Relationship Of Storm-time Changes In Thermospheric Mass Density With Solar Wind-magnetosphere Coupling Parameters And Modeling

The investigation of thermospheric mass densities is one of the most popular research topics. The study of mass density variations during geomagnetic storms is not only important for the precise orbit determination and tracking of the LEO satellites, but also important for understanding the solar wind - magnetosphere ionosphere-thermopshere coupling processes.The thermosphere structure is mainly influenced by the solar radiation and the geomagnetic activities, where the solar radiation determines the background mass density and the long-term variation of the thermosphere. During geomagnetic storms, enhanced solar wind carrying southward IMF interacts with the Earth's magnetosphere, a large amount of solar wind energy impinges into the inner magnetosphere through reconnection. Large-scale solar wind electric field penertrates into the magnetosphere and even the low latitude ionosphere. Enhanced convection electric field is mapped to the high latitude ionosphere and drives the plasma convection there. Meanwhile the ionospheric conductivity is enhanced by the participation of the charged energetic particles into the high latitude ionosphere. Strong convection electric field and enhanced ionospheric conductivity produce enhanced Joule heating at the high latitude region. Other heating sources include the impact between the high-speed plasma and the neutral particles. The atmosphere is heated and expands, causing strong disturbances of the local densities. The disturbances then propagate equatorward in the form of large-scale circulation and traveling atmospheric disturbances. At low latitudes, the coulomb impact between enhanced ring current ions and electrons also produces heat, which leads to the heating of the thermosphere on one hand, and the ionization of the neutral gas, hence the change of neutral densities on the other hand. These physical processes lead to the strong disturbances of the thermopsheric mass density during storm-time, sometimes the disturbances reach to 200%-400% of the quiet values.This dissertation makes use of the thermopsheric mass density measurements from the LEO satellites CHAMP and GRACE to study the storm-time density variations during geomagnetic storms in the decline phase of the Solar Cycle 23 from 2002 to 2005. At high latitudes the polar cap density anomaly is found during storm-time. Based on the study of some typical anomaly events, explanations for possible driving mechanism are provided. A statistical study is performed to analyse the salient features of the anomaly. At mid and low latitudes the dependence of the storm-time mass density disturbances on the merging electric field is studied. The optimal delay of the density behind the merging electric field is determined by correlation analysis, then the empirical relation between the thermopsheric density and the merging electric field is build by linear regression. A linear density predicting model is established for the storm-time thermospheric density at altitudes around 400 km and 500 km.The main studies performed and results obtained in this dissertation can be concluded as follows:1. Using CHAMP density data of four years from 2002 to 2005, the storm-time density disturbances at high latitudes are studied. Practically in every geomagnetic storm with Dst(min)≤-100 nT the polar cap density anomaly is found, i.e. the mass density enhancement reaches 120% of the background values poleward±78°Mlat. The statistical study shows that the polar cap density anomaly is characterized by:a) The anomalies are of medium scale size (typically<900 km) and seem to have a short dwell time (<1.5 h) in the polar cap.b) The ratio of density enhancement, on average by a factor of 2, shows little dependence on the solar flux level (F10.7). The peak density in the Northern Hemisphere is by a factor of 1.4 larger than in the southern. However, the relative enhancement is comparable in both hemispheres.c) Mass density anomalies in the polar cap occur under all interplanetary magnetic field (IMF) directions. They show some preference for a southward IMF orientation or for strong northward. No clear local time dependence on IMF orientation is observed.d) About half of the anomalies are accompanied by strong FAC for northward IMF.e) Some density anomalies are accompanied by intense ion upflows, which could be a possible cause for the density enhancement.The local mass density anomaly in the polar cap during storm time is a complex phenomenon. There are probably several different mechanisms responsible for their formation. The possible driving mechanisms are Joule heating, the trans polar propagation of TADs across the polar cap and the interaction with ion upflow. 2. Based on the four-year observations of CHAMP, the dependence of low and mid latitude densities on the merging electric field Em during geomagnetic storms is investigated. Altogether 32 major storms are included in a statistical study. A linear empirical model is established for the relationship between thermopsheric density at 400 km and the merging electric field. The findings in this study can be concluded as:a) The properly preprocessed merging electric field tracks the density changes in all phases of magnetic storms sufficiently well. It is possible to utilize it to predict density variations during a storm. In order to account for the memory effect of the thermosphere a 3 h integrated Em value is applied.b) A truncation of Em for strong solar wind driving, according to the saturation of the cross-polar cap potential,φpc, does not reflect the variation of thermospheric density well. Only if the full Em swing is considered the simple linear relation holds also for super storms.c) The delay time of density changes behind Em depends on various factors. Basically, it depends on geomagnetic latitude, local time and the storm intensity. The delay time of the low latitude density with respect to Em is around 4.5 h, showing no clear dependence on MLT. The delay time of the mid latitude density varies between 1.5 to 6 h, where the nightside delay time is longer than the dayside.The mean delay of orbit-averaged density is almost the same as the low latitude delay.d) The storm-induced mass density perturbation at constant altitude has been found to be an additive enhancement on top of the quiet-time density. It can be expressed as linear relation between the density and Em:ρ= aEm+ρamb (a=0.5), independent of local time and for all the storms during 2002 to 2005. The ambient density,ρamb, is determined from the quiet day before the storm, and the solar flux influence on pamb during the storm is taken into account.e) The linear relation can reproduce the storm-time density changes, especially the orbit-averaged density sufficiently well. For that reason the proposed model can, for example, be used for calculating the storm effect on satellite drag.3. The established predicting model mentioned above is used to reproduce the orbit-averaged density at mid and low latitudes along CHAMP and GRACE orbits. The performances of the model at 400 km and 500 km are compared, indicating good predict capability at both altitudes. The findings in this study can be concluded as:a) The merging electric field is better correlated with the orbit-averaged density than the segment-averaged one. The model gives very satisfactory prediction for the orbit-averaged density.b) The orbit-averaged density along CHAMP and GRACE orbits both show a delayed response of 4.5 h to the solar wind inputs during the geomagnetic storms.c) The a value at GRACE altitude is around 30% of that at CHAMP altitude. The logarithm of the ratio axHAMP/aGRAXE (derived from each storm individually) varies almost linearly with respect to the altitude difference of the two satellites.d) For an arbitrarily selected storm in Dec.2006, the prediction of the linear model is better than the prediction of the NRLMSISE-00 model as well as the JB2008 model, especially at GRACE attitude.e) The CIR-driven storms have commonly larger a values than the ICME-driven storms.The innovation of this dissertation can be concluded as:To our knowledge the polar cap density anomaly during geomagnetic storms is studied comprehensively for the first time. The main feature of the anomaly is investigated in detail by a statistical study, and possible driving mechanisms are provided. The concept of using the merging electric field as the main input for a storm-time density predict model is also introduced for the first time. The results show that the model can reproduce the storm-time thermospheric density disturbances at both 400 km and 500 km successfully. The findings in this dissertation are of importance not only to the theoretical studies of the thermosphere, but also to the precise orbit determination of the LEO satellites.