelvin-Helmholtz instability with driven boundary conditions at the magnetopause can lead to the formation of plasma vortices observed in the magnetopause-boundary layer. In the later stage of development, a density plateau is formed in the central part of the boundary layer. Second, the coupling of plasma vortices formed in the boundary layer to the polar ionosphere is studied based on a magnetosphere-ionosphere coupling model. The finite ionospheric conductivity provides a dragging force to the plasma flow and leads to the decay of plasma vortices in the boundary layer. In the model, the ionospheric conductivity is allowed to be enhanced due to accelerated electrons precipitating in upward field-aligned current regions. The competing effect of the formation and decay of vortices leads to the formation of strong vortices only in limited regions. Several enhanced conductivity regions are formed along the post-noon auroral oval, which may account for the observed auroral bright spots. In addition, the evolution of localized plasma vortices, as well as magnetic flux ropes, along magnetic field lines is studied. The evolution leads to the generation of large-amplitude Alfven waves, which carry field-aligned currents and provide the link for the coupling of plasma vortices and magnetic flux ropes in the magnetosphere to the polar ionosphere.
    • Reversability Of Arctic Sea Ice Retreat - A Conceptual Multi-Scale Modeling Approach

      Mueller-Stoffels, Marc; Wackerbauer, Renate (2012)
      The ice-albedo feedback has been identified as an important factor in the decay of the Arctic sea ice cover in a warming climate. Mechanisms of transition from perennial ice cover to seasonal ice cover are discussed in the literature; the existence of a tipping point is disputed. A newly developed regular network model for energy exchange and phase transition of an ice covered ocean mixed layer is introduced. The existence of bistability, a key ingredient for irreversibility, on local and regional scales is explored. It is shown in a spatially confined model that the asymptotic behavior and the existence of a parameter region of bistability strongly depend on the albedo parametrization. The spatial dynamics of sea ice retreat are studied for a high resolution latitudinal model of the ocean mixed layer. This regional model suggests that sea ice retreat is reversible. It is shown that laterally driven melt of thick multi-year sea ice, and thus, ice-albedo feedback, is an important mechanism in the transition from perennial to seasonal ice cover at the pole. Results are used to interpret observed changes in the recent ice extent and ice volume record. It is shown that the effectiveness of ice-albedo feedback strongly depends on the existence of lateral heat transfer mechanisms in the ocean.
    • Role Of Conductivity Spatial Structure In Determining The Locations Of Sprite Initiation

      Tavares, Fernanda De Sao Sabbas; Jeffries, Martin O. (2003)
      Sprites are transient optical signatures of mesospheric electrical breakdown in response to lightning discharges. Multiple sprites are often observed to occur simultaneously, laterally displaced from the underlying causative cloud-to-ground (CG) lightning discharge. The causes of this lateral displacement are presently not understood. This dissertation investigates the role of neutral density perturbations in determining the locations of sprite initiation. The work was performed in three interrelated studies. (1) A detailed statistical study of the temporal-spatial relationships between sprites and the associated CG was performed for July 22, 1996. The distribution of sprite offsets relative to the underlying lightning had a mean of ~40 km. The distribution of sprite onset delays following the parent lightning had a mean of ~20--30 ms, consistent with theoretical estimates for the electron avalanche-to-streamer transition in the mesosphere. (2) A follow-up study for the same observations was performed to investigate the relationship of the sprites to convective activity in the underlying thunderstorm, using GOES-8 infrared imagery of cloud-top temperatures. The sprite generating thunderstorm was a Mesoscale Convective System (MCS). The maximum sprite and -CG production of the system were simultaneously reached at the time of maximum contiguous cloud cover of the coldest region, corresponding to the period of greatest convective activity of the system. Thunderstorm convective activity is a potential source of gravity waves and mesospheric turbulence. (3) Computer simulations of the temporal-spatial evolution of lightning-induced electric fields in a turbulent upper atmosphere were performed. The modeled turbulence in the simulations spanned the amplitude range 10% to 40% of the ambient background neutral density, with characteristic scale sizes of 2 km and 5 km, respectively. The results indicate that neutral density spatial structure, similar to observed turbulence in the mesosphere, facilitates electrical breakdown in isolated regions of density depletions at sprite initiation altitudes. These spatially distributed breakdown regions provide the seed electrons necessary for sprite generation, and may account for the observed sprite offsets.
    • Structure of reconnection layers in the magnetosphere

      Lin, Yu; Lee, Lou-Chuang; Hawkins, J. G.; Sentman, D. D.; Smith, R. W.; Swift, D. W. (1993)
      Magnetic reconnection can lead to the formation of observed boundary layers at the dayside magnetopause and in the nightside plasma sheet of the magnetosphere. In this thesis, the structure of these reconnection layers is studied by solving the one-dimensional Riemann problem for the evolution of a current sheet. Analytical method, resistive MHD simulations, and hybrid simulations are used. Based on the ideal MHD formulation, rotational discontinuities, slow shocks, slow expansion waves, and contact discontinuity are present in the dayside reconnection layer. Fast expansion waves are also present in the solution of the Riemann problem, but they quickly propagate out of the reconnection layer. Our study provides a coherent picture for the transition from the reconnection layer with two slow shocks in Petschek's model to the reconnection layer with a rotational discontinuity and a slow expansion wave in Levy et al.'s model. In the resistive MHD simulations, the rotational discontinuities are replaced by intermediate shocks or time-dependent intermediate shocks. In the hybrid simulations, the time-dependent intermediate shock quickly evolves to a steady rotational discontinuity, and the contact discontinuity does not exist. The magnetotail reconnection layer consists of two slow shocks. Hybrid simulations of slow shocks indicate that there exists a critical number, $M\sb{c}$, such that for slow shocks with an intermediate Mach number $M\sb{I} \ge M\sb{c}$, a large-amplitude rotational wavetrain is present in the downstream region. For slow shocks with $M\sb{I} < M\sb{c}$, the downstream wavetrain does not exist. Chaotic ion orbits in the downstream wave provide an efficient mechanism for ion heating and wave damping and explain the existence of the critical number $M\sb{c}$ in slow shocks.
    • The Dynamics And Morphology Of Sprites

      Moudry, Dana; Sentman, Dave (2003)
      In 1999 the University of Alaska Fairbanks fielded a 1000 fields-per-second intensified CCD camera to study sprites and associated upper atmospheric phenomena occurring above active thunderstorms as part of the NASA Sprites99 campaign. The exceptional clarity and definition obtained by this camera the night of August 18, 1999, provides the most detailed image record of these phenomena that has been obtained to date. The result of a frame-by-frame analysis of the data permits an orderly classification of upper atmospheric optical phenomena, and is the subject matter of this thesis. The images show that both elves and halos, which are diffuse emissions preceding sprites, are largely spatially unstructured. Observations of sprites initiating outside of main parts of halos, and without a halo, suggest sprites are initiated primarily from locations of atmospheric composition and density inhomogeneities. All sprites appear to start as tendrils descending from approximately 75 km altitude, and may form other dynamic or stationary features. Dynamic features include downward developing tendrils and upward developing branches. Stationary features include beads, columns, and diffuse "puffs," all of which have durations greater than 1 ms. Stationary sprite features are responsible for a significant fraction of the total optical emissions of sprites. Velocities of sprite tendrils were measured. After initial speeds of 106--107 m/s, sprite tendrils may slow to 105 m/s. Similarly, on some occasions the dim optical emission left behind by the descending tendrils may expand horizontally, with speeds on the order of 105 m/s. The volume excited by the sprite tendrils may rebrighten after 30--100 ms in the form of one of three different sprite after effects collectively termed "crawlers." A "smooth crawler" consists of several beads moving upward (~105 m/s) without a large vertical extent, with "smooth" dynamics at 1 ms timescale. "Embers" are bead-like forms which send a downward-propagating luminous structure towards the cloudtop at speeds of 106 m/s, and have irregular dynamics at 1 ms timescales. In TV-rate observations, the downward-propagating structure of an ember is averaged out and appears as a vertically-extended ribbon above the clouds. The third kind of crawler, so-called "palm tree," appears similar to an ember at TV-rates, but with a wider crown at top.
    • The generalized Ohm's law in collisionless magnetic reconnection

      Cai, Heng-Jin; Lee, L. C.; Jiang, T. M.; Morack, J. L.; Sentman, D. D.; Swift, D. W. (1995)
      Magnetic reconnection is an important process in space environments. As a result of magnetic reconnection, the magnetic field topology changes, which requires the breakdown of the frozen-in condition in ideal magnetofluids. In a collisional plasma, the resistivity associated with Coulomb collisions of charged particles is responsible for the breakdown of frozen-in condition. In a collisionless plasma, however, the cause of the breakdown of frozen-in condition remains unanswered. We address this problem by investigating the generalized Ohm's law and the force balance near magnetic neutral lines based on two-dimensional particle simulations. In a particle simulation with one active species, it is found that a weakly anisotropic and skewed velocity distribution is formed near the magnetic X line, leading to the presence of off-diagonal elements of plasma pressure tensor. The gradients of the off-diagonal pressure terms transport plasma momentum away from the X line to balance the reconnection electric field. The presence of the reconnection electric field results in the breakdown of frozen-in condition. The importance of both electron and ion off-diagonal pressure tensor terms in the generalized Ohm's law near neutral lines is further confirmed in full particle simulations. The generation of the off-diagonal pressure terms can be explained in terms of the thermal dispersion of particle motions and the response of particle distribution function in the electric and magnetic fields near the neutral lines. In the particle simulations, we also find the presence of a new dynamo process, in which a large amount of new magnetic flux near the magnetic O line is generated. This dynamo process is not allowed in resistive magnetofluids. However, in a collisionless plasma, the plasma inertia and momentum transport due to the off-diagonal plasma pressure terms can lead to E $\cdot$ J < 0 near the magnetic O line and make the dynamo process possible.