Analysis of GPS Signals for the Determination of Plate Movements and Environmental Modeling

Transcription

1 ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 Analysis of GPS Signals for the Determination of Plate Movements and Environmental Modeling Ikram-e-Khuda, Umer Zia Abstract : Our focus of this paper is to discuss a method to determine the dynamics of two different continental plates, using GPS signals. The continental plates that are selected for the analysis are N.American and S.American plates. To determine the dynamics or plate movements, navigational signals of GPS are used. Access was made possible by Onsala Space Research Laboratories, Sweden [1]. Data from GPS measurements are observed for a time span of about eight to ten years. Second part of this paper deals with the use of GPS signals to help us measure the change in the troposphere. Thus by having some prior knowledge and applying sensible calculations, future climate and several atmospheric processes can be predicted. Index Terms Continental plates, GPS, troposphere, climate, plate movements. T I. INTRODUCTION HE surface of the earth is divided into seven major plates and several minor plates made up from the lithosphere. Seven major plates include; African Plate, Antarctic Plate, Eurasian Plate, Indo-Australian Plate, North American Plate, Pacific Plate, South American Plate. Similarly the seven secondary plates include; Arabian Plate, Caribbean Plate, Cocos Plate, Juan de Fuca Plate, Nazca Plate, Philippine Sea Plate, Scotia Plate. In addition to these primary and secondary plates, there are tertiary plates as well. Earlier it was assumed that earth s major features were fixed. However in the 19s, scientists found that relative movements between the continents do exist, however with a very slow acceleration. These movements are quite random but can be determined precisely on most areas of the earth's surface. There exist different types of plate movements; vertical, horizontal, divergent, convergent and lateral spinning. Most of the earth s seismic activities occur at the plate boundaries as they interact. This project was submitted in December 1. This work was part of the Masters Program course work in Satellite Positioning course at Chalmers University of Technology Ikram-e-Khuda and Umer Zia were the Masters student in Chalmers University of Technology (Chalmers), Gothenburg, Sweden ( ekhuda@student.chalmers.se) and (umer@student.chalmers.se) respectively. Ikram-e-Khdua is currently working as Assistant Professor in Iqra University (Karachi) in Faculty of Engineering Science and Technology. Umer Zia is presently working as Research Assistant in LUMS, Pakistan. Relative density of oceanic lithosphere and the relative weakness of the asthenosphere provides major contribution in plate movements. Dissipation of heat from the mantle is as well, acknowledged to be the original source of energy driving plate tectonics, through convection or large scale upwelling and doming. By using the GPS data our first task is to determine whether the plates are really moving and in which orientation, and to verify such movement between the S.American and North American plate. We then predict, by using GPS data to show how to predict atmospheric condition for the next day. For GPS, details about its elements and operations can be studied from a good amount of literature available [,] II. GPS ERRORS Ionosphere & troposphere disturbances: The troposphere is characterized having wet and dry parts, where the wet one is the most problematic part of the troposphere to model. If the reference stations are far away from the receiver or have a large height difference in comparison with the receiver, the errors from the troposphere increase significantly. To decrease troposphere errors, one should survey when the weather is similar, at the location of the receiver. The ionosphere is the upper part of the atmosphere and the impact on the ionosphere comes primarily from solar activity, contributing to the number of free electrons in the ionosphere which disturbs the measurements. These disturbances involve radio communication loss, initialization problems, loss of tracking of GNSS satellites, low precision of the measurements, etc., and they might occur more or less in different locations and at different times of the day and year. Ionosphere and troposphere layers slow down the satellite signal as it passes down. However the GPS system has a built in model that accounts for an average amount of these disturbances. DOP (Dilution of Precision): The geometry of the observable satellite constellation has a major effect on the position error. The visibility of GPS satellites with small separation angles leads to high DOP values and viseversa. The calculation shows that the best geometry comes from measurements with one satellite directly overhead and three on evenly separated elevation angles. As

2 ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 the GNSS system are increasing, this type of error can easily be eliminated by collaboration between satellites and by making receiver which can operate on multiple GNSS systems [5]. Geoid Model: The user should make sure the best geoid model is downloaded into the receiver to be able to determine accurate orthometric heights. III. OVERVIEW OF DATA PROCESSING GPS Receiver IGS GIPSY OASIS Parameter Estimation Ambiguity Resolution Figure 1.Dilution of Precision Signal reflection/multipath: Signal hits and is bounced back from objects like tall buildings, rocks, trees, walls etc. This causes the signal to be delayed and faded before it reaches the receiver. Multipath effects are different for each receiver and therefore do not cancel in differential GPS. These errors can be minimized by taking signal from those satellites that are at almost 1 above the horizon []. Ephemeris errors: Ephemeris errors are also known as orbital errors. These are errors in the satellite s reported position against its actual position. Clock errors: The built in clock of the GPS receiver is not as accurate as the atomic clocks of the satellites and the timing errors leads to corresponding errors in calculations. Visibility of Satellites: GPS system is designed in such a way that the receiver sees as many satellites as possible to make its accuracy better. But in real environment this is seldom the case. GPS receivers do not work indoors, underwater and underground. Satellite Shading: For the signals to work properly the satellites have to be placed at wide angles from each other. Poor geometry resulting from grouping can result in signal interference. Antenna Type: It is important for accurate positioning. Different antennas have varying sensitivities of measurements for the disturbances. One type of antenna might be more appropriate receiving low elevation signals from satellites, but worse at mitigating multipath errors Output Stacovs Tars Tkts Figure. Basic Block Diagram of GPS Data Processing IGS It is defined as International GNSS Service (IGS). Formerly it was defined only for GPS. It is a consortium of more than worldwide agencies who provides high quality geodetic data products to meet the objectives of a wide range of scientific, research and engineering applications and studies. High accuracy is achieved by deploying a dual frequency phase measuring receiver at an unknown location and recording all four observables (two phases and two codes) for periods of several hours to several days. The GPS data is provided in RINEX format (Receiver Independent Exchange format). Following data products are generated: Satellite Ephemerides Earth Rotation Parameters: Polar Motion (PM) Polar Motion Rates (PM rate) Length-of-day (LOD) Coordinates & velocities of the IGS tracking stations GPS satellite and tracking station clock information Earth rotation parameters Atmospheric Parameters Ionospheric information Tropospheric information o Final tropospheric zenith path delay o Ultra-Rapid tropospheric zenith path delay o Final ionospheric TEC grid o Rapid ionospheric TEC grid

3 ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 GIPSY-OASIS It is an automated, fast, ultra-precise high precision GPS data processing software package with strict data quality control, provided by National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (JPL). This data is used here for processing. The whole set of parameters being determined with GIPSY consists of the coordinates of the receiver s antenna, phase ambiguities, receiver clock errors, wet zenith delays and troposphere gradients. The following are the steps involved in GIPSY processing. Input: RINEX format satellite navigation data are the two input file requirement for GIPSY as is retrieved from IGS servers Modeling : This involves modeling on the basis of accurate and precise Geodetic models. Table 1. GPS stations and their locations Estimation: this involves estimation of parameters, and mapping of data points by proper smoothing and filtering of them. Kalman filtering with LMS filter are used for this purpose Ambiguity Resolution: This gives accuracy with higher level of precision by removing errors because of station coordinates, orbital parameters and differentiating techniques. Output: The output file is converted to a human readable format from Binary Format, and it goes to directories like stacovs, tars and tkts. The coordinate information is saved in the STACOVS directory; the other directory is TARS and contains tropospheric data. The third directory TKTS contain clock information. The STACOVS file contains not only the XYZ coordinates of the stations, but also the possible amount of error. IV. DESCRIPTION OF TASKS GPS provides us coordinates of a particular point over a specific period of time. If the coordinates are changing with the time, then one can say that dynamics exist or there is movement. In task 1 we took stations (locations) on both of the plates which are under observations to collect and manipulate that data to get to our conclusions. The stations were selected with the consideration that they were far from earthquake sensitive regions and data from them were taken of the duration of 1 years, summer, of 7 to 1 consecutive days from each year. Summer was selected because it avoids snow and wet climate conditions which will cause error in our data from GPS. Taking data for consecutive days would help decrease uncertainty due to sudden or abrupt atmospheric changes Figure 3. Map showing GPS stations It can be seen from the above table and map that a minimum of three stations have been selected at each of the plate, covering north, south and central part of the plate. Moreover the stations at the two extremes, that is the north and the south of a plate may also confirm the difference in the movement between them. Task or second part of the paper is related to a very important application of GPS to detect the change in the troposphere and predict the future atmosphere. We shall accomplish this task by taking data for four days for two different stations, and then predict the troposphere for the 5th day. The two stations which were chosen, were KOUR and KIRU. We selected these stations as they had different climatic conditions. KIRU is in the north (Sweden) which is expected to be cold and less humid, the other station is in the mild latitudes (S.America) and expected to be warm and humid, near equator. In this paper only simulations from KIRU are presented to save space.

4 ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 V. OBSERVATIONS, RESULTS AND DISCUSSIONS A. Task 1. Plate Movements GPS data is in the form of latitude (Lat), longitude (Long) and Radius (Rad) information. Station movements vs. year durations for N.American earth stations are shown in fig, fig 5 and fig. S.American stations observations are similarly shown in fig 7,fig 8 and fig 9. Fig.5 WES Station The graphs of three stations of the North American Plate, shows their movement in the past ten years. For all the stations, a decrease in the longitude is observed from 1cm to -1cm, 5cm to -1 cm, 1 to cm, in the stations, BAKE,WES and MDO1 respectively. This clearly shows the westwards movement of the North American plate For the three stations we see a decrease in laitude in Bake and MDO1, and comparing this with the station WES one can judge that N.America plate tends to move in the counterclockwise direction. To conclude, one can say that the movement between different parts of the plate is not the same and we see a westwards drift with counterclockwise rotation. We can also measure the drift velocity and angular rotation by applying simple mathematics. 1 Fig. BAKE Station Figure 5. Simulation of GPS data from WES Station 1 8 Fig. MDO1 Station Figure. Simulation of GPS data from Bake Station Figure. Simulation of GPS data from MDO1 Station

5 ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 8 Fig.7 KOUR Station 8 Fig.9 LPGS Station Figure7. Simulation of GPS from KOUR station Figure9. Simulation of GPS from LGPS station Fig.8 CHPI Station In figures 7,8 and 9, we can see the graphs of latitude, longitude and radius for S.American plate. Lets consider the stations in the two extremes of the continent, which are, KOUR and LPGS. Their increase in latitude and decrease in longitude shows the drift of the S.American plate in north western direction vetor. Now considering the stations in the central portion, CHPI; the graph shows drift more in latitude and almost not very much decelerating drift in the longitude. But this does show similarity in the directions with the previous two stations. So net drift is in the north western direction for S.American plate. This analysis was made with the simulations we performed on Matlab for the GPS data that was provided to us, using the latitude, longitude and radius information. In the plots/ figures they are shown and mentioned with the blue, red and green plots respectively. Figure8. Simulation of GPS from CHPI station

6 ZTD (m) ZTD (m) ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 Errors and Uncertainty Fig 1. WES Station with Error Uncertainity Plot Fig 1. Estimation/ Prediction of ZTD for KIRU Figure 1. Simulation of GPS data from WES Uncertainty Plot Station with Error Lets now consider the errors and uncertainty. We show one station WES (in N. America) and have plotted the data with the error uncertainty in the form of error bars. As seen in the plot, there is some error percentage for the radius or altitude variation, longitude and latitude, which after calculation comes out to be.9%, 1.% and.9% respectively. As in our case, plate motion, altitude measurement was not crucial and hence can be neglected and from a general trend in such measurements 5- percent error for Lat and Long appears to be acceptable as compared to the error sources defined above Time (s) x Figure 11(top) Data for Four days. (bottom). Estimations Results Fig 11. Estimation/ Prediction of ZTD for KIRU B. TASK Environmental Modeling Fig 11 shows the Tropospheric delay parameter for four days of data(from KIRU). To estimate(atmosphere or water vapor content) for the 5 th day, we extended time axis to plot the slope and then we estimated the value of Zenith Total Delay (ZTD) at the mean position of the next day, which comes out to be equal to.195. From this we obtain actual slope of the next day, which is shown with green line in fig 11 (b). Now the ZTD again at the mean position of the next day comes out to be.1. The difference between predicted and actual values is equal to;.1.195= Time (s) x 1 8 VI. CONCLUSIONS In this paper we have studied and analyzed the simulations of actual data about the kinematics of the continents among each other. With the above calculations and plots, it is certain that the continents are really moving with the passage of time. Results revealed the movement of the North American plate in westward direction and a movement variation was observed between

7 ASIAN JOURNAL OF ENGINEERING, SCIENCES & TECHNOLOGY, VOL., ISSUE 1. MARCH 1 the north and south parts of a same plate, thus giving it a rotation in circular direction. Regarding the S.American, their movement appears to be in the direction of increasing latitude and decreasing longitude. More stations can definitely be added in our work to find movement of different part of late and to find more accurate results. Whereas regarding the second task results in predicting the troposphere; the methods that were used were quite reasonable. Hence we present a simple and easy way to analyze GPS data for extracting useful information about plate tectonics and environmental modeling with predictions of future weather. ACKNOWLEDGMENTS The authors are thankful to the concerned faculty in Earth and Space Science Department of Chalmers University, Sweden for helping them accesses the related data and using it for the analysis. Sincere regards to PhD student Tong Ning, for providing technical assistance in the completion of this project work. The authors are also thankful to Adjunct Professor Jan Johansson for his supervision of the project work. REFERENCES [1] Official website of Onsala Space Observatory in association with Chalmers University of Technology is available at: [] Elliott Kaplan, Christopher Hegarty, Understanding GPS: Principles and Applications, Artech House Publishers; edition (November 3, 5) [3] Ahmed El-Rabbany, Introduction to GPS: The Global Positioning System, Artech House Publishers; edition (August 31, ) [] Jan Van Sickle, P.L.S. GPS for Land Surveyors, CRC Press; 3 edition (May 5, 8) [5] Hans-Jürgen EULER, Joachim WIRTH, Advanced Concept in Multiple GNSS Network RTK Processing 1 st International Conference on Machine Control & Guidance 8 [] G W ROBERTS, X MENG, E COSSER and A H DODSON, The Use of Single Frequency GPS to Measure the Deformations and Deflections of Structures,