the HY-2 satellite and Technology,Nanjing, Jiangsu, PRChina Jiangsu,PRChina

Transcription

1 Indian Journal of Geo-Marine Sciences Vol. 4(2), February 216, pp The ionospheric VTEC inversion and results analysis based on the HY-2 satellite Xinzhi Wang 1,2,3 Dongjie Yue 3 & Fuyang Ke 2 1 Jiangsu Key Laboratory of Meteorological Observation and Information Processing,Nanjing University of Information Science and Technology,Nanjing,2144 Jiangsu, PRChina 2 School of Geography and Remote Sensing, Nanjing University of Information Science and Technology, Nanjing, 2144 Jiangsu,PRChina 3 School of Earth Sciences and Engineering, Hohai University,Nanjing,2198 Jiangsu, PRChina. *[ wangxinzhi@126.com] Received 17 October 214; revised 14 January 21 In this paper, a new data processing method and flow is put forward to calculate the ionospheric VTEC inversely according to the features of the HY-2 satellite radar altimeter data. In order to verify the accuracy and feasibility of the data processing method and flow, the paper compares and analyzes the processed data from the aspects of time, data amount, latitude and longitude. HY-2 VTEC is also compared with the GIM VTEC which is interpolated from the GIM, and it is found that there is a good agreement between the HY-2 VTEC and the GIM VTEC. It shows that the data processing method and flow is suitable and correct. Meanwhile, the differences between HY-2 VTEC and GIM VTEC are larger in the areas where the IGS observation stations are fewer, while the agreement is good and the differences are small in the areas where the IGS observation stations are more. [Keywords: HY-2, Ocean Satellite, GIM, Ionospheric, VTEC] Introduction Global navigation satellite system (GNSS) is one of the most important means of probing the ionosphere 1. Global Ionospheric Map (GIM) is a global ionospheric product which is computed by the International GNSS Service (IGS) using the daily observation data from the global distributed GNSS observation stations, and is published in the Ionosphere map exchange format (IONEX). GIM has wide applications in the areas of navigation, precise point positioning, pre-earthquake ionospheric anomalies detection and ionospheric model 2-6, et al. However, GNSS ground stations are not evenly distributed globally, particular in ocean areas, which leads to relatively low accuracy and reliability of GIM in these areas. Ocean satellite is one of the earth observation satellites dedicated to observe the ocean phenomena and processes, and it can provide the global ocean dynamic information for monitoring ocean environment and developing ocean resources and research, et al. The use of ocean satellites to detect ionosphere can effectively compensate the lacks of GNSS observation stations in the ocean areas. Since the United States and France launched the TOPEX/Poseidon ocean satellite in 1992, many scholars have carried out research by using the TOPEX/Poseidon to detect the ionosphere. David A. Imel (1994) made use of GPS data to evaluate the Ku-band and C-band ionospheric corrections of the TOPEX/Poseidon satellite radar altimeter. M.V. Codrescua et al (1999) researched the ionospheric VTEC

2 198 INDIAN J MAR SCI VOL 4, NO.2 FEBRUARY 216 calculated inversely by using the TOPEX/Poseidon satellite radar altimeter data 7.Y.O. MigoyaOrue et al (28), Francisco Azpilicueta et al (29), and Yu.V. Yasyukevich et al (21) researched the differences of ionospheric VTEC from TOPEX/Poseidon, International Reference Ionosphere (IRI), GPS and GIM 8-1. S. Todorova (27, 28), E. Yizengaw (28), M. M. Alizadeh (211), Denise Dettmering (211), Chen peng (213) fused multi-source data from GNSS, TOPEX/Poseidon and Jason-1 ocean satellite, COSMIC and CHAMP to research the global ionospheric model, and the results indicated that the fusion effectively improved the accuracy and reliability of GIM in ocean areas Compared with the Western countries, China started late in ocean satellite project. On August 16, 211, China launched the first domestic ocean dynamic environment satellite named HY-2. Satellite is equipped with radar altimeter and many other types of detective equipment and is used to detect the ionosphere. At present, there has been no research about the ionospheric VTEC inversion using the HY-2 satellite in China. Latitudes and longitudes of the same pass in different cycles of the HY-2 satellite are basically the same, while the time for observation is a bit different. According to this characteristic, this paper puts forward a data processing method and flow by using the HY-2 satellite radar altimeter data to calculate the ionospheric VTEC data inversely. In order to illustrate the accuracy and feasibility of the data processing method and flow, the preprocessed data are compared and analyzed from the aspects of time, data amount, latitude and longitude. With the comparison between the HY-2 VTEC, which is calculated from the HY-2 altimeter data inversely, and the GIM VTEC, which is interpolated from the GIM, it is found that there is a good agreement between HY-2 VTEC and GIM VTEC. It also indicates that the data processing method and flow is suitable and correct. Data processing method and flow putted forward in this paper has important referencial significance for the ionosphere research of using the HY-2 satellite radar altimeter data. Using the domestic ocean satellite to research the detection of ionosphere can give full play of the ocean satellite s benefits and is also good to expand the application and research fields of the ocean satellite. Materials and Methods HY-2 satellite The orbit of HY-2 satellite is sun-synchronous, the altitude is 971km and the inclination is Satellite, equipped with Radar Altimeter, Microwave Scatter meter, Scanning Microwave Radiometer, Correction Microwave Radiometer, DORIS, Dual Frequency GPS and Laser Range Finder, is used to measure sea surface height, significant wave height, wind speed and other basic elements of the ocean. The corrected accuracy of sea surface measurement can be less than 8cm. When the satellite completes a whole ascent or descend trajectory, it makes a pass. Different passes are named with different numbers. When the pass is in an ascent trajectory, the number is single; when the pass is in a descend trajectory, the number is double. Satellite running around the earth for a circuit contains two passes and each pass will cost about 2.2 minutes. Period for the satellite to repeatedly visit the same area is 14 days. Data of the satellite radar altimeter are stored in cycles which are also named in numbers. Each cycle contains 386 passes data. Because there is only one HY-2 satellite, the observed time and area of each pass in the same cycle are not the same. However, the orbit is sun-synchronous, so the observed time and area in the repeated cycles are basically the same. Data products of the HY-2 satellite radar altimeter Data of the HY-2 satellite radar altimeter are divided into 4 levels: they are level, level 1, level 2 and level 3. Level is also called as crude data since it can generate the other three levels after pretreatment, data inversion and statistical average. Level 1 contains two types of 1A and 1B. 1A is one kind of data conducted by time marking

3 WANG et al: THE IONOSPHERIC VTEC INVERSION AND RESULTS ANALYSIS BASED ON THE HY-2 SATELLITE 199 and geographical positioning. 1B is another type of data with locational and descriptive information, conducted by pass dividing, FFT format conversing, height tracking and slope value format conversing. Level 2 is conducted from level 1 by inverse calculation, sea and land marking and quality controlling. Meanwhile level 2 includes three different types: Interim Geophysical Data Records (IGDR), Sensor Geophysical Data Records (SGDR) and Geophysical Data Records (GDR).Level 3 is gridded from level 2 by monthly, quarterly and annual averaging and it includes significant wave height, sea surface wind speed and gridded sea surface height anomaly and monthly, quarterly, annual averaging data.in this paper, we mainly use the IGDR and GDR data. Analysis of the HY-2 satellite radar altimeter To analyze the same pass data from different cycles, 1 cycles of pass 33 from cycles 42-1 are selected at random. Table 1 shows the related information of pass 33 of the 1 cycles. From the table, it can be found that the start and end time of these 1 cycles are not the same. Sampling frequency of the HY-2 satellite radar altimeter is 1 second, and the amount of seconds between the start and end observed time of each cycle is the theoretical amount,which is relatively stable or has nuance and is coincide with the runtime ofthe pass (2.2 minutes * 6 seconds). But the actual amount is very different. It is found that there are no data in some epochs after checking the original data. Latitude direction of the start observation is relatively stable, but longitude is slightly different. However, Latitude direction of the end observation is slight different because the end time for observation is different. Considering this factor, latitude direction of the end observed location is also regarded as stable, but longitude direction is slightly different too. In order to analyze the data of different passes from the same cycle, 6 sets data of passes from cycle is selected. Table 2 shows the related information of the 6 sets data. From the table, it can be found that the single number pass moves from the southern hemisphere to the northern, and it shows an ascent trajectory. While the double number pass moves from the northern hemisphere to the southern, and it shows a descend trajectory. The observed time of each pass is different from the start and end observed time. Meanwhile the observed areas of each pass are also not the same by the latitude and longitude. These differences can be found visually from Figure 1. 4 N 18 W 13 W 9 W 4 W 4 E 9 E 13 E 18 E 4 S 9 N 9 S Fig.1 Paths and locations of passes in cycle Attention: In Figure 1, different passes are showed in blue line, and from left to right they are passes 16, 14, 12, and 1,13,11. Cycle Start time Endtime Table1-Information of different cycles in pass33 Start position End position Latitude Longitude Latitude Longitude Real amount Theoretical amount 42 ::3 1:42: ::11 1:42: ::18 1:42: ::29 1:42: ::38 1:42: ::42 1:42: ::49 1:43:

4 2 INDIAN J MAR SCI VOL 4, NO.2 FEBRUARY ::8 1:43:2 :1:4 1:43:16 1 :1:7 1:43: Table2-Information of different passes in cycle Cycle Start time End time 11 1:39:37 11:31: 12 11:31:1 12:24: :24: 13:16: :16:21 14:8: :8:32 1::2 16 1::46 1:2:9 Start position End position Latitude Longitude Latitude Longitude Real amount Theoretical amount Inversion method and data process of HY-2 satellite radar altimeter Inversion method of HY-2 satellite radar altimeter HY-2 satellite radar altimeter,with dual frequency operation mode of 13.8GHZ (Ku-band) and.2ghz (G-band), gets the height from satellite to the sea surface by measuring the spreading timeof the electromagnetic wavepropagated from its antenna to the sea surface. Because ionosphere is dispersive medium, when electromagnetic waves spread in it, each frequency will spread atdifferentt speeds. So the ionosphericvtec can be calculated inverselybased on the ionosphericcorrections through formula: 2 dr f VTEC 43 dual-frequency the following (1) In this formula, the unit of VTEC is TECU; dr is the ionosphericcorrectionn of different frequency, and the unit of itis mm; fis frequency of the electromagnetic waves,and the unit of it is GHZ. It is noticed that the distances observed by the HY-2 satellite radar altimeter are vertical distances between the satellite and the sea level. So the results are the VTEC which does not need to be calculated by the projection function. Data process of HY-2 satellite radar altimeter HY-2 satellite is mainly used to observe the ocean areas. Ionospheric corrections in these areas are negative, while in the land areas the correctionsare usually set to be default or non-negative. So the corrections in the land areas should be removed during data processing. In addition,theoretical amount of thepass is 3133, and the amount of the land areas being removed,it normally keepsmore than 2, which is still a large number. In order to reduce the amountand eliminate data errors, thedataof the HY-2 satellite radar altimeter need to be processed. There are five steps of the data processing. (1) Extracting the ionospheric corrections and removingthe default and non-negative values. Since the accuracy of the Ku-band ionospheric correctionsis higherthanthe C-band, the Ku-band corrections is often chosen to be calculated. (2) Detecting the gross errors of the extracted ionospheric corrections with (variance) as a criterion. When the default and non-negative values are removed, the data still contain gross errors. In this circumstance, detecting the gross errors can minimize its impact on the final results. (3) Smoothing the detected data. Smoothing can effectively eliminate the influences of the random errors and is conducted mainly by time intervals. Theintervals can beset independently. In this paper, it is 2 seconds, s, whichmakes that the latitude direction can be divided intoseveral zonesaccording to 1 and makes it easier to

5 WANG et al: THE IONOSPHERIC VTEC INVERSION AND RESULTS ANALYSIS BASED ON THE HY-2 SATELLITE 21 contrast and analyze. (4) Calculating the observed times, latitudes and longitudesof the smoothed data. The calculation mainly adopts the method of averaging. () Calculating the ionospheric VTEC. The ionospheric VTEC is calculated inversely by using the formula (1). Results and Discussion Global ionosphere maps Since 1998, the ionosphere working group of the IGS has been using the daily observed data from global distributed GNSS(GPS/GLONASS) observation stations to calculate thegim and publishing it in the format of IONEX. GIM gives the worldwide VTEC in latitude and longitude grids. The time resolution of GIM is 2 hours, and the spatial resolution is 2. in latitude and in longitude. In order to compare and analyze the HY-2 VTEC calculated from the HY-2 satellite inversely, the study makes use of the GIM published by Center for Orbit Determination in Europe (CODE) and uses interpolation to calculate the GIM VTEC along the passes of HY-2 satellite in time, latitude and longitude. The method of interpolation is four-point interpolation formula in reference17. GIM is a global ionospheric measurement model, it has good reliability and is used widely. Table 1 shows the GIM RMS of the same location (latitude 3.3, longitude 13) in different days and different times in March -April213. From table 3, it can be seen that the accuracy of GIM model changed little and basically uniform. So the GIM VTEC can be used to compare with the HY-2 VTEC. Table 3-GIM RMS of the same location in different days and different times(unit:tecu) Times (UTC) March 31st April 4th April 8th April 12th April 16th : : : : : : : Data analysis In order to analyze the changes of the ionosphere in the same period and area, it is necessary to analyzethe same pass data in different cycles. This study selectsthe pass33 incycles 33-7 at random and uses the IGDR data to process. Due to the data of pass 33 in cycles 4 and 41 are blank, the data are 23 groups in all. Route of pass33 in ocean areas is shown in Figure2. After data processing, the data of each cycle achieves better unity in time, data amount, latitude and longitude.table4 lists the th data of Cycles It can be seen from the table that the maximal difference values are.187 and.11 separately in latitudeand longitude directions. So the spatial locations are in good agreement. Maximal differences of observed times are 64 seconds, and the ionosphere will keep stably in a short period. So theobserved time of the data is also in good agreement. Amounts of the results in the 1 cyclesare same, which indicates that the processed results are satisfactory. different passes and 12 different cycles are selected to be analyzed. They are pass1 in cycles 42-4, pass21 in cycles -3, pass33 in cycles 42-4,pass16 in cycles 42-4 and pass1 in cycles -3. Passes 1, 21 and 33 are ascent trajectory while passes 16 and 1are decent trajectory.

6 22 INDIAN J MAR SCI VOL 4, NO.2 FEBRUARY 216 Table shows the observed date and time of passes 1,16 and 33 in different cycles and Table 6 shows those of the passes 1 and 21. The observed date and time in tables 4 and are Universal Time (UT). Figure 2 shows distributions of the global IGS observation stations and paths of the different passes. 9 N 4 N 18 W 13 W 9 W 4 W 4 E 9 E 13 E 18 E 4 S Table 4-The processed results of the data Cycle Time Latitude Longitude Amount 48 1:6: :6: :6: :6: :7: :7: :7: :7: :7: :7: Because the data of the passes from the HY-2 satellite radar altimeter is blank, so the passes are blank in high latitude sea areas, south of the earth in Figure 2. This maybe that the weather is not good in that day or others. Fig.2-Distributions of the global IGS observation stations and routes of different passes Attention: In Figure 2, the blue lines are different passes, and from left to right there are passes 1,16,33,1 and 21;the red points are IGS stations Table-Observed date and time of passes 1, 16 and 33 in different cycles Date Start and Pass Cycle Cycle Cycle Cycle end time /27 /11 /2 6/8 14:7--14:9 16 4/27 /11 /2 6/8 14:9--1:1 33 /8 /22 6/ 6/19 :--1:42 Table6-Observed date and time of passes 21 and 1 in different cycles Pass 9 S Cycle Cycle 1 Date Cycle 2 Cycle 3 Start and end time 21 8/24 9/7 9/21 1/ 8:2--8: 1 8/2 9/3 9/17 1/1 16:7--17:. 2 Cycle 42 PASS 1 3 Cycle Cycle 44 2 Cycle HY2 VTEC GIM VTEC Fig.3-Comparison and analysis of HY-2 VTEC and GIM VTEC of pass1

7 WANG et al: THE IONOSPHERIC VTEC INVERSION AND RESULTS ANALYSIS BASED ON THE HY-2 SATELLITE 23 8 Cycle PASS 21 Cycle Cycle 2 Cycle HY2 VTEC GIM VTEC Fig.4-Comparison and analysis of HY-2 VTEC and GIM VTEC of pass 21 Cycle 42 PASS 33 Cycle Cycle 44 6 Cycle Fig.-Comparison and analysis of HY-2 VTEC and GIM VTEC of pass33 Attention: In Figures3-, abscissa represents latitude, ordinate represents VTEC (unit: TECU); the vertical dashed line is the equatorial position. The orbit altitude of the HY-2 satellite is 971 km, whilethe GPS is more than 2, km.so the GIM VTEC islarger than the HY-2 VTEC, because the GIM VTEC contains the VTECofthe regions between the orbitsof HY2 and GPS. It can be seen fromfigures 3- that the GIM VTEC is larger than the HY-2 VTECin most cases. This is HY2 VTEC GIM VTEC consistent with the theory. In addition, wefindthat the tendency of HY-2 VTEC is basically the same as GIM VTEC. There is a fluctuating phenomenon of HY-2 VTEC near the equator in cycles of Figure3, in cycles -3 of Figure4 and in cycles 43-4 of Figure. This coincides with the equatorial anomaly found in

8 24 INDIAN J MAR SCI VOL 4, NO.2 FEBRUARY 216 the reference 18. In Figure 3, the HY-2 VTECof Cycles 42 and 43 arelarger than the GIM VTECin north areas of the equator, and this is inconsistent with the theory. The situations of HY-2 VTEC larger than GIM VTEC also appear in latitude areas higher than S of Cycles 1 and 3 in Figure 4 and in mid and high latitude areas of Cycles42, 44 and 4 in Figure. It can be seen from Figure 2 that there are less IGS stations near pass 21, while the distances between the pass and the near IGS stations are large. So the accuracy of the GIM VTEC in these areas is low. 2 Cycle 42 PASS 16 Cycle Cycle 44 3 Cycle HY2 VTEC GIM VTEC Fig.6-Comparison and analysis of HY-2 VTEC and GIM VTEC of pass 16 Cycle PASS 1 Cycle Cycle 2 4 Cycle HY2 VTEC GIM VTEC Fig.7-Comparison and analysis of HY-2 VTEC and GIM VTEC of pass1 Attention: In Figure 7, abscissa represents latitude, and ordinate represents VTEC (unit: TECU); the vertical dashed line is the equatorial position

9 WANG et al: THE IONOSPHERIC VTEC INVERSION AND RESULTS ANALYSIS BASED ON THE HY-2 SATELLITE 2 In Figure 6, the tendency of HY-2 VTEC in the latitude areas lower than 18.6 S is consistent with the GIM VTEC. However, both HY-2 VTEC and GIM VTEC have larger fluctuations in latitude areas higher than 18.6 S. In Figure 7, the HY-2 VTEC also has large fluctuations. The differences between HY-2 VTEC and GIM VTEC are larger in latitude areas higher than S. It can be seen from Figure 2 that passes 16 and 1 locate in the Pacific Ocean area and the IGS observation stations are very scarce near the passes. So the accuracy of the GIM VTEC in these areas is low. Distributions of land and sea are different in north and south hemisphere. The land is wider in north hemisphere and the sea is wider in south. As it can be seen from Figure 2 that the land of the route of passes 1 and 33 are wider in north hemisphere and the ocean is more in south. Because the ocean satellites only observe the oceans, the data amounts of passes 1 and 33 in south hemisphere are bigger than those in the north in Figures 3 and. This is consistent with the status reflected in Figure 2. Intermediate areas of pass 21 include lands, so the data of pass 21 appear discontinuities in Figure 4. Conclusions In this paper, HY-2 satellite radar altimeter data are detailed analyzed. It is found that latitudes and longitudes of the same pass are basically the same, but the observed time has a bit differences. According to this feature, a new data processing method and flow are put forward to inverse the altimeter data into ionospheric VTEC. Then the paper compares and analyses the preprocessed data from the aspects of time, data amount, latitude and longitude, and verifies the accuracy and feasibility of the data processing method and flow. The HY-2 VTEC is compared with the GIM VTEC interpolated from the GIM, and it is found that there is a good agreement between the HY-2 VTEC and the GIM VTEC. It also shows that the data processing method is suitable and correct. Meanwhile, it finds that the differences between HY-2 VTEC and GIM VTEC are larger in the areas of fewer IGS observation stations. In the areas of more IGS observation stations, however, the differences are smaller. The trends of HY-2 VTEC and GIM VTEC are good. Acknowledgements This research work was supported by the open project of Jiangsu Key Laboratory of Meteorological Observation and Information Processing (KDXS149), the National Natural Science Foundation of China (413436, 61721), the research project of Surveying Mapping and Geoinformation of Jiangsu Province (JSCHKY218). Authors are grateful to the National Marine Satellite Center of China for providing the HY2 satellite data and to the CODE for providing GIM. References 1 Wu Yewen. Characteristics of the ionospheric total electron content based on the measurements of global navigation satellites. PhD dissertation,xidian University (in Chinese), Yuan Yunbin, OuJikun. Establish the GPS grid ionospheric model based on different areas for different stations.chinese Science Bulletin (in Chinese), 47(22) Zhang Hongping. Study on the regional ionospheric monitoring and delay of the Chinese region based on the ground-based GPS.PhD dissertation, Shanghai astronomical observatory, Chinese academy of sciences(in Chinese), Liu Jun, Chai Hongzhou, Liu Changjian, et al.analysis based on code GIM of ionospheric TEC anomaly before earthquake. Journal of Geodesy and Geodynamics(in Chinese),31(211) Wang Cheng, Wang Jiexian. Assessment of Global Ionospheric Maps and International Reference Ionosphere in China.Journal of Geodesy and Geodynamics (in Chinese),33(213) Wang Ping. Theoretical research and algorithm implementation of GPS single-frequency precise point positioning. PhD dissertation, PLA Information University (in Chinese), M.V. Codrescua, S.E. Palo, X. Zhang, et al. TEC climatology derived from TOPEX/POSEIDON measurements. Journal of Atmospheric and Solar-Terrestrial Physics,61(1999) Y.O. MigoyaOrue, S.M. Radicella, P. Coisson, et al. Comparing TOPEX TEC measurements with IRI predictions. Advances in Space Research, 42 (28) Francisco Azpilicueta,Claudio Brunini. Analysis of the bias between TOPEX and GPS VTEC determinations. J Geod, 83 (29)

10 26 INDIAN J MAR SCI VOL 4, NO.2 FEBRUARY Yu.V. Yasyukevich, E.L. Afraimovich, K.S. Palamartchouk,et al. Cross testing of ionosphere models IRI-21 and IRI-27, data from satellite altimeters (Topex/Poseidon and Jason-1) and global ionosphere maps. Advances in Space Research, 46 (21) Chen Peng, Chen Jiajun. Global ionospheric grid modeling using GNSS and space-based observations. The Fourth China Satellite Navigation Conference Proceedings, E.Yizengaw, M.B.Moldwin, D.Galvan, et al. Global plasmaspheric TEC and its relative contribution to GPS TEC. Journal of Atmospheric and Solar-Terrestrial Physics, 7 (28) S. Todorova, T. Hobiger, H. Schuh. Using the Global Navigation Satellite System and satellite altimetry for combined Global Ionosphere Maps. Advances in Space Research,42 (28) M. M. Alizadeh, H. Schuh, S. Todorova, et al. Global Ionosphere Maps of VTEC from GNSS, satellite altimetry, and formosat-3/cosmic data. J Geod,8 (211) Denise Dettmering, Michael Schmidt,Robert Heinkelmann, et al. Combination of different space-geodetic observations for regional ionosphere modeling. J Geod,8 (211) O.A. Maltseva, G.A. Zhbankov, T. Trinh Quang. Improvement of the real time total electron content based on the International Reference Ionosphere model. Advances in Space Research, 46 (21) Stefan S., Werner G., Feltens, J. IONEX:The Ionosphere Map Exchange Format Version1. Proceeding of the IGS Analysis Center Workshop,1998, CAI Chaojun, CAOJing, HUANG Jiang, et al.gps Observation Data and IRI-Model-based Contrastive Analysis on Ionospheric TEC Change of Guangzhou Region during low Solar Activity Years.South China Journal of Seismology (in Chinese),32(212)1-8.