IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 1 First Ionospheric Radio-Occultation Measurements From GNSS Occultation Sounder on the Chinese Feng-Yun 3C Satellite Tian Mao, Lingfeng Sun, Guanglin Yang, Xinan Yue, Tao Yu, Cong Huang, Zhongchao Zeng, Yungang Wang, and Jingsong Wang Abstract The Global Navigation Satellite System Occultation Sounder (GNOS) has been planned for the five Feng-Yun 3 series (FY3) weather satellites since 2013, the first of which, the FY3C satellite, was launched successfully at 03:07 UTC on September 23, 2013 from the Taiyuan Satellite Base, Shanxi province, China, into the orbit of 836-km altitude and 98.75 inclination. In addition to the Global Positioning System (GPS), the FY3C/GNOS is capable of tracking the occultation signal of the BeiDou Navigation Satellite System (BDS) (also called COMPASS) from space for the first time. The quality of BDS radio occultation (RO) has been verified in terms of signal-to-noise ratio. In this paper, the electron density profiles (EDPs) observed by FY3C/GNOS from bothgpsroandbdsro,whichwereprocessedandarchived in the National Satellite Meteorological Center of China Meteorological Administration, are compared with 32 globally distributed ionosonde observations, and then, we compare GPS RO EDPs with ionosonde observations at Mohe (52.0 N, 122.5 E), Beijing (40.3 N, 116.2 E), Wuhan (31.0 N, 114.5 E), and Sanya (18.3 N, 109.6 E). FY3C/GNOS EDPs show good agreement with ionosonde measurements, with larger discrepancies near the equatorial ionization anomaly region at Wuhan and Sanya. The ionospheric peak density (NmF2) and peak height (hmf2) derived from FY3C/GNOS EDPs are also compared with those obtained from the globally distributed ionosondes for the day of year 274 365 in 2013. In general, NmF2 and hmf2 have a higher correlation coefficient in the middle high latitude than in the lower latitude region, due to the difference of ionospheric horizontal inhomogeneity. We also compared the NmF2 and hmf2 maps between FY3C/GNOS and the International Reference Ionosphere 2012 (IRI-2012) model. However, the wavenumber-4 structure, Manuscript received January 14, 2015; revised June 13, 2015, October 29, 2015, December 31, 2015, and February 5, 2016; accepted February 17, 2016. This work was supported in part by the National High-tech R&D Program of China (863 Program) under Grant 2014AA123503 and in part by the Natural Science Foundation of China under Grant 41574178. T. Mao, C. Huang, Z. Zeng, Y. Wang, and J. Wang are with the Key Laboratory of Space Weather, National Center for Space Weather, China Meteorological Administration, Beijing 100081, China. L. Sun is with the Key Laboratory of Earth and Planetary Physics, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China (e-mail: sunlingfeng@mail.iggcas.ac.cn). G. Yang is with the Key Laboratory of Space Weather, National Center for Space Weather, China Meteorological Administration, Beijing 100081, China, with the Center for Space Science and Applied Research, Chinese Academy of Sciences, Beijing 100080, China, and also with the University of Chinese Academy of Sciences, Beijing 100049, China. X. Yue is with the COSMIC Program Office, University Corporation for Atmospheric Research, Boulder, CO 80301 USA. T. Yu was with the Key Laboratory of Space Weather, National Center for Space Weather, China Meteorological Administration, Beijing 100081, China. He is now with the China University of Geoscience, Wuhan 430000, China. Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TGRS.2016.2546978 which can be indicated clearly from FY3C/GNOS observations, could not be reproduced well by IRI-2012. Further investigations show that the nighttime EDPs have obvious ionization enhancement around the ionospheric E layer over the Aurora and the South Atlantic Anomaly regions due to the energetic particle precipitation indicated by the Space Environment Monitor observations onboard FY3C. Index Terms Feng-Yun 3C (FY3C)/Global Navigation Satellite System (GNSS) Occultation Sounder (GNOS), ionosonde, radio occultation (RO). I. INTRODUCTION THE story of radio occultation (RO) began at the dawn of interplanetary space exploration in the 1960s when a team of scientists from Stanford University and the Jet Propulsion Laboratory proposed to use the RO technique to probe the atmosphere of Mars based on the Mariner 3 and 4 satellites [1]. In the 1990s, when the constellation of Global Positioning System (GPS) was nearly accomplished and the corresponding techniques became mature, it was realized by the community that the RO technique could be used to sound the Earth s atmosphere and ionosphere based on the GPS signals [2], [3]. This concept was soon after demonstrated by the proof-of-concept GPS/Meteorology experiment in 1995[4]. This led to several successful additional missions hereafter, including CHAMP, GRACE, C/NOFS, FORMOSAT-3/COSMIC, Metop-A/B, TerrSAR-X/TanDEM-X, and so on. Of these missions, COSMIC was the first constellation of satellites dedicated primarily to RO and delivering RO data in near real time [5]. The Feng-Yun 3C (FY3C) satellite, equipped with the Global Navigation Satellite System (GNSS) Occultation Sounder (GNOS), is capable of tracking the occultation signal of the BeiDou Navigation Satellite System (BDS) (also called COMPASS) from space for the first time. GNSS RO measurement has been proven a powerful technique to remotely sense the Earth s atmospheric temperature and water vapor and ionospheric electron density [6]. In comparison with traditional ionosphere measurement techniques, such as ionosonde, rocket, and incoherent scatter radar, the GNSS RO measurement has the following advantages: 1) limb sounding geometry complementary to ground or space nadir viewing instruments; 2) high accuracy and precision; 3) high vertical resolution; and 4) full global coverage. Because the effect of the ionosphere on the radio wave is frequency dependent, the dual-frequency observations can be used to derive the total 0196-2892 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
2 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING Fig. 1. FY3C/GNOS configuration. electron content (TEC) along the GNSS ray with an accuracy of 2 3 TEC units [7]. Ionospheric scientific users are often interested in the altitudinal electron density profile (EDP), which is derived through an Abel inversion under certain assumptions based on slant TEC for each RO event. Yue et al. [8] [10] have done a series of simulations and real data evaluation to evaluate the RO EDP quality and assess its error distribution. Generally, the Abel inversion can give reasonable EDPs in the F and above region as well as peak height and density [9]. It has degraded performance in the region that has larger horizontal inhomogeneity such as the equatorial ionization anomaly (EIA) region and the lower altitude of the Aurora region with precipitation-induced ionization enhancement. Yue et al. [7] have summarized the details of RO ionospheric data processing, algorithms, and error distributions. The ionosondes can obtain vertical EDPs below the peak of the F 2 layer, which can be used to validate the RO measurement. Several investigators have compared ionospheric data measured by the RO measurement with those obtained by ionosonde and incoherent scatter radar observation located in the equatorial, low-latitude, and midlatitude regions in the past years [11] [22]. Their results show that NmF2 and hmf2 retrieved from COSMIC data are in agreement with the ISR and ionosonde measurements. This paper illustrates the first ionospheric RO measurements from the GNOS on the Chinese FY3C satellite. The main contents of this paper include the following: 1) systematic description of FY3C/GNOS and data processing and BDS RO signal quality assessment in terms of signal-to-noise ratio (SNR) in Section II; 2) FY3C/GNOS ionospheric EDP evaluation through comparison with ionosonde observations and the International Reference Ionosphere (IRI) model [23] in Section III; and 3) summary of the study results in Section IV. II. FY3C/GNOS DESCRIPTION Feng-Yun means Wind and Cloud in Chinese. Feng-Yun 3 series (FY3), which consists of a series of seven secondgeneration meteorological satellites in sun-synchronous polar orbit, is a follow-on mission of the China FY1 series satellites. The FY3 series represents a cooperative program between the China Meteorological Administration (CMA) and the China National Space Administration [24]. The first two satellites of FY3, named FY3A and FY3B, were launched in 2008 and 2010, respectively, with a designed life of two years and dedicated primarily to research and development. The FY3 series second batch of satellites consists of two weather satellites, FY3C and FY3D, with a designed life of five years and operational purpose. The FY3C is a morning sun-synchronous satellite with an orbit altitude of 836 km and an inclination of 98.75. Its orbit period is about 104 min, and the equatorial crossing time shifts between 10:00 and 10:20 [25], [26]. The FY3C satellite was launched at 03:07 UTC on September 23, 2013 from the Taiyuan Satellite Base, Shanxi province, China. The FY3D is planned to be an afternoon (14:00 local time (LT) over Beijing) sun-synchronous satellite and will be launched in 2016. Twelve different instruments make up the science payloads of the FY3C satellite. The instrument that interests us here is GNOS, which was developed by the Center for Space Science and Applied Research, Chinese Academy of Sciences. It consists of three physically separated antennas, the positioning antenna, the rising occultation antenna, and the setting occultation antenna (see Fig. 1). The positioning antenna is a widebeam low-gain antenna pointing at zenith with hemispherical coverage and a 1-Hz sampling rate. It can track up to six BDS satellites and more than eight GPS satellites simultaneously. Both occultation antennas are utilized to sample the ionospheric occultation in 1 Hz and the atmospheric occultation in 50 Hz of
MAO et al.: FIRST IONOSPHERIC RO MEASUREMENTS FROM GNOS ON CHINESE FENG-YUN 3C SATELLITE 3 Fig. 2. Distribution of (green triangles) ionosonde stations and (the black line) a random FY3C orbit with the position of tangent point (blue lines) of BDS RO events and (red lines) of GPS RO events. the phase-locked loop (PLL) and/or 100 Hz of the open loop. They can track up to four BDS satellites and six GPS satellites simultaneously. The detailed parameters of GNOS can be found in [25]. GNOS has been scheduled to fly on five of the FY3 series satellites, from FY3C to FY3G. FY3C/GNOS aims to obtain atmospheric temperature, density, air pressure [4], [27], and ionospheric electron density [2], [16], [28] through measuring the Doppler shift of radio signals. Several investigators have evaluated GNOS capability through simulations like the End-to-End GNSS Occultation Performance Simulator software and mountain-based experiments. Bai et al. [25] presented the instrument performance as derived from the analysis of measurement data in laboratory and mountain-based occultation validation experiments at Mount Wuling in Hebei province. The comparison showed that the atmospheric refractivity profiles obtained by FY3C/GNOS were consistent with those from the radiosonde. The rms of the differences between the GNOS and radiosonde refractivities is less than 3%. The most outstanding feature of FY3C/GNOS is that it can track BDS signals in addition to GPS signals from the low Earth orbit (LEO) for the first time. Both the phase and amplitude of B1 and B2 of BDS could be received in PLL [25]. In fully operational mode, FY3C/GNOS can observe 500 GPS ROs and 200 BDS ROs globally per day. Fig. 2 shows the distribution of the GPS and BDS occultation events in terms of tangent points of ionospheric EDPs during one FY3C orbit as an example. There are more BDS ROs occurring over the Asia region because, currently, BDS has been operational only around the Asia region. With the continuing launch of FY3 series satellites in the future, there will be more and more RO data contributed to the community from China. Before processing the data, it is important to look at the data quality of the measured data themselves before proceeding since the BDS RO signal is first observed. We checked the B1/B2 SNR for 169 atmosphere RO events and over 1500 ionosphere RO events, for we are focused on ionosphere results. Generally, they behave similarly to GPS. For example, they show significant oscillations if there is a sporadic E layer occurrence around 100 km. In the troposphere, the occultation SNR is significantly reduced due to the enhanced atmospheric absorption, water vapor refraction, and refractive index gradients from dry air near the surface [27]. Fig. 3 show two such examples. The left panel is an atmospheric occultation B1C SNR in 50 Hz versus straight-line height. We can see clearly an oscillation around 108 km due to the sporadic E layer. It indicates that FY3C/GNOS occultation SNR could be used to study the global ionospheric sporadic Fig. 3. Left panel shows typical SNR oscillations (20.1 S, 145.2 E, 2013/10/02 03:31 UT) of GNOS BDS B1C due to sporadic E layer around 108 km observed by the atmospheric occultation (50 Hz) antenna. The right panel shows typical SNR oscillations (13.7 N, 1.9 W, 2013/10/02 04:25 UT) of GNOS BDS B1C due to spread F observed by the ionospheric occultation (1 Hz) antenna. E layer occurrence [29]. The right panel is for an ionospheric occultation example with 1 Hz, which shows scintillation at the bottom of the ionosphere probably due to spread F. Please note that, although ionospheric and atmospheric ROs are tracked by the same physical antenna, they are actually processed by different electric systems in the receiver, which can explain the lower SNR of the ionospheric occultation than that of the atmospheric occultation [25]. The FY3C/GNOS data are officially processed and archived by the National Satellite Meteorological Center (NSMC) of CMA [26]. A detailed description of neutral atmospheric RO processing is beyond the scope of this paper. The ionospheric EDPs are retrieved by the standard Abel inversion [28]. Generally, the following assumptions are made in the retrieval: the straight-line propagation of the signal, circular LEO orbit, spherical symmetry electron density, and coplane of occultation [8]. The spherical symmetry assumption is believed to be the main retrieval error source. According to Yue et al. [7], [8], the Abel inversion error of EDP is systematic. It does not depend very much on the LEO orbit or the receiver type. A. Data III. RESULTS AND DISCUSSION The ionosonde data used in this paper are retrieved from 32 global distributed ionosonde stations, including 30 stations from the Digital Ionogram Data Base and two ionosonde stations from the Chinese Meridion Project [30]. The comparison was conducted for the interval of the day of year (DOY) 274 to 365 in 2013. All the ionograms used in this paper are manually scaled. In the comparison, we selected the FY3C/GNOS EDPs and ionograms following the criteria of Ely et al. [13]. Specifically, the maximum time difference between the two observations is 7.5 min, and the maximum space difference is 2.5. Furthermore, the ionograms with strong spread F or without well-defined traces were discarded. To avoid possible uncertainty that may arise from penetration electric fields, disturbance dynamo electric fields, and neutral wind, the data
4 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING Fig. 4. (a) Comparison of the FY3C/GNOS EDP (red solid lines, 129.74 E, 64.15 N, 10:33 UT) with those measured by the YAKUTSK ionosonde (blue solid line, 129.6 E, 62.0 N, 10:30 UT) and COSMIC/RO (black solid line, 130.25 E, 59.71 N, 10:36 UT) at almost the same time and location. The title gives the longitude and latitude of YAKUTSK. The error bars are the standard deviations for ionosonde data within a 1-h time bin (from t 30 min to t+30 min). The numbers in the right upper corner describe the DOY and universal time (UT). (b) Comparison of the EDPs observed by GPS RO of FY3C (red solid lines, 35.2 S, 16.8 E, 22:10 UT), BDS RO of FY3C (black solid line, 36.9 S, 17.1 E, 22:09 UT), and the Hermanus ionosonde (blue solid line, 34.4 S, 19.2 E, 22:15 UT) at almost the same time and location. The title gives the longitude and latitude of Hermanus. The error bars are the standard deviations for ionosonde data within a 1-h time bin. The numbers in the right upper corner describe the DOY and UT. for periods (until 3 h after or before) with a Kp index greater or equal to 5 were also discarded. B. Comparison With Ionosonde Fig. 4(a) describes a colocated EDP obtained by FY3C/GNOS (red solid line, 129.74 E, 64.15 N, 10:33 UT), COSMIC/RO (black solid line, 130.25 E, 59.71 N, 10:36 UT), and ionosonde (blue solid line, 129.6 E, 62.0 N, 10:30 UT) at YAKUTSK (62 N, 129.6 E) at almost the same time and location. As shown in this figure, the EDPs observed by the FY3C and COSMIC are within error bars of the ionosonde data below the F peak height. The error bar indicates the electron density difference between the maximum value and the minimum value at the same height from t 30 min to t+30 min. The electron densities obtained by FY3C are consistent with those measured by COSMIC at all heights. The rms of the differences between the FY3C and COSMIC EDPs (from 230 to 700 km, for in low altitude both EDPs are negative or very small) is less than 10%. The EDPs of the ionosonde are manually scaled using the SAO Explorer software package, and the profiles are automatically generated by the ARTIST program in the SAO software [31]. It should be mentioned that the ionosonde profile above the F-region peak is modeled and typically has some differences with real EDPs. The rms of the differences between the FY3C EDP and the ionosonde EDP from 230 km to the peak height is about 15%, and from 240 km to the peak height, it is about 10%. The rms of the differences between the COSMIC EDP and the ionosonde EDP is about 10% and 7%, respectively. Fig. 4(b) is the same as Fig. 4(a) except for the EDP of COSMIC replaced by that of BDS onboard FY3C (black solid line) at Hermanus (34.4 S, 19.2 E). It is shown that the electron profile observed by the BDS RO shows good agreement with that measured by the GPS RO. Moreover, the profiles of the RO measurement are consistent with that observed by ionosonde between 270 km and the peak height of the F2 layer. The rms of the differences between the FY3C GPS RO EDP and the ionosonde EDP from 270 km to the peak height is about 9%, and that between the FY3C BDS RO EDP and the ionosonde EDP is about 7%. There are still some differences on the EDPs observed by GPS RO, BDS RO, and the Hermanus ionosonde. It is mainly because of the differences of position of RO observation and the azimuth angle of RO. Then, next in Fig. 5, the FY3C EDPs are compared with those measured by the ionosonde at Mohe (the first column), Beijing (the second column), Wuhan (the third column), and Sanya (the fourth column). In Fig. 5, the first column is almost within the error bars of ionosonde observations at all heights, expect below the 200-km height where the FY3C data become negative. In the second column, the difference values obtained by the two techniques are close to each other, where the error is about 10 4 10 5 cm 3. The EDPs in the third and fourth columns are in the same order of magnitude but are different obviously. The FY3C NmF2 is larger (smaller) than those obtained from the ionosonde at Wuhan (Sanya). It is obvious that the differences become evident with decreasing latitude. This is because Wuhan and Sanya are in the northern crest of the EIA, while Mohe and Beijing are not. The EIA region has larger ionospheric horizontal gradients, which will result in more significant Abel inversion error [8]. In Fig. 6, we compare the NmF2 and hmf2 measured by 32 globally distributed ionosondes (the locations of ionosondes in both geographic and geomagnetic coordinates are listed in Table I) with those obtained by RO from FY3C during DOY 274 to 365 in 2013. The maximum difference is 7.5 min in time and 2.5 in space when selecting the colocated EDP pairs. In each panel, the black solid line represents the diagonal line, and the blue solid line is the linear fit to the data. The correlation coefficient R and the data number N are embedded in the right lower corner in each panel. The x-axis represents the value measured by the ionosonde, while the y-axis represents the value measured by FY3C. In the figure, the left, middle,
MAO et al.: FIRST IONOSPHERIC RO MEASUREMENTS FROM GNOS ON CHINESE FENG-YUN 3C SATELLITE 5 Fig. 5. Comparison of the (red solid lines) FY3C/GNOS EDPs with (blue circles) those measured by the ionosondes over Mohe, Beijing, Wuhan, and Sanya. Error bars are the standard deviations for ionosonde data within a 1-h time bin. The numbers in the right upper corner of each panel show the ionosonde name station, DOY, and UT. Fig. 6. Comparison of (up) NmF2 and (bottom) hmf2 obtained by FY3C/GNOS and those from ionosondes during DOY 274 365 in 2013. The left, middle, and right panels represent the geomagnetic latitude region of 90 S 90 N, 90 S 35 S, and 35 N 90 N, and 35 S 35 N, respectively. The black line represents the diagonal line, and the blue line is a linear fit to the data points. The values in the right lower corner of each panel show the correlation coefficient and the number of data points.
6 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING TABLE I LIST OF THE IONOSONDE STATIONS USED IN THE ANALYSIS and right panels represent the geomagnetic latitude region of 90 S 90 N, 90 S 35 S and 35 N 90 N, and 35 S 35 N, respectively. The top column and bottom column show the comparison results for NmF2 and hmf2, respectively. In the left panel, the correlation coefficients for NmF2 (top) and hmf2 (bottom) are 0.88 and 0.83, respectively. The agreement is a little better for NmF2 than for hmf2. The correlation coefficient for NmF2 is better than that obtained by Lei et al. [16] for COSMIC, which is probably because of the different criteria of data selection. Good agreement is found in the middle panels, in which the correlation coefficients for NmF2 and hmf2 are 0.95 and 0.82, respectively. The equal value line and the linear fit line are almost on top of each other in the top middle panel. It shows that the ionosonde NmF2 is consistent with the observation of FY3C/GNOS above 35 N or under 35 S. The correlation coefficients in the right panels for NmF2 and hmf2 are 0.82 and 0.77, respectively. The correlation coefficients in the middle high latitudes are larger than those in the middle lower latitudes. The correlations for the peak height are lower than those for the peak density. In the regions far away from the EIA, the horizontal gradients of the electron density and vertical gradients are lower than those in the EIA regions, so the retrieved error of electron density is smaller. Moreover, the FY3C/GNOS data and ionosonde data used in this paper are mainly in the northern winter. There are strong transequatorial neutral winds blowing from the summer hemisphere to the winter hemisphere at altitudes near the F2 peak, which tend to reinforce the gradients of electron density in the winter hemisphere. Therefore, the correlation of corresponding ionospheric parameters is lower in the middle lower latitude region than in the middle-to-high latitude region. What is more, in Fig. 7, we compare the NmF2 and hmf2 measured by ionosondes with those obtained by GPS RO and BDS RO from FY3C during DOY 274 to 327 in 2013. The criteria of data selection are the same as that in Fig. 6. In each panel, the black solid line represents the diagonal line, the blue squares are GPS RO observations, and the red stars are BDS RO observations. The values in the right lower corner of each panel show the correlation coefficients and the numbers
MAO et al.: FIRST IONOSPHERIC RO MEASUREMENTS FROM GNOS ON CHINESE FENG-YUN 3C SATELLITE 7 Fig. 7. Comparison of (left) NmF2 and (right) hmf2 of GPS RO and BDS RO with ionosondes during DOY 274 327 in 2013. The black line represents the diagonal line, the blue squares are GPS RO observations, and the red stars are BDS RO observations. The values in the right lower corner of each panel show the correlation coefficients and the numbers of data points. Fig. 8. Maps of noontime NmF2 derived from (upper panel) FY3C/GNOS measurements and the IRI model in the Northern Hemisphere (days 274 365 in 2013). The black solid line indicates the location of the magnetic equator. of data points. The x-axis represents the value measured by the ionosonde, while the y-axis represents the value measured by FY3C. The left column and right column show the comparison results for NmF2 and hmf2, respectively. In the left panel, the correlation coefficients for NmF2 (GPS RO) and (BDS RO) are 0.88 and 0.90. In the right panel, the correlation coefficients for hmf2 (GPS RO) and (BDS RO) are 0.90 and 0.94. The agreement is a little better for hmf2 than for NmF2. C. Comparison With IRI Figs. 8 and 9 show the comparison of corresponding ionospheric characters (NmF2 and hmf2) obtained by the FY3C and IRI model during the LT 11 to 14 of 274 to 365. Unfortunately, during the 11 to 14 LT interval, the ionospheric parameters obtained by the FY3C are mainly in the Northern Hemisphere because the satellite orbits in a solarsynchronization orbit. Fig. 8 gives the NmF2 maps as functions of geographic latitude and longitude from FY3C measurements (upper panel) and IRI model results (lower panel) in the Northern Hemisphere. The black line describes the location of the magnetic equator. In general, the FY3C measurements for NmF2 are consistent with the IRI results. However, there are indeed some differences. In the EIA region, FY3C/GNOS NmF2 is relatively higher at 60 E 120 E, 10 W 40 E, and 90 W 170 W than those at other longitude segments. The low-latitude ionosphere shows the remarkable longitudinal wavenumber-4 patterns [17], [32], [33]. Wan et al. [33] pointed out that it was recently noticed that the low-latitude ionosphere shows remarkable longitudinal wavenumber-4 patterns, i.e., the EIA crests are enhanced over West Africa, Southeast Asia, Central Pacific Ocean, and South America. Because FY-3C is a morning sun-synchronous satellite, its observation concentrates in the Northern Hemisphere, so we can only see three enhancements over West Africa, Southeast Asia, and Central Pacific Ocean, but we cannot see the enhancement over South America. In Fig. 8, the FY3C results are lower than those modeled by IRI, namely, the IRI results overestimate the peak density in the crest region, which was discussed by many researchers [16], [34]. The hmf2 maps of FY3C/GNOS measurements and IRI results are given in Fig. 9. In general, there is also a good agreement between the FY3C data and
8 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING Fig. 9. Same as Fig. 8 but for hmf2. the IRI model results. However, we can see that FY3C/GNOS hmf2 is relatively lower between 100 Wand 180 Wthan those modeled by IRI. D. Ionization Enhancement Around E Layer Due to the high inclination angle, FY3C/GNOS occultation should have a better coverage than that of COSMIC in the high latitude and polar region. According to Tsai et al. [35] and Yue et al. [10], the EDPs sometimes show significant ionization enhancement around the E layer in the Aurora region due to the energetic particle precipitation. In addition, there also exists the enhancement of energetic particle precipitation in the South Atlantic Anomaly (SAA) due to the magnetic field minimum there. These precipitations will cause the additional ionization enhancement of the ionospheric E layer [36]. Fortunately, FY3C also carries a Space Environment Monitor (SEM) instrument to measure the energetic particle precipitation flux at the orbit altitude. It enables us to verify the E layer enhancement above the Aurora and SAA regions. The bottom panel of Fig. 10 shows the global distribution of the energetic electron (0.15 0.35 MeV) flux obtained by FY3C/SEM. This is the lowest energy channel of FY3C energetic electron detection, whose energy is closest to precipitation electrons that penetrate down to the E layer altitude. As a comparison, we show three typical FY3C/GNOS EDPs at nighttime from the Auroral region, the SAA region, and a reference point. The latitude and longitude of the RO EDP at the E layer height are shown at the top of each panel. There are two electron density peak values in the Auroral zone and the SAA region at nighttime. One is at about 300 km, and the other is at 120 km. The lower peak in both profiles is probably due to the energetic particle precipitationinduced ionization enhancement. This result further validates the quality of FY3C/GNOS-derived EDPs. IV. SUMMARY GNSS RO technique has been a powerful technique in remotely sensing global ionosphere and atmosphere. China, represented by NSMC/CMA, is planning to make RO measurements in a series of five polar meteorological satellites, which Fig. 10. Top three panels show the EDPs in the nighttime obtained by FY3C/GNOS in the (the left panel) Auroral oval, (the middle panel) SAA, and (the right panel) reference data. The latitude and longitude of the RO EDP at the equivalent location are shown at the top of each panel. The numbers in the right upper corner of each panel show the DOY, LT, and UT. The bottom panel shows the global distribution of energetic electron (0.15 0.35 MeV) flux obtained by FY3C/SEM. will significantly contribute to the global RO observations in the future. The first satellite of this series, named FY3C, was launched in 2013, with a RO payload called GNOS. In addition to GPS, GNOS is capable of tracking the BDS signals from the LEO for the first time. The SNR test shows the reasonable quality of BDS measurements. In this paper, we first gave an overview of the FY3C/GNOS measurements. We then compared EDPs retrieved from FY3C/GNOS with those observed by ionosondes at Mohe, Beijing, Wuhan, and Sanya for a limited number of overhead passes. These preliminary comparisons show that there
MAO et al.: FIRST IONOSPHERIC RO MEASUREMENTS FROM GNOS ON CHINESE FENG-YUN 3C SATELLITE 9 is agreement between FY3C/GNOS profiles and ionosonde profiles in the middle-to-high latitude and there are obviously differences in the EIA region. Comparisons have also been made between ionospheric peak parameters retrieved by FY3C and those measured by globally distributed ionosondes, and agreement is also obtained in this case. The results indicate that NmF2 and hmf2 retrieved from FY3C/GNOS measurements are reliable and can be used for ionospheric physics studies. The comparison between the FY3C/GNOS data and the IRI model is also reasonably good, but the IRI model tends to overestimate NmF2 at the crests of the equatorial anomalies. Furthermore, FY3C/GNOS EDPs show ionization enhancement around the E layer during nighttime due to the energetic particle precipitation over the Aurora and SAA regions. We expect that the FY3 series RO data can make significant contributions to ionospheric studies and space weather monitoring. ACKNOWLEDGMENT The authors would like to thank the National Satellite Meteorological Center of the China Meteorological Administration for the radio occultation (RO) data of Feng-Yun 3C/GNOS and the Data Centre of Meridian Project in the Chinese Academy of Sciences for the access of ionosonde data. The authors would also like to thank DIDBase (http://umlcar.uml.edu/didbase) for the open access of the ionosonde data, the COSMIC Data Analysis and Archive Center (http://cdaac-www.cosmic.ucar. edu/) for the RO data, and the National Space Science Data Center (http://nssdc.gsfc.nasa.gov/) and D. Bilitza for making the International Reference Ionosphere model codes available. REFERENCES [1] T. P. Yunck, C. H. Liu, and R. Ware, A history of GPS sounding, Terr. Atmos. Ocean. Sci., vol. 11, no. 1, pp. 1 20, Mar. 2000. [2] G. A. Hajj and L. L. Romans, Ionospheric electron density profiles obtained with the Global Positioning System: Results from the GPS/MET experiment, Radio Sci., vol. 33, no. 1, pp. 175 190, Jan./Feb. 1998. [3] S. G. Jin, G. P. Feng, and S. Gleason, Remote sensing using GNSS signals: Current status and future directions, Adv. Space Res., vol. 47, no. 10, pp. 1645 1653, 2011. [4] C. Rocken et al., Analysis and validation of GPS/MET data in the neutral atmosphere, J. Geophys. Res., vol. 102, no. D25, pp. 29 849 29 866, Dec. 1997. [5] W. Schreiner, C. Rocken, S. Sokolovsky, S. Syndergaard, and D. Hunt, Estimates of the precision of GPS radio occultations from the COSMIC/FORMOSAT-3 mission, Geophys. Res. Lett., vol. 34, no. 4, pp. 1 5, Feb. 2007, DOI: 10.1029/2006GL027557. [6] R. A. Anthes, Exploring Earth s atmosphere with radio occultation: Contributions to weather, climate and space weather, Atmos. Meas. Tech., vol. 4, pp. 1077 1103, Jun. 2011, DOI: 10.5194/amtd-4-1077-2011. [7] X. Yue, W. S. Schreiner, Y. H. Kuo, D. C. Hunt, and C. Rocken, GNSS radio occultation technique and space weather monitoring, in Proc. 26th ION GNSS, Nashville, TN, USA, 2013, pp. 2508 2522. [8] X. Yue et al., Error analysis of Abel retrieved electron density profiles from radio occultation measurements, Ann. Geophys., vol. 28, no. 1, pp. 217 222, Jan. 2010. [9]X.Yue,W.S.Schreiner,C.Rocken,andY.H.Kuo, Evaluationof the orbit altitude electron density estimation and its effect on the Abel inversion from radio occultation measurements, Radio Sci., vol. 46, Feb. 2011, Art. no. RS1013, DOI: 10.1029/2010RS004514. [10] X. Yue et al., GNSS radio occultation derived electron density quality in high latitude and polar region: NCAR-TIEGCM simulation and real data evaluation, J. Atmos. Sol.-Terr. Phys., vol. 98, pp. 39 49, Sep. 2013b, DOI: 10.1016/j.jastp.2013.03.009. [11] Y. H. Chu, C. L. Su, and H. T. Ko, A global survey of COSMIC ionospheric peak electron density and its height: A comparison with ground-based ionosonde measurements, Adv. Space Res., vol. 46, no. 4, pp. 431 439, Aug. 2010. [12] Y. J. Chuo, C. C. Lee, W. S. Chen, and B. W. Reinisch, Comparison between bottomside ionospheric profile parameters retrieved from FORMOSAT3 measurements and ground-based observation collected at Jicamarca, J. Atmos. Sol.-Terr. Phys., vol. 73, pp. 1665 1673, Aug. 2011. [13] C. V. Ely, I. S. Batista, and M. A. Abdu, Radio occultation electron density profiles from the FORMOSAT-3/OSMIC satellites over the Brazilian region: A comparison with Digisonde data, Adv. Space Res., vol. 49, pp. 1553 1562, Jun. 2012. [14] L. H. Hu et al., Comparison between ionospheric peak parameters retrieved from COSMIC measurement and ionosonde observation over Sanya, Adv. Space Res., vol. 54, pp. 929 938, Sep. 2014. [15] M. C. Kelley et al., Comparison of COSMIC occultation-based electron density profiles and TIP observations with Arecibo incoherent scatter radar data, Radio Sci., vol. 44, Art. no. RS4011, Aug. 2009, DOI: 10.1029/2008RS004087. [16] J. Lei et al., Comparison of COSMIC ionospheric measurements with ground-based observations and model predictions: Preliminary results, J. Geophys. Res., vol. 112, Jul. 2007, Art. no. A07308, DOI: 10.1029/ 2006JA012240. [17] G. Liu, T. J. Immel, S. L. England, K. K. Kumar, and G. Ramkumar, Temporal modulations of the longitudinal structure in F2 peak height in the equatorial ionosphere as observed by COSMIC, J. Geophys. Res., vol. 115, Apr. 2010, Art. no. A04303, DOI: 10.1029/2009JA014829. [18] J. Y. Liu, C. C. Lee, J. Y. Yang, C. Y. Chen, and B. W. Reinisch, Electron density profiles in the equatorial ionosphere observed by the FORMOSAT-3/COSMIC and a digisonde at Jicamarca, GPS Solut., vol. 14, pp. 75 81, Jan. 2010, DOI: 10.1007/s10291-009-0150-3. [19] L. Liu et al., Features of the middle and low latitude ionosphere during solar minimum as revealed from COSMIC radio occultation measurements, J. Geophys. Res., vol. 116, Sep. 2011, Art. no. A09307, DOI: 10.1029/2011JA016691. [20] Y. Sahai et al., Effects observed in the equatorial and low latitude ionospheric F-region in the Brazilian sector during low solar activity geomagnetic storms and comparison with the COSMIC measurements, Adv. Space Res., vol. 50, no. 10, pp. 1344 1351, Nov. 2012. [21] L. C. Tsai and W. H. Tsai, Improvement of GPS/MET ionospheric profiling and validation using the Chung-Li ionosonde measurements and the IRI model, Terr. Atmos. Ocean. Sci., vol. 15, pp. 589 607, Nov. 2004. [22] L. F. Sun, B. Q. Zhao, X. A. Yue, and T. Mao, Comparison between ionospheric character parameters retrieved from FORMOSAT3 measurement and ionosonde observation over China, (in Chinese), Chin. J. Geophys., vol. 57, pp. 3625 3632, Nov. 2014. [23] D. Bilitza et al., The International Reference Ionosphere 2012 a model of international collaboration, J. Space Weather Space Clim., vol.4, pp. 1 12, Jul. 2014, DOI: 10.1051/swsc/2014004. [24] Z. Meng, Y. Sun, and Y. Dong, FY-3 China s second generation polar orbit meteorological satellite, presented at the Proceedings 53rd IAC World Space Congress, Houston, TX, USA, Oct. 10 19, 2002, paper IAC-02- B.2.10. [25] W. H. Bai et al., An introduction to the FY3 GNOS instrument and mountain-top tests, Atmos. Meas. Tech., vol. 7, pp. 1817 1823, Jun. 2014. [26] Y. M. Bi et al., An introduction to China FY3 radio occultation mission and its measurement simulation, Adv. Space Res., vol. 49, no. 7, pp. 1191 1197, Apr. 2012, DOI: 10.5194/amt-7-1817-2014. [27] E. R. Kursinski, G. A. Hajj, J. T. Schofield, and R. P. Linfield, Observing Earth s atmosphere with radio occultation measurements using the Global Positioning System, J. Geophys. Res., vol. 102, pp. 23 429 23 465, Oct. 1997. [28] W. S. Schreiner, S. V. Sokolovsky, C. Rocken, and D. C. Hunt, Analysis and validation of GPS/MET radio occultation data in the ionosphere, Radio Sci., vol. 34, pp. 949 966. Jul./Aug. 1999. [29] C. Arras et al., A global climatology of ionospheric irregularities derived from GPS radio occultation, Geophys. Res. Lett., vol. 35, no. 14, Jul. 2008, Art. no. L14809, DOI: 10.1029/2008GL034158. [30] C. Wang, New chains of space weather monitoring stations in China, Space Weather, vol. 8, Aug. 2011, Art. no. s08001, DOI: 10.1029/ 2010SW000603. [31] M. K. Grigori et al., Exploring digisonde ionogram data with SAO- X and DIDBase, in Proc. AIP Conf., 2008, vol. 974, p. 175, DOI: 10.1063/1.2885027. [32] T. J. Immel et al., Control of equatorial ionospheric morphology by atmospheric tides, Geophys. Res. Lett., vol. 33, no. 15, Aug. 2006, Art. no. L15108, DOI: 10.1029/2006GL026161.
10 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING [33] W. Wan et al., Wavenumber-4 patterns of the total electron content over the low latitude ionosphere, Geophys. Res. Lett., vol. 35, no. 12, Jun. 2008, Art. no. L12104, DOI: 10.1029/2008GL033755. [34] G. Jee, R. W. Schunk, and L. Scherliess, Comparison of IRI-2001 with TOPEX TEC measurements, J. Atmos. Sol.-Terr. Phys., vol. 67, no. 4, pp. 365 380, Mar. 2005. [35] H. F. Tsai, J. Y. Liu, C. H. Lin, and M. L. Hsu, FORMOSAT-3/COSMIC observations of the ionospheric auroral oval development, GPS Solut., vol. 14, pp. 91 97, Jan. 2010, DOI: 10.10 07/s10291-0 09-0137-0. [36] M. A. Abdu, I. S. Batista, A. J. Carrasco, and C. G. M. Brum, South Atlantic magnetic anomaly ionization: A review and a new focus on electrodynamic effects in the equatorial ionosphere, J. Atmos. Sol.-Terr. Phys., vol. 67, pp. 1643 1657, Dec. 2005. Tian Mao was born in Hubei province, China, in 1981. She received the B.Sc. degree in Physics from Central China Normal University, Wuhan, China, in 2002 and the Ph.D. degree in space physics from the Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan, China, in 2007. Since 2007, she has been with the National Center for Space Weather, China Meteorological Administration, Beijing, China, where she is currently an Associate Researcher. Her research interests include ionospheric remote sensing and 3-D structure of the ionosphere. Lingfeng Sun was born in Inner Mongolia province, China, in 1975. He received the Ph.D. degree in space physics from the Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan, China, in 2007. Since 2007, he has been an Associate Researcher in the Key Laboratory of Earth and Planetary Physics, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China. His research interests are in ionosphere physics and forecasting of space weather. Guanglin Yang received the M.S. degree in atmosphere physics from Peking University, Beijing, China, in 2002. He is currently a Senior Engineer in ionosphere observation, working in the National Space Weather Center, China Meteorological Administration, Beijing. His research interests include Global Navigation Satellite System (GNSS) radio occultation, GNSS/MET, and coupling of upper atmosphere and ionosphere. Xinan Yue received the Ph.D. degree in space physics from the Graduate School of the Chinese Academy of Sciences, Beijing, China, in 2008. He is currently a Project Scientist in the Constellation Observing System for Meteorology, Ionosphere, and Climate Program Office, University Corporation for Atmospheric Research, Boulder, CO, USA, where he is responsible for the ionospheric data process and evaluation. His scientific interests include ionospheric/thermospheric modeling, data assimilation, Global Navigation Satellite System applications, remote sensing, and space weather. He has published nearly 80 science-citation-index papers as either first author or coauthor in related fields. Tao Yu received the Ph.D. degree in space physics from the Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences,Wuhan, China, in 2003. From 2003 to 2014, he was a Research Assistant, Associate Researcher, and Researcher in space physics, respectively. His research interests include modeling and Global Navigation Satellite System monitoring of ionosphere. Since 2015, he has been working as a Professor with the China University of Geosciences. Cong Huang received the Ph.D. degree in astrophysics from Peking University, Beijing, China, in 2006. Since 2006, he has been with the Division of Space Weather, National Satellite Meteorological Center, China Meteorological Administration, Beijing, as a Research Associate. His research interests are in the space environment monitoring of low-earth-orbit satellites, the space weather forecast model building, the observation data assimilation, and the climate change induced by space weather. He is the author or coauthor of more than ten science citation index and ISTP papers published in astrophysics and space physics journals and presented at international conferences. Zhongchao Zeng received the Ph.D. degree in electronic engineering from the Institute of Electronics, Chinese Academy of Sciences, Beijing, China, in 2011. Since then, he has been an Associate Researcher with the National Satellite Meteorological Center, China Meteorological Administration, Beijing, China. His research interests include radio wave propagation in ionosphere and ray tracing. Yungang Wang received the M.S. degree in particle astrophysics from the Institute of High Energy Physics, Chinese Academy of Sciences, Beijing, China, 2007. He is currently a Senior Engineer in ionosphere observation, working in the National Space Weather Center, China Meteorological Administration, Beijing. His research interests include ionosphere observation, ionospheric irregularity, and HF communication. Jingsong Wang was born in Hubei, China, in 1970. He received the B.S and M.S. degrees in space physics from Peking University, Beijing, China, in 1991 and 1994, respectively, and the Ph.D. degree in space physics from the Chinese Academy of Sciences, Beijing, China, in 1997. Since 2009, he has been the Director-General of the National Center for Space Weather, China Meteorological Administration, Beijing. His research interests include terrestrial and Martian ionospheres and space weather sciences.
本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 (www.xuebalib.com) 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具