Longitudinal dependence of the forecast accuracy of the ionospheric total electron content in the equatorial zone

The longitudinal dependence of the behavior of ionospheric parameters has been the subject of a number of works where significant variations are discovered.This also applies to the prediction of the ionospheric total electron content(TEC),which neural network methods have recently been widely used.However,the results are mainly presented for a limited set of meridians.This paper examines the longitudinal dependence of the TEC forecast accuracy in the equatorial zone.In this case,the methods are used that provided the best accuracy on three meridians:European(30°E),Southeastern(110°E)and American(75°W).Results for the stations considered are analyzed as a function of longitude using the Jet Propulsion Laboratory Global Ionosphere Map(JPL GIM)for 2015.These results are for 2 h ahead and 24 h ahead forecast.It was found that in this case,based on the metric values,three groups of architectures can be distinguished.The first group included long short-term memory(LSTM),gated recurrent unit(GRU),and temporal convolutional networks(TCN)models as a part of unidirectional deep learning models;the second group is based on the recurrent models from the first group,which were supplemented with a bidirectional algorithm,increasing the TEC forecasting accuracy by 2-3 times.The third group,which includes the bidirectional TCN architecture(BiTCN),provided the highest accuracy.For this architecture,according to data obtained for 9 equatorial stations,practical independence of the TEC prediction accuracy from longitude was observed under the following metrics(Mean Absolute Error MAE,Root Mean Square Error RMSE,Mean Absolute Percentage Error MAPE):MAE(2 h)is 0.2 TECU approximately;MAE(24 h)is 0.4 TECU approximately;RMSE(2 h)is less than 0.5 TECU except Niue station(RMSE(2 h)is 1 TECU approximately);RMSE(24 h)is in the range of 1.0-1.7 TECU;MAPE(2 h)<1%except Darwin station,MAPE(24 h)<2%.This result was confirmed by data from additional 5 stations that formed latitudinal chains in the equatorial part of the three meridians.The complete correspondence of the observational and predicted TEC values is illustrated using several stations for disturbed conditions on December 19-22,2015,which included the strongest magnetic storm in the second half of the year(min Dst=-155 nT).

Total Electron Content Diurnal and Seasonal Variations and Response to Solar Events at Koudougou Station in Burkina Faso

In this paper, we studied the seasonal behavior of the total electron content (TEC) during a part of solar cycle 24 ascending, maximum and decreasing phases at Koudougou station (Latitude: 12°15'09"N Longitude: 2°21'45"W). Response of TEC to solar recurrent events is presented. The highest values of the TEC in 2014, 2015 and 2016 were recorded on March and October, while in 2013 they were recorded on April and November, corresponding to equinox months. This observation shows that TEC values at the equinoxes are higher than those of solstices. Moreover, the monthly TEC varies in phase with the sunspots number showing a linear dependence of the TEC on solar activity. The ionospheric electron contents are generally very low both before noon and during the night, but quite high at noon and after noon. This pattern of TEC variation is due to the fluctuation of incident solar radiation on the Earth’s equatorial ionosphere. During quiet periods, the number of free electrons generated is lower than that generated during recurrent periods, which shows a positive contribution of recurrent activity to the level of the TEC. Investigations have also highlighted a winter anomaly and equinoctial asymmetry in TEC behavior at Koudougou station.

Total Electron Content during Recurrent and Quiet Geomagnetic Periods at the Koudougou Station in Burkina Faso

In this work, the comparative study of total electron content (TEC) between recurrent and quiet geomagnetic periods of solar cycle 24 at Koudougou station with geographical coordinates 12°15'N;- 2°20'E was addressed. This study aims to analyze how geomagnetic variations influence the behavior of TEC in this specific region. The geomagnetic indices Kp and Dst were used to select quiet and recurrent days. Statistical analysis was used to interpret the graphs. The results show that the mean diurnal TEC has a minimum before dawn (around 0500 UT) and reaches a maximum value around 1400 UT, progressively decreasing after sunset. In comparison, the average diurnal TEC on recurrent days is slightly higher than on quiet days, with an average difference of 7 TECU. This difference increases with the level of geomagnetic disturbance, reaching 21 TECU during a moderate storm. The study also reveals significant monthly variations, with March and October showing the highest TEC values for quiet and recurrent days, respectively. Equinox months show the highest mean values, while solstice months show the lowest. Signatures of semi-annual, winter and equatorial ionization anomalies were observed. When analyzing annual variations, it was found that the TEC variation depends significantly on F10.7 solar flux, explaining up to 98% during recurrent geomagnetic activity and 92% during quiet geomagnetic activity.

Performance of GPS slant total electron content and IRI-Plas-STEC for days with ionospheric disturbance

Total Electron Content （TEC） is an important observable parameter of the ionosphere which forms the main source of error for space based navigation and positioning systems. Since the deployment of Global Navigation Satellite Systems （GNSS）, cost-effective esti- mation of TEC between the earth based receiver and Global Positioning System （GPS） sat- ellites became the major means of investigation of local and regional disturbance for earthquake precursor and augmentation system studies. International Reference Iono- sphere （IRI） extended to plasmasphere （IRI-Plas） is the most developed ionospheric and plasmaspheric climatic model that provides hourly, monthly median of electron density distribution globally. Recently, IONOLAB group {www.ionolab.org） has presented a new online space weather service that can compute slant TEC （STEC） on a desired ray path for a given date and time using IRI-Plas model （IRI-Plas-STEC）. In this study, the performance of the model based STEC is compared with GPS-STEC computed according to the estimation method developed by the IONOLAB group and includes the receiver bias as IONOLAB-BIAS （IONOLAB-STEC）. Using Symmetric Kullback-Leibler Distance （SKLD）, Cross Correlation （CC） coefficient and the metric norm （L2N） to compare IRI-Plas-STEC and IONOLAB-STEC for the month of October 2011 over the Turkish National Permanent GPS Network （TNPGN- Active）, it has been observed that SKLD provides a good indicator of disturbance for both earthquakes and geomagnetic storms.

基于注意力机制LSTM的电离层TEC预测

电离层总电子含量(Total Electron Content,TEC)的监测与预报是空间环境研究的重要内容,对卫星通讯和导航定位等有重要意义.TEC值影响因素较多,很难确定精确物理模型来对其进行预测.本文设计了基于注意力机制的LSTM模型(Att-LSTM),采用过去24小时TEC观测数据对未来TEC进行预测.选择北半球东经100°上,每2.5°纬度选择一个位置,共计36个位置来验证本文提出模型的性能,并与主流的深度学习模型如DNN、RNN、LSTM进行对比实验.取得了如下成果:(1)在选定的36个地区未来2小时单点预测上,基于本文的Att-LSTM模型的TEC预测性能明显优于其他对比模型;(2)讨论了纬度对Att-LSTM预测未来2小时TEC值时性能的影响,发现在北纬0°到60°之间,Att-LSTM预测性能随着纬度的升高而略有降低,在北纬62.5°~87.5°之间,模型预测性能出现扰动,预测效果略差;(3)讨论了磁暴期和磁静期模型的预测性能,发现无论是磁暴期还是磁静期,本文模型预测性能均较好;(4)还讨论了对未来多时点预测效果,实验结果表明,本文所提出的模型对未来2、4个小时的预测拟合度R-Square均超过0.95,预测结果比较可靠,对未来6、8、10个小时预测拟合度最高为0.7934,预测拟合度R-Square下降迅速,预测结果不可靠.

基于GPS TEC的2022年1月15日汤加火山喷发激起的电离层行扰分析

2022年1月15日南太平洋汤加海底火山发生剧烈喷发,是近30年来最大规模的火山爆发,产生的强烈大气波动为开展火山喷发电离层扰动研究提供了难得的机会。本文利用GPS数据探测火山附近、新西兰、澳大利亚和中国地区的电离层扰动,从波形、频率、传播速度和时空分布等角度分析汤加火山喷发电离层行扰(TIDs)的特征,并利用电离层测高站、海平面监测站和大气压监测站的观测数据进一步验证TIDs的传播特征。研究结果发现,汤加火山喷发在其附近区域、新西兰、澳大利亚和中国地区引起了3类TIDs。在火山附近东、西、南、北方向上均探测到第一类TIDs, TIDs的传播速度为617~972 m/s,该类TIDs极有可能由火山喷发产生的声波引起。汤加火山喷发仅在火山附近东、西方向引起第二类TIDs,其传播速度分别为472 m/s和418 m/s,可能由声波传播过程中衍生的声重力波或者混合波引起,形成机理有待进一步研究。汤加火山喷发在新西兰、澳大利亚和中国地区引发了第三类TIDs,其传播速度为328~352 m/s,该类TIDs与Lamb波密切相关。

Analysis of Anomalous Enhancement in TEC and Electron Density in the China Region Prior to the 17 March 2015 Geomagnetic Storm Based on Ground and Space Observations

Total Electron Content(TEC)and electron density enhancement were observed on the day before 17 March 2015 great storm in the China Region.Observations from ground-and space-based instruments are used to investigate the temporal and spatial evolution of the pre-storm enhancement.TEC enhancement was observed from 24°N to 30°N after 10:00 UT at 105°E,110°E and 115°E longitudes on March 16.The maximum magnitude of TEC enhancement was more than 10 TECU and the maximal relative TEC enhancement exceeded 30%.Compared with geomagnetic quiet days,the electron density of Equatorial Ionization Anomaly(EIA)northern peak from Swarm A/C satellites on March 16 was larger and at higher latitudes.NmF2 enhanced during 11:30—21:00 UT at Shaoyang Station and increased by 200%at~16:00 UT.However,TEC and electron density enhancement were not accompanied by a significant change of hmF2.Most research has excluded some potential mechanisms as the main driving factors for storm-time density enhancements by establishing observational constraints.In this paper,we observed pre-storm enhancement in electron density at different altitudes and Equatorial Electrojet(EEJ)strength results derived from ground magnetometers observations suggest an enhanced eastward electric field from the E region probably played a significant role in this event.

基于北斗GEO卫星观测的桂林地区TEC空间梯度研究

小尺度的总电子含量(TEC)梯度异常会显著影响区域电离层模型以及地基增强系统(GBAS)的精度。因此,为了提高区域电离层模型的准确性和GBAS导航定位的精度,深入研究小尺度的TEC空间梯度显得尤为重要。为了研究中小尺度的空间梯度,利用2019—2022年间桂林地区全球卫星导航系统(GNSS)接收机的北斗地球静止轨道(GEO)卫星数据,对中国低纬地区(23°N 107.9°~113.1°E)电离层的TEC空间梯度进行研究分析。根据3颗北斗GEO卫星对地形成的3个相邻穿刺点(从西至东,依次编号为C02,C03,C01)的TEC观测值,构成C02-C03与C03-C01两个约为300 km尺度区域(107.9°~110.3°E,110.3°~113.1°E)的TEC梯度观测。统计结果显示,C02-C03区域的空间梯度基本为西向;而C03-C01区域的空间梯度在1~4月与9~12月为东向,5~8月为西向。这表明2个相邻区域的空间梯度在季节变化上存在差异,揭示了中小尺度空间梯度具有显著的局地效应。此外,地方时差并非空间梯度事件的主要原因。在太阳活动高年,背景TEC较高,空间梯度事件更易发生。梯度事件发生率的半年变化是TEC半年异常的体现。空间梯度事件在不同地磁条件下的发生率相似。

Ionospheric total electron content disturbance associated with May 12, 2008,Wenchuan earthquake

Possible ionospheric disturbances relating to the May 12, 2008, MsS.0 Wenchuan earthquake were identified by Global Positioning System （GPS）-derived total electron content （TEC）, ion- osonde observations, the global ionospheric map （GIM）, and electron density profiles detected by the Constellation Observation System for Meteorology Ionosphere and Climate （COSMIC）. We applied a statistical test to detect anomalous TEC signals and found that a unique enhancement in TEC, recorded at 16 GPS stations, appeared on May 9, 2008. The critical fre- quency at F2 peak （foF2）, observed by the Chinese ionosondes, and maximal plasma frequency, derived from COSMIC data, revealed a characteristic similar to GPS TEC variations. The GIM showed that the anomalous variations of May 9 were located southeast of the epicenter. Using GPS data from 13 stations near the epicenter, we analyzed the TEC variations of satellite orbit traces during 04：00-11：00 UT. We found that TEC decreased to the east and increased to the southeast of the epicenter during this period. Results showed that the abnormal disturbance on May 9 was probably an ionosphenc precursor of the Wenchuan earthquake of May 12, 2008.

Manifestations of the earthquake preparations in the ionosphere total electron content variations

Recent ionospheric observations report anomalous total electron content (TEC) deviations prior strong earthquakes. We discuss common fetures of the pre-earthquake TEC disturbances on the basis of statistics covering 50 strong seismic events during 2005-2006. The F2-layer ionospheric plasma drift under action of the electric fields of seismic origin is proposed as the main reason of producing TEC anomalies. The origin of such electric fields is discussed in terms of the lithosphere-atmosphere-ionosphere coupling system. This theory is supported by numerical simulations using global Upper Atmosphere Model (UAM). UAM calculations show that the vertical electric current with the density of about 20 - 40 nA/m2 flowing between the Earth and ionosphere over an area of about 200 by 2000 km is required to produce the TEC disturbances with the amplitude of about 30% - 50% relatively to the non-disturbed conditions. Ionosphere responses on the variations of the latitudinal position, direction and configuration of the vertical electric currents have been investigated. We show that not only the vertical component of the ionospheric plasma drift but also horizontal components play an important role in producing pre-earthquake TEC disturbances.

Ionospheric vertical total electron content prediction model in low-latitude regions based on long short-term memory neural network

Ionosphere delay is one of the main sources of noise affecting global navigation satellite systems, operation of radio detection and ranging systems and very-long-baseline-interferometry. One of the most important and common methods to reduce this phase delay is to establish accurate nowcasting and forecasting ionospheric total electron content models. For forecasting models, compared to mid-to-high latitudes, at low latitudes, an active ionosphere leads to extreme differences between long-term prediction models and the actual state of the ionosphere. To solve the problem of low accuracy for long-term prediction models at low latitudes, this article provides a low-latitude, long-term ionospheric prediction model based on a multi-input-multi-output, long-short-term memory neural network. To verify the feasibility of the model, we first made predictions of the vertical total electron content data 24 and 48 hours in advance for each day of July 2020 and then compared both the predictions corresponding to a given day, for all days. Furthermore, in the model modification part, we selected historical data from June 2020 for the validation set, determined a large offset from the results that were predicted to be active, and used the ratio of the mean absolute error of the detected results to that of the predicted results as a correction coefficient to modify our multi-input-multi-output long short-term memory model. The average root mean square error of the 24-hour-advance predictions of our modified model was 4.4 TECU, which was lower and better than5.1 TECU of the multi-input-multi-output, long short-term memory model and 5.9 TECU of the IRI-2016 model.

基于SSA-PSO-LSTM模型的电离层TEC预报

受多种因素影响,电离层电子总含量(TEC)时间序列具有非线性、非平稳性特征,为提升长短期记忆(LSTM)神经网络模型在电离层TEC预报中的精度,本文在该神经网络模型的基础上,引入奇异谱分析(SSA)与粒子群优化(PSO)算法,构建了新的SSA-PSO-LSTM模型。一方面,利用了SSA对TEC时间序列进行数据预处理;另一方面,利用粒子群优化算法改进LSTM神经网络模型参数。选用欧洲地球参考框架(EUREF)提供的格网点电离层TEC时间序列数据进行实验,实验结果表明,在磁平静期与磁暴期,该组合模型的TEC预报结果均方根误差分别为0.28个总电子含量单位(TECu)、0.83个TECu,平均相对精度分别为96.35%、91.33%,均优于对比模型,验证了本文提出的组合预报模型的有效性与优越性。平均相对精度分别为96.35%、91.33%,均优于对比模型,验证了本文提出的组合预报模型的有效性与优越性。

Questions on Optimization of Measurement of Total Electron Content in Ionosphere with GPS

The analysis of existing method for calculation of total content of electrons (TEC) in ionosphere using GPS occultation method does show that due to different values of signal/noise ration in GPS signals ?and , the new method of optimum measurements of relevant frequency components of TEC measured by phase and code methods should be developed. The optimum quantity of measurements of the above-mentioned frequency components is determined taking into account the limitation imposed on general number of necessary measurements.

基于滑窗初始点dSTEC定权的全球实时电离层综合

自2017年起,国际GNSS服务组织(IGS)部分电离层分析中心先后提供全球实时电离层校正服务。针对实际应用中单一分析中心实时电离层产品存在的可靠性差问题,本文提出了基于滑窗初始点差分斜向总电子含量(dSTEC)定权的全球实时电离层综合方法。以中国科学院(CAS)、法国空间研究中心(CNES)、西班牙加泰罗尼亚理工大学(UPC)及武汉大学(WHU)实时产品为基础,实现了实时全球电离层地图(RT-GIM)产品的例行综合。从电离层延迟误差修正、单频标准单点定位(SF-SPP)及单频精密单点定位(SF-PPP) 3个方面,分析2022年2月15日至2022年3月15日综合RT-GIM产品的应用性能。以IGS事后GIM为参考,本文综合RT-GIM与UPC综合RT-GIM对应的RMS为分别为3.30和3.20 TECU,二者精度相当;以定位残差95%分位数为统计量,与IGS事后GIM产品相比,综合RT-GIM对应的SF-SPP及SF-PPP精度分别低7.7%和4.9%;与北斗三号广播电离层模型BDGIM相比,综合RT-GIM对应的SF-SPP及SF-PPP精度分别提升了15.9%和9.5%。自2022年起,CAS综合RT-GIM产品已例行提交至IGS。

基于TEC数据的含电离层闪烁卫星导航信号模拟

电离层闪烁导致卫星导航信号在传播过程中幅度与相位发生随机波动,严重影响接收机的性能。为模拟真实的受电离层闪烁影响的卫星导航信号,以供接收机进行性能测试,提出了基于电离层总电子含量(Total Electron Content,TEC)数据的电离层闪烁仿真方法。该方法通过输入卫星观测文件与导航电文获得电离层的TEC和仰角,利用TEC数据和各卫星仰角,估计出受电离层闪烁影响的各卫星导航信号幅度闪烁指数和相位闪烁指数,结合Cornell模型实现卫星导航信号模拟。该方法充分考虑了卫星导航信号闪烁指数获取困难,以及电离层闪烁对不同卫星导航信号的影响,能够高保真反演卫星导航信号。试验结果表明,该方法反演的电离层闪烁与实际发生的闪烁具有良好的一致性。

太阳活动下降年昆明地区TEC变化特征与IRI-2020模拟比较分析

利用昆明低纬度测站(24.7°N,102.9°E,磁纬15.1°N)2016-2019年的观测数据和最新版的国际参考电离层(IRI-2020)模拟结果,对昆明地区电离层总电子含量(TEC)在太阳活动下降年期间的变化特征及与模型输出进行对比研究。结果表明,昆明TEC存在明显的春秋高值、夏冬低值的半年异常;白天高值、夜间低值的日变化特点突出,日峰值出现在06:30-08:00 UT(约13:00-15:00 LT);TEC随太阳活动减弱而明显下降,年平均峰值在2016-2019年分别为48,33,27,24 TECU;日峰值TEC与F10.7存在显著相关,月均值相关系数达到0.86,而与Ap指数则表现为弱相关;IRI-2020能较好地模拟昆明地区TEC的季节变化,但与观测值存在较大差异;均方根偏差值多集中在2~15 TECU,相对偏差百分比值主要在–85%~50%范围变化。对比结果表明IRI-2020的预测精度仍有待提高。

An artificial neural network model in predicting VTEC over central Anatolia in Turkey

This research investigates the capability of artificial neural networks to predict vertical total electron content(VTEC)over central Anatolia in Turkey.The VTEC dataset was derived from the 19 permanent Global Positioning System(GPS)stations belonging to the Turkish National Permanent GPS NetworkActive(TUSAGA-Aktif)and International Global Navigation Satellite System Service(IGS)networks.The study area is located at 32.6°E-37.5°E and 36.0°N-42.0°N.Considering the factors inducing VTEC variations in the ionosphere,an artificial neural network(NN)with seven input neurons in a multi-layer perceptron model is proposed.The KURU and ANMU GPS stations from the TUSAGA-Aktif network are selected to implement the proposed neural network model.Based on the root mean square error(RMSE)results from 50 simulation tests,the hidden layer in the NN model is designed with 41 neurons since the lowest RMSE is achieved in this attempt.According to the correlation coefficients,absolute and relative errors,the NN VTEC provides better predictions for hourly and quarterly GPS VTEC.In addition,this paper demonstrates that the NN VTEC model shows better performance than the global IRI2016 model.Regarding the spatial contribution of the GPS network to TEC prediction,the KURU station performs better than ANMU station in fitting with the proposed NN model in the station-based comparison.

基于KF-LSTM组合模型的短期电离层TEC预测

针对电离层总电子含量(TEC)时间序列非线性、非平稳性等特点,提出以卡尔曼滤波对电离层TEC数据进行预处理为基础,融合长短期记忆神经网络模型,构建KF-LSTM短期电离层组合模型预测TEC的方法,并利用该模型预测2016年和2018年4个时段的高、中、低纬度及赤道地区36个格网点提前1 d的电离层TEC。结果表明,KF-LSTM组合模型预测效果优于传统BP神经网络模型和单纯的LSTM模型。在赤道地区,其预测效果与C1PG模型相当;在15°~75°N地区,其预测效果优于C1PG模型。

2001号台风“黄蜂”路径电离层TEC异常分析

该文以2020年太平洋台风季首个被命名的风暴--强台风“黄蜂”为例,采用国际GNSS服务组织发布的全球垂直总电子含量数据,结合双线性插值、滑动时窗及四分位距的方法,分析台风发生发展过程前后,台风路径特征点上空电离层TEC的异常情况。研究发现,电离层异常集中发生在台风起编日及其前后各一日的时间,具体时段在0:00—10:00 UTC,异常属性全部为负异常,且扰动范围先增大后减小,综合判断认为2001号台风“黄蜂”是引起这几日台风路径上空电离层形扰的原因。

基于CEEMD的电离层TEC组合预报模型

针对单一电离层总电子含量(TEC)预报模型存在的缺陷,如受外界因素干扰较大、预报精度随预报时间的增加明显降低等,本文提出一种基于补充集合经验模态分解(CEEMD)电离层TEC组合预报模型。该模型实现电离层TEC预报的关键途径为:首先,利用CEEMD对TEC原始序列进行自适应分解,得到具有不同频率的分量并依据分量复杂度分析结果进行重构;其次,使用广义回归神经网络(GRNN)模型对高频分量进行建模与预报,使用Holt-Winters模型对低频分量进行建模与预报;最后,重构高频分量预报结果与低频分量预报结果得到电离层TEC预报值。根据太阳活动选取两段不同年积日、不同纬度电离层TEC序列进行实验,结果表明本文提出组合预报模型较单一的Holt-Winters模型、GRNN模型预报精度更高,在太阳活动平静期预报结果的平均相对精度为92.83%,在太阳活动剧烈期预报结果的平均相对精度为84.35%,对于长时间TEC预报也具有较好的效果,稳定性高。