Geospatial environmental data modelling applications using remote sensing, GIS and spatial statistics

This thesis presents novel modelling applications for environmental geospatial data using remote sensing, GIS and statistical modelling techniques. The studied themes can be classified into four main themes: (i) to develop advanced geospatial databases. Paper (I) demonstrates the creation of a geospatial database for the Glanville fritillary butterfly (Melitaea cinxia) in the Åland Islands, south-western Finland; (ii) to analyse species diversity and distribution using GIS techniques. Paper (II) presents a diversity and geographical distribution analysis for Scopulini moths at a world-wide scale; (iii) to study spatiotemporal forest cover change. Paper (III) presents a study of exotic and indigenous tree cover change detection in Taita Hills Kenya using airborne imagery and GIS analysis techniques; (iv) to explore predictive modelling techniques using geospatial data. In Paper (IV) human population occurrence and abundance in the Taita Hills highlands was predicted using the generalized additive modelling (GAM) technique. Paper (V) presents techniques to enhance fire prediction and burned area estimation at a regional scale in East Caprivi Namibia. Paper (VI) compares eight state-of-the-art predictive modelling methods to improve fire prediction, burned area estimation and fire risk mapping in East Caprivi Namibia. The results in Paper (I) showed that geospatial data can be managed effectively using advanced relational database management systems. Metapopulation data for Melitaea cinxia butterfly was successfully combined with GPS-delimited habitat patch information and climatic data. Using the geospatial database, spatial analyses were successfully conducted at habitat patch level or at more coarse analysis scales. Moreover, this study showed it appears evident that at a large-scale spatially correlated weather conditions are one of the primary causes of spatially correlated changes in Melitaea cinxia population sizes. In Paper (II) spatiotemporal characteristics of Socupulini moths description, diversity and distribution were analysed at a world-wide scale and for the first time GIS techniques were used for Scopulini moth geographical distribution analysis. This study revealed that Scopulini moths have a cosmopolitan distribution. The majority of the species have been described from the low latitudes, sub-Saharan Africa being the hot spot of species diversity. However, the taxonomical effort has been uneven among biogeographical regions. Paper III showed that forest cover change can be analysed in great detail using modern airborne imagery techniques and historical aerial photographs. However, when spatiotemporal forest cover change is studied care has to be taken in co-registration and image interpretation when historical black and white aerial photography is used. In Paper (IV) human population distribution and abundance could be modelled with fairly good results using geospatial predictors and non-Gaussian predictive modelling techniques. Moreover, land cover layer is not necessary needed as a predictor because first and second-order image texture measurements derived from satellite imagery had more power to explain the variation in dwelling unit occurrence and abundance. Paper V showed that generalized linear model (GLM) is a suitable technique for fire occurrence prediction and for burned area estimation. GLM based burned area estimations were found to be more superior than the existing MODIS burned area product (MCD45A1). However, spatial autocorrelation of fires has to be taken into account when using the GLM technique for fire occurrence prediction. Paper VI showed that novel statistical predictive modelling techniques can be used to improve fire prediction, burned area estimation and fire risk mapping at a regional scale. However, some noticeable variation between different predictive modelling techniques for fire occurrence prediction and burned area estimation existed.

The influence of the Coriolis force on flux tubes rising through the solar convection zone

In order to study the effect of the Coriolis force due to solar rotation on rising magnetic flux, the authors consider a flux ring, azimuthally symmetric around the rotation axis, starting from rest at the bottom of the convection zone, and then follow the trajectory of the flux ring as it rises. If it is assumed that the flux ring remains azimuthally symmetric during its ascent, then the problem can be described essentially in terms of two parameters: the value of the initial magnetic field in the ring when it starts, and the effective drag experienced by it. For field strengths at the bottom of the convection zone of order 10,000 G or less, it is found that the Coriolis force plays a dominant role and flux rings starting from low latitudes at the bottom are deflected and emerge at latitudes significantly poleward of sunspot zones.

Penetration of hydromagnetic energy deep into the magnetosphere

Magnetometer data, acquired on spacecraft and simultaneously at high and low latitudes on the ground, are compared in order to study the propagation characteristics of hydromagnetic energy deep into the magnetosphere. Single events provide evidence that wave energy at L ∼ 3 can at times be only one order of magnitude lower than at L ∼ 13. In addition, statistical analyses of the H-component groundbased data obtained during local daytime hours of 17 July-3 August 1985 show that wave amplitudes at L ∼ 3 are generally 10-30 times lower than at L ∼ 13. The L-dependence of near-equator magnetic field fluctuations measured on ISEE-2 show a sharp drop in energy near the magnetopause and a more gradual fall-off of energy deeper inside the magnetosphere. Such high levels of wave power deep in the magnetosphere have not been quantitatively understood previously. Our initial attempt is to calculate the decay length of an evanescent wave generated at a thick magnetopause boundary. Numerical calculations show that fast magnetosonic modes (called magnetopause and inner mode) can be generated under very restrictive conditions for the field and plasma parameters. These fast compressional modes may have their energy reduced by only one order of magnitude over a penetration depth of about 8RE. More realistic numerical simulations need to be carried out to see whether better agreement with the data can be attained.

Effect of Turbulence on Emerging Magnetic Flux Tubes in the Convection Zone

Working under the hypothesis that magnetic flux in the sun is generated at the bottom of the convection zone, Choudhuri and Gilman (1987; Astrophys. J. 316, 788) found that a magnetic flux tube symmetric around the rotation axis, when released at the bottom of the convection zone, gets deflected by the Coriolis force and tends to move parallel to the rotation axis as it rises in the convection zone. As a result, all the flux emerges at rather high latitudes and the flux observed at the typical sunspot latitudes remains unexplained. Choudhuri (1989; Solar Physics, in press) finds that non-axisymmetric perturbations too cannot subdue the Coriolis force. In this paper, we no longer treat the convection zone to be passive as in the previous papers, but we consider the role of turbulence in the convection zone in inhibiting the Coriolis force. The interaction of the flux tubes with the turbulence is treated in a phenomenological way as follows: (1) Large scale turbulence on the scale of giant cells can physically drag the tubes outwards, thus pulling the flux towards lower latitudes by dominating over the Coriolis force. (2) Small scale turbulence of the size of the tubes can exchange angular momentum with the tube, thus suppressing the growth of the Coriolis force and making the tubes emerge at lower latitudes. Numerical simulations show that the giant cells can drag the tubes and make them emerge at lower latitufes only if the velocities within the giant cells are unrealistically large of if the radii of the flux tubes are as small as 10 km. However, small scale turbulence can successfully suppress the growth of the Coriolis force if the tubes have radii smaller than about 300 km which may not be unreasonable. Such flux tubes can then emerge at low latitudes where sunspots are seen.

Can equipartition fields produce the tilts of bipolar magnetic regions?

The effect of turbulence on the nonaxisymmetric flux rings of equipartition field strength in bipolar magnetic regions is studied on the basis of the small-scale momentum exchange mechanism and the giant cell drag combined with the Kelvin-Helmholtz drag mechanism. It is shown that the giant cell drag and small-scale momentum exchange mechanism can make equipartition flux loops emerge at low latitudes, in addition to making them exhibit the observed tilts. However, the sizes of the flux tubes have to be restricted to a couple of hundred kilometers. An ad hoc constraint on the footpoints of the flux loops is introduced by not letting them move in the phi direction, and it is found that equipartition fields of any size can be made to emerge at sunspot latitudes with the observed tilts by suitably adjusting the footpoint separations.

Solar dynamo models with realistic internal rotation

Solar dynamo models based on differential rotation inferred from helioseismology tend to produce rather strong magnetic activity at high solar latitudes, in contrast to the observed fact that sunspots appear at low latitudes. We show that a meridional circulation penetrating below the tachocline can solve this problem.

IS A DEEP ONE-CELL MERIDIONAL CIRCULATION ESSENTIAL FOR THE FLUX TRANSPORT SOLAR DYNAMO?

The solar activity cycle is successfully modeled by the flux transport dynamo, in which the meridional circulation of the Sun plays an important role. Most of the kinematic dynamo simulations assume a one-cell structure of the meridional circulation within the convection zone, with the equatorward return flow at its bottom. In view of the recent claims that the return flow occurs at a much shallower depth, we explore whether a meridional circulation with such a shallow return flow can still retain the attractive features of the flux transport dynamo (such as a proper butterfly diagram, the proper phase relation between the toroidal and poloidal fields). We consider additional cells of the meridional circulation below the shallow return flow-both the case of multiple cells radially stacked above one another and the case of more complicated cell patterns. As long as there is an equatorward flow in low latitudes at the bottom of the convection zone, we find that the solar behavior is approximately reproduced. However, if there is either no flow or a poleward flow at the bottom of the convection zone, then we cannot reproduce solar behavior. On making the turbulent diffusivity low, we still find periodic behavior, although the period of the cycle becomes unrealistically large. In addition, with a low diffusivity, we do not get the observed correlation between the polar field at the sunspot minimum and the strength of the next cycle, which is reproduced when diffusivity is high. On introducing radially downward pumping, we get a more reasonable period and more solar-like behavior even with low diffusivity.

Tracking Pacific bluefin tuna (Thunnus thynnus orientalis) in the northeastern Pacific with an automated algorithm that estimates latitude by matching sea-surface-temperature data from satellites with temperature data from tags on fish

Data recovered from 11 popup satellite archival tags and 3 surgically implanted archival tags were used to analyze the movement patterns of juvenile northern bluefin tuna (Thunnus thynnus orientalis) in the eastern Pacific. The light sensors on archival and pop-up satellite transmitting archival tags (PSATs) provide data on the time of sunrise and sunset, allowing the calculation of an approximate geographic position of the animal. Light-based estimates of longitude are relatively robust but latitude estimates are prone to large degrees of error, particularly near the times of the equinoxes and when the tag is at low latitudes. Estimating latitude remains a problem for researchers using light-based geolocation algorithms and it has been suggested that sea surface temperature data from satellites may be a useful tool for refining latitude estimates. Tag data from bluefin tuna were subjected to a newly developed algorithm, called “PSAT Tracker,” which automatically matches sea surface temperature data from the tags with sea surface temperatures recorded by satellites. The results of this algorithm compared favorably to the estimates of latitude calculated with the lightbased algorithms and allowed for estimation of fish positions during times of the year when the lightbased algorithms failed. Three near one-year tracks produced by PSAT tracker showed that the fish range from the California−Oregon border to southern Baja California, Mexico, and that the majority of time is spent off the coast of central Baja Mexico. A seasonal movement pattern was evident; the fish spend winter and spring off central Baja California, and summer through fall is spent moving northward to Oregon and returning to Baja California.

大气季节内振荡对东亚夏季风活动的影响研究

利用ERA40逐日再分析资料、NCEP/NCAR2逐日再分析资料、中国740个测站日降水资料、上海台风研究所提供的西太平洋热带气旋资料、Kaplan等重建的月平均SSTA资料、NOAA逐日长波辐射（OLR）等资料，应用离散功率谱分析、带通滤波、EOF分析等统计方法，研究了东亚夏季风(EASM）的移动特征、东亚地区季节内振荡(ISO)的基本特征、季节内振荡对东亚夏季风活动的影响、季节内振荡对东亚夏季风异常活动的影响机理。主要结论如下： （1）综合动力和热力因素定义了可动态描述东亚夏季风移动和强度的指数，并利用该指数研究了东亚夏季风的爆发和移动的季节内变化及其年际和年代际变化特征。研究发现，气候平均东亚夏季风前沿分别在28候、33候、36候、38候、40候、44候出现了明显的跳跃。东亚夏季风活动具有显著的年际变率，主要由于季风前沿在某些区域异常停滞和突然跨越北跳或南撤引起，造成中国东部旱涝灾害频繁发生。东亚夏季风的活动具有明显的年代际变化，在1965年、1980年、1994年发生了突变，造成中国东部降水由“南旱北涝”向“南涝北旱”的转变。 （2）东亚季风区季节内变化具有10~25d和30~60d两个波段的季节内振荡周期，以30-60d为主。存在三个主要低频模态，第一模态主要表征了EASM在长江中下游和华北地区活动期间的低频形势；第二模态印度洋-菲律宾由低频气旋式环流控制，主要表现了ISO在EASM爆发期间的低频形势；第三模态主要出现在EASM在华南和淮河活动期间的低频形势。第一模态和第三模态是代表东亚夏季风活动异常的主要低频形势。 （3）热带和副热带地区ISO总是沿垂直切变风的垂直方向传播。因此，在南海-菲律宾东北风垂直切变和副热带西太平洋北风垂直切变下，大气热源激发菲律宾附近交替出现的低频气旋和低频反气旋不断向西北传播，副热带西太平洋ISO以向西传播为主。中高纬度地区，乌拉尔山附近ISO以向东、向南移动或局地振荡为主；北太平洋中部ISO在某些情况下向南、向西传播。 （4）季风爆发期，伴随着热带东印度洋到菲律宾一系列低频气旋和低频反气旋， 冷空气向南输送，10~25天和30~60天季节内振荡低频气旋同时传入南海加快了南海夏季风的爆发。在气候态下，ISO活动表现的欧亚- 太平洋（EAP）以及太平洋-北美（PNA）低频波列分布特征(本文提出的EAP和PNA低频波列与传统意义上的二维定点相关得到的波列不同)。这种低频分布形式使得欧亚和太平洋中高纬度的槽、脊及太平洋副热带高压稳定、加强，东亚地区的低频波列则成为热带和中高纬度ISO相互作用影响东亚夏季风活动的纽带。不同的阶段表现不同的低频模态，30~60d低频模态的转变加快了EASM推进过程中跳跃性；30-60d低频模态的维持使得EASM前沿相对停滞。 （5）30-60d滤波场，菲律宾海域交替出现的低频气旋和低频反气旋不断向西北传播到南海-西太平洋一带。当南海-西太平洋地区低频气旋活跃时，季风槽加强、东伸，季风槽内热带气旋（TC）频数增加；当南海-西太平洋低频反气旋活跃时，季风槽减弱、西退，TC处于间歇期，生成位置不集中。 （6）在El Nino态下，大气季节内振荡偏弱，北传特征不明显，但ISO由中高纬度北太平洋中部向南和副热带西太平洋向西的传播特征显著，东亚地区ISO活动以第三模态为主，EASM集中停滞在华南和淮河流域，常伴随着持续性区域暴雨的出现，易造成华南和江淮流域洪涝灾害，长江和华北持续干旱。在La Nina态下，大气季节内振荡活跃，且具有明显的向北传播特征，PNA低频波列显著，东亚地区ISO活动以第一模态单峰为主；EASM主要停滞在长江中下游和华北地区，这些地区出现异常持续强降水，华南和淮河流域多干旱；在El Nino态向La Nina态转换期，ISO活动以第一模态双峰为主，长江中下游常常出现二度梅。

中低纬度变化磁场的暴时特征研究

Magnetic storm is a kind of severe disturbances in the whole solar-earth electromagnetic space. It has significant effects on communication, electric power, oil transport pipe and human activities in space. Therefore, magnetic storms are worth for applications systems, not only being a favorable issue for scientists. In this paper, the spatial and temporal distributions of the magnetic fields produced by the magnetosphere-ionosphere current systems during storms are studied. Four parts are included in this paper decomposion of different disturbances with different origins, topological structure of the ring current, the asymmetric characteristics of the ring current, and the statistic peculiarities of the day-to-day variability (DTD) of Sq. 1 The decomposition of magnetic disturbances at mid-low latitudes and its evolutions during storms Transient variations in the geomagnetic field recorded at mid-low latitudes mainly include the storm-time variation (Dst), solar quiet daily variation (Sq) and disturbance daily variation (SD). With the data of the geomagnetic meridian chain observatories in China, 25 storms during the period of 1997 to 1999 have been analyzed. According to the features of different variations, a method of “three-steps decomposition” is developed by using the method of Natural Orthogonal Components (NOC), Correlation Analysis and Fourier Analysis to separate those three components in turn. The results show that, the first eigenmode by the MNOC clearly describing the special distribution and temporal evolution of storm-time variation, in addition, Correlation Analysis and Fourier Analysis offer a useful method to extract the Sq and SD variations. The latitudinal shift of the Sq current focus seems to be the principal reason of the day-to-day variaitons in the daily range of Sq. The magnitude of SD reaches a maximum during the main phase, and then gradually decreases. 2 The topology structure of the ring current during storms Both the mechanism of the ring current and the geomagnetic data suggest that the central plane of the ring current is declining to the geomagnetic equator plane with a tilt angle δ. Using the H and Z component data at two stations in a meridian chain, we deduce a new parameter describing the invariable peculiarity of different storms. Then the δ angle is calculated by using the data from a meridian chain and tested with the ERC model. Finally the deduced tilt angles are used to modify Dst index. 3 The asymmetric characteristics of the ring current during storms The variations of the geomagnetic field at mid-low latitudes show a significant dawn-dusk asymmetry, resulting from the superposition of the fields from the symmetric ring current and the partial ring current. On the basis of the data from the 20°E, 30°E meridian chains and 30°N latitudinal chain, the dawn-dusk asymmetry is investigated by using three methods, namely, statistic analysis, ring current model calculation and typical event analysis. This characteristic implies the asymmetry of the spatial distribution of the ring current. In addition, during the main phase after the sudden commencement (SC), H field increases and reaches maximum around noontime, implying the effect of the Chapman-Ferraro current. 4 The statistic characteristics of the day-to-day variability and its mechanism The day-to-day variability of the geomagnetic Sq field is studied by using the magnetic data from a meridian chain of magnetometers along 120° E longitude. The method of NOC is applied to separate the Sq variation from complicated disturbances. The first eigenmode with the largest eigenvalue represents fairly well the Sq variation with a conspicuous day-to-day variability in the daily range. For the stations on the same north- or south-side of the Sq current system focus, the day-to-day variations show a positive correlation. In contrast, for the stations on the different sides of the Sq focus, they show a negative correlation, suggesting an important role of latitudinal shift of the Sq current system focus to the day-to-day variability of the Sq daily range. The Sq daily range is correlated with the magnetic indices Ap and Dst in a peculiar way: on some severe disturbed days, noticeably enhancements of the Sq are observed, implying increases of the ionospheric conductivities and/or tidal wind velocities; on other severe disturbed days, however, dramatically reduced Sq variations occur, suggesting dominant effects of the ‘disturbance dynamo’ process.

太阳辐射对热层和电离层变化性的影响

Solar ultraviolet (UV) radiation at wavelengths less than 400 nm is an important source of energy for aeronomic processes throughout the solar system. Solar UV photons are absorbed in planetary atmospheres, as well as throughout the heliosphere, via photodissociation of molecules, photoionization of molecules and atoms, and photoexcitation toexcitation including resonance scattering. In this paper, the solar irradiances data measured by TIMED SEE, as well as the solar proxies such as F10.7 and Mg II, thermosphere neutral density of CHAMP measurements and topside ionospheric plasmas densities from DMSP, are used to analyze solar irradiance effects on the variabilities of the thermosphere and the ionosphere. First, thermosphere densities near 410 km altitude are analyzed for solar irradiance variability effects during the period 2002-2004. Correlations between the densities and the solar irradiances for different spectral lines and wavelength ranges reveal significantly different characteristics. The density correlates remarkably well with all the selected solar irradiances except the lower chromospheric O I (130.4 nm) emission. Among the chosen solar proxies, the Mg II core-to-wing ratio index, EUV (30-120 nm) and F10.7 show the highest correlations with the density for short-term (< ~27 days) variations. For both long- (> ~27 days) and short-term variations, linear correlation coefficients exhibit a decreasing trend from low latitudes towards high latitudes. The density variability can be effectively modeled (capturing 71% of the variance) using multiple solar irradiance indices, including F10.7, SEUV (the EUV 30-120 nm index), and SFUV (the FUV 120-193 nm index), in which a lag time of 1 day was used for both F10.7 and SEUV, and 5 days for SFUV. In our regression formulation SEUV has the largest contribution to the density variation (40%), with the F10.7 having the next largest contribution (32%) and SFUV accounting for the rest (28%). Furthermore, a pronounced period of about 27.2 days (mean period of the Sun's rotation) is present in both density and solar irradiance data of 2003 and 2004, and a pronounced period of about 54.4 days (doubled period of the solar rotation) is also revealed in 2004. However, soft X-ray and FUV irradiances did not present a pronounced 54.4 day period in 2004, in spite of their high correlation with the densities. The Ap index also shows 54-day periodicities in 2004, and magnetic activity, together with solar irradiance, affects the 54-day variation in density significantly. In addition, NRLMSISE00, DTM-2000 and JB2006 model predictions are compared with density measurements from CHAMP to assess their accuracy, and the results show that these models underestimate the response of the thermosphere to variations induced by solar rotation. Next, the equatorial topside ionospheric plasmas densities Ni are analyzed for solar irradiance variability effects during the period 2002-2005. Linear correlations between Ni and the solar irradiances for different wavelength ranges reveal significantly different characteristics. XUV (0-35 nm) and EUV (115-130 nm) show higher correlation with Ni for the long-term variations, whereas EUV (35-115 nm) show higher correlation for the short-term variations. Moreover, partial correlation analysis shows that the long-term variations of Ni are affected by both XUV (0-35 nm) and EUV (35-115 nm), whereas XUV (0-35 nm) play a more important role; the short-term variations of Ni are mostly affected by EUV (35-115 nm). Furthermore, a pronounced period of about 27 days is present in both Ni and solar irradiance data of 2003 and 2004, and a pronounced period of about 54 days is also revealed in 2004. Finally, prompted by previous studies that have suggested solar EUV radiation as a means of driving the semiannual variation, we investigate the intra-annual variation in thermosphere neutral density near 400 km during 2002-2005. The intra-annual variation, commonly referred to as the ‘semiannual variation’, is characterized by significant latitude structure, hemispheric asymmetries, and inter-annual variability. The magnitude of the maximum yearly difference, from the yearly minimum to the yearly maximum, varies by as much as 60% from year to year, and the phases of the minima and maxima also change by 20-40 days from year to year. Each annual harmonic of the intra-annual variation, namely, annual, semiannual, ter-annual and quatra-annual, exhibits a decreasing trend from 2002 through 2005 that is correlated with the decline in solar activity. In addition, some variations in these harmonics are correlated with geomagnetic activity, as represented by the daily mean value of Kp. Recent empirical models of the thermosphere are found to be deficient in capturing most of the latitude dependencies discovered in our data. In addition, the solar flux and geomagnetic activity proxies that we have employed do not capture some latitude and inter-annual variations detected in our data. It is possible that these variations are partly due to other effects, such as seasonal-latitudinal variations in turbopause altitude (and hence O/N2 composition) and ionosphere coupling processes that remain to be discovered in the context of influencing the intra-annual variations depicted here. Our results provide a new dataset to challenge and validate thermosphere-ionosphere general circulation models that seek to delineate the thermosphere intra-annual variation and to understand the various competing mechanisms that may contribute to its existence and variability. We furthermore suggest that the term “intra-annual” variation be adopted to describe the variability in thermosphere and ionosphere parameters that is well-captured through a superposition of annual, semiannual, ter-annual, and quatra-annual harmonic terms, and that “semiannual’ be used strictly in reference to a pure 6-monthly sinusoidal variation. Moreover, we propose the term “intra-seasonal” to refer to those shorter-term variations that arise as residuals from the above Fourier representation.

中低纬电离层暴的统计分析与经验建模

As a key issue of ionospheric weather study, systemic studies on ionospheric storms can not only further improve our understanding of the response of the ionosphere to solar and geomagnetic disturbances, but also help us to reveal the chemical, dynamic and electro-dynamic mechanisms during storms. Empirical modelling for regional ionospheric storm is also very useful, because it can provide us with tools and references for the forecasting and further practical application of ionospheric activity. In this thesis, we focus on describing and forecasting of ionospheric storms at middle and low latitudes. The main points of my investigations are listed as follows. (1) By using magnetic storms during the period over 50 years, the dependence of the type, onset time and time delay of the ionospheric storms on magnetic latitude, season and local time at middle and low latitudes in the East-Asian sector are studied. The results show that the occurrences of the types of ionospheric disturbances differ in latitude and season. The onset of the ionospheric storms depends on local time. At middle latitudes, most negative phase onsets are within the local time interval from night to early morning, and they rarely occurred in the local noon and afternoon sectors. At low latitudes, positive phases commence most frequently in the daytime sector as well as pre-midnight sector. The average time delays for both the positive and negative ionospheric storms increase with descending latitudes. The time delay has significant dependence on the local time of main phase onset (MPO). The time delay of positive response is shorter for daytime MPO and longer for night-time MPO, whereas the opposite applies for negative response. (2) Based on some previous researches, a primary empirical model for mid-latitude ionospheric disturbance is set up. By fitting to the observed data, we get a high accuracy with a mean RMSE of only 12-14% in summer and equinox. The model output has been compared with the output of STORM model, and the results show that, our model is much better than STORM in summer and a little better for some mid-latitude stations at equinox. Especially, for the type of two-step geomagnetic storm, our model can present twice descending of foF2 very well. In addition, our model can forecast positive ionospheric storms.

太阳活动变化对中低纬电离层的影响

With the variations of solar activity, solar EUV and X-ray radiations change over different timescales (e.g., from solar cycle variation to solar flare burst). Since solar EUV and X-ray radiations are the primary energy sources for the ionosphere, theirs variations undoubtedly produce significant and complicated effects on the ionosphere. So the variations of solar activity significantly affect the ionosphere. It is essential for both ionospheric theory and applications to study solar activity effects on the ionosphere. The study about solar activity variations of the ionosphere is an important part of the ionospheric climatology. It can enhance the understanding for the basic processes in the ionosphere, ionospheric structure and its change, ionosphere/thermosphere coupling, and so on. As for applications, people need sufficient knowledges about solar activity variations of the ionosphere in order to improve ionospheric models so that more accurate forecast for the ionospheric environments can be made. Presently, the whole image about the modalities of ionospheric solar activity variations is still unknown, and related mechanisms still cannot be well understood. This paper is about the effects of the 11-year change in solar activity to the low- and mid-latitude ionosphere. We use multi-type ionospheric observations and model to investigate solar activity effects on the electron density and ionospheric spatial structure, and we focus on discussing some related mechanisms. The main works are as follows: Firstly, solar activity variations of ionospheric peak electron density (NmF2) around 1400 LT were investigated using ionosonde observations in the 120°E sector. The result shows that the variation trend of NmF2 with F107 depends on latitudes and seasons. There is obvious saturation trend in low latitudes in all seasons; while in middle latitudes, NmF2 increases linearly with F107 in winter but saturates with F107 at higher solar activity levels in the other seasons. We calculated the photochemical equilibrium electron density to discuss the effects induced by the changes of neutral atmosphere and dynamics processes on the solar activity variations of NmF2. We found that: (1) Seasonal variation of neutral atmosphere plays an important role in the seasonal difference of the solar activity variations of NmF2 in middle latitudes. (2) Less [O]/[N2] and higher neutral temperature are important for the saturation effect in summer, and the increase of vibrational excited N2 is also important for the saturation effect. (3) Dynamics processes can significantly weaken the increase of NmF2 when solar activity enhances, which is also a necessary factor for the saturation effect. Secondly, solar activity variations of nighttime NmF2 were investigated using ionosonde observations in the 120°E sector. The result shows that the variation trends of NmF2 with F107 in nighttime are different from that in daytime in some cases, and the nighttime variation trends depend on seasons. There is linear increase trend in equinox nighttime, and saturation trend in summer nighttime, while the increase rate of NmF2 with F107 increases when solar activity enhances in winter nighttime (we term it with “amplification trend”). We discussed the possible mechanisms which affect the solar activity variations of nighttime NmF2. The primary conclusions are as follows: (1) In the equatorial ionization anomaly (EIA) crest region, the plasma influx induced by the pre-reversal enhancement (PRE) results in the change of the variation trend between NmF2 and F107 from “saturation” to “linear” after sunset in equinoxes and winter; while the recombination process at the F2-peak is the primary factor that affects the variation trend of NmF2 with F107 in middle latitudes. (2) The recombination coefficient at the F2-peak height reaches its maximum at moderate solar activity level in winter nighttime, which induces NmF2 attenuates more quickly at moderate solar activity level. This is the main reason for the amplification trend. (3) The change of the recombination process at the F2-peak with solar activity depends on the increases of neutral parameters (temperature, density et al.) and the F2-peak height (hmF2). The seasonal differences in the changes of neutral atmosphere and hmF2 with solar activity are the primary reasons for the seasonal difference in the variation trend of nighttime NmF2 with F107. Finally, we investigated the solar activity dependence of the topside ionosphere in low latitudes using ROCSAT-1 satellite (at 600 km altitude) observations. The primary results and conclusions are as follows: (1) Latitudinal distribution of the plasma density is local time, seasonal, and solar activity dependent. In daytime, there is a plasma density peak at the dip equator. The peak is obviously enhanced at high solar activity level, and the strength of the peak strongly depends on seasons. While at sunset, two profound plasma density peaks (double-peak structure) are found in solar maximum equinox months. (2) Local time dependence of the latitudinal distribution is due to the local time variation of the equatorial dynamics processes. Double-peak structure is attributed to the fountain effect induced by strong PRE. Daytime peak enhances with solar activity since the plasma density increases with solar activity more strongly at the dip equator due to the equatorial vertical drift, and its seasonal dependence is mainly due to the seasonal variations of neutral density and the equatorial vertical drift. In the sunset sector, seasonal and solar activity dependences of the latitudinal distribution are related to the seasonal and solar activity variations of PRE. (3) The variation trend of the plasma density with solar activity shows local time, seasonal, and latitudinal differences. That is different from the changeless amplification trend at the DMSP altitude (840 km). Profound saturation effect is found in the dip equator region at equinox sunset. This saturation effect in the topside ionosphere is realated to the increase of PRE with solar activity. Solar activity variation trend of the topside plasma density was discussed quantitatively by Chapman-α function. The result shows that the effect induced by the change of the scale height is dominant at high altitudes; while the variation trend of ROCSAT-1 plasma density with solar activity is suggested to be related to the changes of the peak height, the scale height, and the peak electron density with solar activity.

太阳耀斑和日食期间电离层行为的统计和模拟研究

The ionosphere is the ionized component of the Earth's upper atmosphere. Solar EUV radiation is the source of ionospheric ionization. Thus the ionosphere is affected strongly by the variations in solar radiation. Solar flares and solar eclipses can induce remarkable short time changes in solar radiation: the solar radiation would increase suddenly during solar flares and decrease significantly during solar eclipses. Solar flare and eclipse events not only affect directly the photochemical processes, but also affect the dynamic processes, and even affect the neutral atmosphere, which is strongly coupled with the ionosphere. The study on the ionospheric response to solar flares and eclipses can advance our knowledge on the ionosphere and its photochemical and dynamic processes and help us to evaluate the ionospheric parameters (such as ion loss coefficients). In addition, the study on the ionospheric responses to solar flares and eclipses is an important part of the ionospheric space weather, which can provide guides for space weather monitoring. This thesis devotes to the study on the ionospheric responses to solar flares and solar eclipses. I have developed two models to simulate the variations of solar EUV radiation during solar flares and solar eclipses, and involved in developing a 2D mid- and low-latitude ionospheric model. On the basis of some observed data and the ionospheric model, I study the temporal and spatial variations of the ionosphere during solar flares and eclipses, and investigate the influences of solar activity, solar zenith angle, neutral gas density, and magnetic dip angle on the ionospheric responses to solar flares and solar eclipses. The main points of my works and results are summarized as follows. 1. The ionospheric response to the X17.2 solar flare on October 28, 2003 was modeled via using a one-dimension theoretical ionospheric model. The simulated variation of TEC is in accordance with the observations, though there are some differences in the amplitude of the variation. Then I carried out a series of simulations to explore the local time and seasonal dependences of the ionospheric responses to solar flares. These calculations show that the ionospheric responses are largely related with the solar zenith angle (SZA). During the daytime (small SZA), most of the increases in electron density occur at altitudes below 300 km with a peak at around 115 km; whereas around sunrise and sunset (SZA>90°), the strongest ionospheric responses occur at much higher altitudes. The TEC increases slower at sunrise than at sunset, which is caused by the difference in the evolution of SZA at sunrise and sunset: SZA decreases with time at sunrise and increase with time at sunset. The ionospheric response is largest in summer and smallest in winter, which is also related to the seasonal difference of SZA. 2. Based on the observations from the ionosondes in Europe and the ionospheric model, I investigated the differences of the ionosphere responses to solar eclipses between the E-layer and F1-layer. Both the observation and simulation show that the decrease in foF1 due to the solar eclipses is larger than that in foE. This effect is due to that the F1 region locates at the transition height between the atomic ion layer and the molecular ion layer. With the revised model of solar radiation during solar flares, our model calculates the radiations from both the inside and outside of photosphere. Large discrepancy can be found between the observations and the calculations with an unrevised model, while the calculations with the revised model consist with the observations. 3. I also explore the effects of the F2-layer height, local time, solar cycle, and magnetic dip angle on the ionospheric responses to solar eclipses via using an ionospheric model and study on the solar zenith angle and the dip dependences by analyzing the data derived from 23 ionosonde stations during seven eclipse events. Both the measured and simulated results show that these factors have significant effect on the ionospheric response. The larger F2-layer height causes the smaller decrease in foF2, which is because that the electron density response decreases with height. The larger dip results in the smaller eclipse effect on the F2 layer, because the larger dip would cause the more diffusion from the top ionosphere which can make up for the plasma loss. The foF2 response is largest at midday and decreases with the increasing SZA. The foF2 response is larger at high solar activity than at low solar activity. The simulated results show that the local time and solar activity discrepancy of the eclipse effect mainly attribute to the difference of the background neutral gas density. 4. I carried out a statistical study on the latitudinal dependence of the ionospheric response to solar eclipses and modeled this latitudinal dependence by the ionospheric model. Both the observations and simulations show that the foF2 and TEC responses have the same latitudinal dependence: the eclipse effects on foF2 and TEC are smaller at low latitudes than at middle latitudes; at the middle latitudes (>40°), the eclipse effect decreases with increasing latitude. In addition, the simulated results show the change in electron temperature at the heights of above 300 km of low latitudes is much smaller than that at the same heights of middle latitudes. This is due to the smaller decrease in photoelectron production rate at its conjugate low heights. 5. By analyzing the observed data during the October 3, 2005 solar eclipse, I find some significant disturbances in the conjugate region of the eclipse region, including a decrease in Te, an increase in foF2 and TEC, and an uprising in hmF2. I also simulated the ionosphere behavior during this eclipse using a mid-low latitude ionospheric model. The simulations reproduce the measured ionospheric disturbances mentioned above in the conjugated hemisphere. The simulations show that the great loss of arriving photoelectron heat from the eclipse region is the principal driving source for the disturbances in the conjugate hemisphere.

电离层垂直TEC映射函数的实验观测与统计特性

When used in the determining the total electron content (TEC), which may be the most important ionospheric parameter, the worldwide GPS observation brings a revolutionary change in the ionospheric science. There are three steps in the data processing to retrieve GPS TEC: (1) to estimate slant TEC from the measurements of GPS signals; (2) to map the slant TEC into vertical; and (3) to interpolate the vertical TEC into grid points. In this scientific dissertation we focus our attention on the second step, the mapping theory and method to convert slant TEC into vertical. This is conventionally done by multiplying on the slant TEC a mapping function which is usually determined by certain models of electron density profile. Study of the vertical TEC mapping function is of significance in GPS TEC measurement. This paper first reviews briefly the three steps in GPS TEC mapping process. Then we compare the vertical TEC mapping function which were respectively calculated from the electron density profiles of the ionospheric model and retrieved from the observation of worldwide GPS TEC. We also perform the statistical analysis on the observational mapping functions. The main works and results are as follows: 1. We calculated the vertical TEC mapping functions for both SLM and Chapman models, and discussed the modulation of the ionosphere height to the mapping functions. We use two simple models, single layer model (SLM) and Chapman models, of the ionospheric electron density profiles to calculate the vertical TEC mapping function. In the case of the SLM, we discuss the control of the ionospheric altitude, i.e., the layer height hipp, to the mapping function. We find that the mapping function decreases rapidly as hipp increases. For the Chapman model we study also the control mapping function by both ionospheric altitude indicated by the peak electron density height hmF2, and the scale height, H, which present the thickness of the ionosphere. It is also found that the mapping function decreases rapidly as hmF2 increases. and it also decreases as H increases. 2. Then we estimate the mapping functions from the GPS observations and compare them with those calculated from the electron density models. We first, proposed a new method to estimate the mapping functions from GPS TEC data. This method is then used to retrieve the observational mapping function from both the slant TEC (TECS) provided by International GPS Service (IGS）and vertical TEC provide by JPL Global Ionospheric Maps (GIMs). Then we compare the observational mapping function with those calculated from the electron density models, SLM and Chapman. We find that the values of the observational mapping functions are much smaller than that from the model mapping functions, when the zenith angle is large enough. We attribute this to the effect of the plasmasphere which is above about 1000 km. 3. We statistically analyze the observational mapping functions and reveal their climatological changes. Observational mapping functions during 1999-2007 are used in our statistics. The main results are as follows. (1) The observational mapping functions decrease obviously with the decrement of the solar activity which is represented by the F10.7 index; (2) In annual variations of the observational mapping functions, the semiannual component is found at low-latitudes, and the remarkable seasonal variations at mid- and high-latitudes. (3) The diurnal variation of the observational mapping functions is that they are large in daytime and small at night, they become extremely small in the early morning before sunrise. (4) The observational mapping functions change with latitudes that they are smaller at lower latitudes and larger at higher. All of the above variations of the observational mapping functions are explained by the existence of the plasmasphere, which changes more slowly with time and more rapidly with latitude than the ionosphere does . In summary, our study on the vertical TEC mapping function imply that the ionosphere height has a modulative effect on the mapping function. We first propose the concept of the 'observational mapping functions' , and provide a new method to calculate them. This is important in improving the TEC mapping. It may also possible to retrieving the plasmaspheric information from GPS observations.