Gan Magnetic Observatory. Maldive Islands South Indian Ocean Annual Report

Transcription

1 Gan Magnetic Observatory Maldive Islands South Indian Ocean 2016 Annual Report 1

2 Institut für Geophysik, Zürich Gan Magnetic Observatory Maldive Islands South Indian Ocean 2016 Annual Report Report Editor John C. Riddick Bibliographic reference Gan Magnetic Observatory, 2016 Annual Report. Institut für Geophysik, Zürich 2016 i

3 List of Contents 1 Introduction Description of the observatory Gan Observatory (IAGA code: GAN) Instrumentation Block diagram of system Data Processing Checks on Daily and Annual Mean Results Check on Definitive Minute Means and Absolute Spot Values Annual Means Gan Annual Means Gan Annual Rates of Change Operational Diary Statistics on Gan Observatory Data Recorded Baselines Total Field (F) Site difference Plots of daily magnetograms Gan Magnetograms January February March April May June July August September October November December Participants in Gan Observatory operation Acknowledgements Publications ii

4 13 References Papers Produced using Gan Data iii

5 1 Introduction This bulletin is a report of the measurements and operations carried out between 1 January and 31 December 2016 at the geomagnetic observatory on the island of Gan in the south Indian Ocean. Gan Observatory was installed by the Institut für Geophysik, ETH Zürich, Switzerland in April Day to day operation of the observatory is the responsibility of the local Meteorological Office staff who also carry out a regular program of absolute observations. The recorded data are automatically transmitted to a database in Zurich and also to the INTERMAGNET Geomagnetic Information Node (GIN) operated by the British Geological Survey (BGS) in Edinburgh, Scotland. All data processing and quality control procedures are carried out by suitably qualified personnel based at various locations in both Gan and in Europe. During 2013 Gan Observatory became a full member of INTERMAGNET. An on-line display of the Gan data can be found at or details of the observatory including photographs and publications at 2 Description of the observatory Gan is the most southerly island of the Maldive Islands and lies in the Indian Ocean almost 1 degree south of the equator. It is a coral atoll and part of the Addu Atol group of islands. Gan is the second largest island of the atoll, after Hithadoo and measures square kilometres (0.87 sq mi) in area. During WWII, Gan Island was developed as an airbase for the Royal Air Force who used it as a refuelling point for flights between the UK and the Far East. The observatory in located on the west side of the main runway and comprises 3 weatherproof shelters for the instruments and a hut used for absolute observations. From December 2014 through to March 2015 major renovation works were carried out on all the instrument huts.. Magnetometers and Control Electronics in waterproof boxes after refurbishment, Left to right, Fluxgate magnetometer, Proton magnetometer, Control Electronics, far right is the Absolute Observing Hut. (This picture courtesy Mohmed Nazmy, cover picture Ahmed Muslim) 1

6 2.1 Gan Observatory (IAGA code: GAN) Location: Latitude : Gan International Airport, Addu Atoll, Maldives 00 41'40.55" S Longitude : 73 09'13.47" E, Altitude: 2 m Local contact: Muslim Ahmed Regional Meteorological Office, Maldives Meteorological Service Gan Addu City Maldives Telephone: E_mail: ahmedmuslim@hotmail.com Zurich, Switzerland contact: Prof. Andrew Jackson Institut fur Geophysik, Sonneggstrasse 5 CH-8092, Zurich Switzerland Telephone: E_Mail: ajackson@ethz.ch 2

7 3 Instrumentation The instruments include a DMI Suspended FGE, 3 component fluxgate magnetometer measuring variations in the horizontal (H) and vertical (Z) intensities and the changes in the declination (D) along with temperatures in the sensor head and fluxgate electronics. Total Field (F) is measured by a GEM Systems GSM 90F1 proton magnetometer which also supports a GPS receiver used to provide accurate timing control for the data sampling and logging. All magnetic components are sampled once a second with magnetometer sampling, data storage and systems housekeeping controlled by a low powered UNIX based PC which operates from a battery backed 12 Volt DC supply deriving its power from a bank of solar cells. As well as storing the data locally a transmission link provides communication between the observatory and the local Meteorological Office in Gan Airport. At this office Internet facilities are available to transmit all recorded data to Zurich where an on-line display and database has been set up ( The observatory is completely self-contained with all power provided by the battery backed solar panels which will support operations for over a week in the event of no solar charge. Gan Observatory Equipment Solar panels used to charge the two 225 AH sealed lead acid batteries. 3

8 3.1 Block diagram of system DMI triaxial suspended fluxgate magnetometer with the protective fibreglass waterproof removed. Power box showing the solar regulator and two, 6 Volt 225 AH batteries. 4

9 4 Data Processing Throughout 2016 systems on Gan automatically updated the archive computer at ETH, Zurich and the BGS Geomagnetic Information Node (GIN) in Edinburgh several times a day. These data were then retrieved and checked, magnetograms plotted and baselines calculated using the GDASView software, (Turbitt, In Press). The daily processing also included a visual check for intermittent, man-made, spikes which were then edited out using the GDASView software. Shown left is a typical magnetogram of the daily variations measured at Gan. The fluxgate data, vertical (Z, blue), horizontal (H, red) and declination (D, green) along with total field (F, pink) calculated from the fluxgate H and Z variations are displayed in the upper 4 traces. The total field (F, black) measured by the proton magnetometer and the difference between F calculated from the fluxgate and F measured by the proton (bottom plot, yellow) are displayed in the lower two traces. These plots give an excellent indication that both the fluxgate and the proton are operating correctly. Following the successful application by Gan Observatory for membership to INTERMAGNET and also the requirement to provide more timely data, following the launch of the SWARM satellites, in 2014 the data processing procedures were revised. These included an hourly transmission of the Gan data to the Edinburgh GIN providing an on-line provisional data set and at monthly intervals, once baselines have been established, the generation and transmission to the GIN of quasi-definitive observatory data in IAGA 2002 format. Using GDASView has greatly simplified the processing of Gan data and the support of BGS Edinburgh is acknowledged in providing regular upgrades to this product. The definitive data and annual mean data for 2016 have all been generated using GDASView V5.04 version software. 5

10 Currently a database of all Gan data is maintained on the ETH server in Zurich. Data Structure on ETH, Zurich Database The files in the directories, shown above, contain 3 component fluxgate (H, D, Z) and proton full field data in a variety of formats and sampling rates from the 1 second recorded data to the baseline corrected, quasi-definitive and definitive data of minute means in IAGA 2002 format. Gan system housekeeping data, including, temperature, humidity, solar panel, battery and load voltages are also stored along with all absolute observations, baseline data and the Monthly and Annual Bulletins, giving details of all site operations and any changes. The Edinburgh GIN database contains day files of Gan minute mean data and makes it available in a variety of formats including IMF V1.22 and IAGA 2002 formats. Note: All access to Gan data in both the ETH and BGS databases is password protected. 6

11 D Component Hourly Mean (mins) H Component Hourly Mean (nt) 4.1 Checks on Daily and Annual Mean Results As a check against the GDASView software and to confirm that the baselines were correctly applied, a program to calculate the daily mean from baseline corrected IAGA 2002 format data was generated. Using this software, the daily mean value of the definitive data of all components for every day in 2016 was calculated and plotted along with the Annual Mean calculated by GDASView. Below are the results obtained from this comparison. The plots show that the daily means (blue diamonds) and the Annual Means (yellow triangle) calculated by GDASView are in good agreement Gan H Component Jan Mar Jun Sep Dec Gan Declination Jan Mar Jun Sep Dec

12 F Component Hourly Means (nt) Z Compoment Hourly Mean (nt) Gan Vertical Component Jan Mar Jun Sep Dec Gan Total Field Jan Mar Jun Sep Dec-16 8

13 Absolute H Component (nt) H Abs - H Minute Mean (nt) Absolute Declination (Mins) D Abs - D Minute Mean 4.2 Check on Definitive Minute Means and Absolute Spot Values As a further check, to ensure that the baselines had been correctly applied, comparisons between the baseline corrected (definitive) minute mean spot values and the absolute values of the magnetic field extracted from the 212 absolute observations made at Gan in 2016, were carried out. The following plots show the results of these comparisons. At the time of each absolute observation of H, D, Z and F the minute values were extracted from the IAGA 2002 format definitive minute mean files and the two values subtracted and plotted as shown below. Declination (D) Jan Mar Jun Sep Dec D Abs D Min.Mean Abs.-Min. Mean Horizontal (H) Component H Abs. H Min. Mean Abs-Min. Mean Jan Apr Jul Oct

14 Absolute F (nt) F Abs - F Minute Mean (nt) Absolute Z Component (nt) Z Abs - Z Minute Mean (nt) Vertical (Z) Component Jan Apr Jul Oct Z Abs. Z Min. Mean Abs. - Min. Mean Total Field (F) 2016 F Abs F Min. Mean Abs - Min.Mean Jan Apr Jul Oct The black dots on each plot show the difference between the absolute spot value and the minute mean value. The standard deviation of the differences for each component is shown below. D mins H nt Z nt F nt The maximum and minimum differences between the absolute observations and the minute means are: Declination mins H nt Z nt F nt Max Min Range These results confirm that there is good agreement between the definitive minute means and the results obtained from the 212 absolute observations made at Gan in

15 5 Annual Means Since the observatory was installed in April 2011 Annual Mean values of all components have been calculated from the baseline corrected minute mean values. In 2011, as the observatory had not been operating for a full year, the Annual Mean was estimated from the baseline corrected daily mean using a linear interpolation at From 2012 onwards GDASView software has been used to calculate Annual Mean values, the results are shown in the table below. Note the results in the tables shown below have been updated from the results given in previous editions of the Gan Year Book. This is as a results of corrections generated by changes in data processing methods and upgrades to the GDASView software. GAN COLATITUDE LONGITUDE ELEVATION 2m 5.1 Gan Annual Means YEAR D I H X Y Z F Deg min Deg. min nt nt nt nt nt Gan Annual Rates of Change D mins I mins H nt Z nt F nt

16 Annual Mean (Degrees) Rate of Chanhe (Minutes/Year) Annual Mean (Degrees) Rate of Change (Minutes/Year) Gan Declination Annual Change Annual Mean Rate of Change Gan Inclination Annual Change Annual Mean Rate of Change

17 Annual Mean (nt) Rate of Change (nt/year) Annual Mean (nt) Rate of Change (nt/year) Annual Mean (nt) Rate of Change (nt/year) Gan Horizontal Component Annual Change Annual Mean Rate of Change Gan Vertical Component Annual Change Annual Mean Rate of Change Gan Total Field Annual Change Annual Mean Rate of Change

18 6 Operational Diary January 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. February 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. March 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. On 31 March tests were started to check the batteries used to power the Gan observatory. This test confirmed that the current batteries, which were installed in 2011, have sufficient capacity to power the observatory for at least 24 hours in the event of a solar cell or regulator failure and the batteries themselves are not needing replaced. April 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. The tests on the system battery capacity concluded that the current battery pack was in good condition and not needing replaced. A report describing these tests can be found on the ETH Server /home/keeling/magdata/jcr Archive/Battery Tests. During April BGS upgraded the GDASView software, the current version used for all Gan data analysis is V4.8 May 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. On 23 May a series of total field (F) site difference measurements were made to check that the total field site difference between the continuous proton and the D/I observing position had not changed. These measurements indicated a 0.5 nt reduction since the previous measurements made in January The current F site difference correction in use since 1 January 2016 is 28.5 nt this value will continue to be used until the next set of measurements in July when if the downward trend continues the correction will be revised. 14

19 F D/I Position - F Observatory Proton (nt) F D/I Position - F Observatory Proton (nt) Gan F Site Difference Jan May Aug Dec Apr Aug-16 Gan total field (F) site differences observations June 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. July 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. On 16 th July a series of total Field (F) sute difference measurements were made. The results indicate a small reduction in this correction since the previous measurements made in May. Gan F Site Difference Jan-1501-May-1529-Aug-1527-Dec-1525-Apr-1623-Aug-16 Total Field site difference measurements for January 2015-July

20 2016 Measurements F D/I F Proton Difference 2016 Average 31-Jan May Jul The mesurements indicate a reduction in this correction of 1.56 nt since January Currently a F Site difference of 29.4 nt is used. This value is under review and will probably be changed in the final analysis of the 2016 data. In July 2016 GDASView V5.01 was released, this version provides improved data editing and data plotting. It also incorporates corrections to minor problems reported in earlier versions. August 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. On 8 th August, 12 minutes and on 9 th August 6 minutes of data were lost due to grass cutting. September 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. During September Ahmed Muslim attended the XVIIth Workshop on Geomagnetic Observatory Instruments and Data Acquisition and Processing at Dourbes, Belguim. Throughout September all Gan operations and absolute observing was the responsibility of Ibrahim Ziyad. October 2016 Throughout the month no problems were experienced with the instruments. On and on 5 th 6 th 24 th 25 th several minutes of data were lost due to grass cutting. In October GDASView was upgraded to V5.04, this version corrected ambiguities in the generation of INTERMAGNET baseline files. November 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. December 2016 Throughout the month no problems were experienced with the instruments. Data losses due to interference spikes were minimal. 16

21 From 1 December a new airline schedule was started with an Airbus A320 landing and taking off 4 times a week. Interference or any artificial disturbances from these operations are not visible on the minute mean plots. 17

22 6.1 Statistics on Gan Observatory Data Recorded In 2016 all the equipment on Gan operated very reliably with minimal data loss. The main cause of data loss was due to the regular grass cutting program carried out by the airport staff. In normal operation 1440 minute samples of all components are generated every day. Shown below is a plot of the data recorded throughout the year Gan 2016 Daily % of data received by GIN Jan Mar Jun Sep Dec-16 In 2016 Gan Observtory data is available for 99.9% of the year. 18

23 Baselines Throughout the year a regular program of absolute measurements of declination (D) and inclination (I) have been carried out by the Meteorological Office staff on Gan and the observers are to be congratulated on their excellent work, during 2016, 212 absolute observations were made. Depending on the weather and available staff time observations are made twice a week. The results from these observations have indicated that after replacing the fluxgate magnetometer in January 2013 the fluxgate has produced very stable results. At Gan there are three trained observers, Ahmed Muslim, Ibrahim Ziyad and Mohmed Nazmy. Using the measured absolute value of D and I and a site corrected value of total field (F), measured by the recording proton, the absolute value of the horizontal (H) and vertical (Z) intensities at the D/I pillar are calculated. From these results the fluxgate H, D and Z baselines are generated. Using the GDASView software, baselines along with observation and instrument (D/I theodolite) quality control measures are derived from the absolute observations. Below the baselines for 2016 are shown. Gan 2016 Baseline plot 19

24 8 Total Field (F) Site difference In standard magnetic observatory practise, total field (F) is measured continuously using a proton magnetometer and the values obtained adjusted to derive the F at the observatory D/I pillar where it is used, along with the measured inclination (I) to calculate the absolute horizontal (H) and vertical (Z) intensities. As the F value is not measured at the D/I pillar a correction has to be applied to the measured value of F to correct it to the F value at the D/I pillar, this is normally referred to as the F Site Difference. In order that a regular check on any changes in the F Site Difference could be made a GEM, GSM19, proton magnetometer was supplied to Gan in Using this instrument regular comparisons are made between the continuous proton and total field measured at the D/I pillar. Gan F Site Difference Jan Apr Jul Oct-16 Plot showing the change in the F Site Difference at Gan in The results of these comparisons indicated that the F Site Difference at Gan, since the refurbishment work was completed in March 2015, has changed by less than 1nT, the current value used by GDASView in the absolute observation reductions is 28.5 nt. The correction: F Observatory proton nT = F D/I pillar Full details and the procedures used in making these measurements can be found at 20

25 9 Plots of daily magnetograms The following plots show the baseline corrected magnetograms for the horizontal (H), declination (D), vertical (Z) fluxgate variations along with the total field (F) from the proton magnetometer. The data are plotted for each month in 6 day sections, dates 1-6 th, 7-12 th, th, th followed by the remaining days of each month as 6 or 7 day plots. Apart from the month of February where the data are plotted as four groups of 6 days and one group of 5 days, 25 th 29 th. 9.1 Gan Magnetograms January

26 22

27 23

28 24

29 Gan Magnetograms February

30 26

31 27

32 28

33 Gan Magnetograms March

34 30

35 31

36 32

37 Gan Magnetograms April

38 34

39 35

40 36

41 Gan Magnetograms May

42 38

43 39

44 40

45 Gan Magnetograms June

46 42

47 43

48 44

49 Gan Magnetograms July

50 46

51 47

52 48

53 Gan Magnetograms August

54 50

55 51

56 52

57 Gan Magnetograms September

58 54

59 55

60 56

61 Gan Magnetograms October

62 58

63 59

64 60

65 Gan Magnetograms November

66 62

67 63

68 64

69 Gan Magnetograms December

70 66

71 67

72 68

73 10 Participants in Gan Observatory operation Gan Operations and Observers Mr Ahmed Muslim, Climatologist, Maldives Meteorological Service. Mr Ibrahim Ziyad, Assistant Meteorological Technician, Maldives Meteorological Service. Mr Mohmed Nazmy, Meteorological Instrument Technician, Maldives Meteorological Service. European Operations. Professor A Jackson, Project Leader, Institute of Geophysics, ETH Zürich, Switzerland. Dr Jakub Velimsky, Dept. of Geophysics, Charles University in Prague, Czech Republic. Dr Alexei Kuvshinov, Institute of Geophysics, ETH Zürich, Switzerland. Mr John C. Riddick, Scotland. 11 Acknowledgements We would like to acknowledge and thank the following people for their help in setting up and the continued running of Gan Observatory. Dr Lars W. Pedersen, National Space Institute, Technical University of Denmark for his continued support and technical expertise. Mr. Abdulla Wahid, Executive Director of Maldives Meteorological Service. Mr. Ali Wafir, Head of Gan Meteorological Office - for logistical and management support; Dr. Juergen Matzka Helmholtz Centre Potsdam, GFZ German Centre for Geosciences - for assistance at initial stages of the project; Dr. Chris Finlay, National Space Institute, Technical University of Denmark - for assistance at initial stages of the project; Mr. K.C.S. Rao and Dr. Kusumita Arora, National Geophysical Research Institute, Hyderabad - for assistance in installation of instruments, and training observers. Mr C. Turbitt, British Geological Survey for his assistance in the development and use of GDASView software. 69

74 12 Publications Gan Observatory Monthly Bulletins, Gan Magnetic Observatory, 2012 Annual Report, Gan Magnetic Observatory, 2013 Annual Report, Gan Magnetic Observatory, 2014 Annual Report, Gan Magnetic Observatory, 2015 Annual Report, Gan Magnetic Observatory, 2016 Annual Report, J. Velímský, A. Muslim, A. Jackson, A. Kuvshinov, F. Samrock, K. Arora, K.C.S. Rao, L. Pedersen, C. Finlay, J. Riddick, Geomagnetic observatory Gan, XV th IAGA Workshop on Geomagnetic Observatory Instruments, Data Acquisition and Processing, Cadiz, J. Velímský, A. Muslim, A. Jackson, A. Kuvshinov, F. Samrock, K. Arora, K.C.S. Rao, L. Pedersen, C. Finlay, J. Riddick, Geomagnetic Observatory Gan, IAGA XII th Scientific Assembly, Merida, J. Velímský, A. Muslim, A. Jackson, A. Kuvshinov, F. Samrock, K. Arora, K.C.S. Rao, L. Pedersen, C. Finlay, J. Riddick, Geomagnetic observatory Gan, XVI th IAGA Workshop on Geomagnetic Observatory Instruments, Data Acquisition and Processing, Hyderabad, A. Muslim, J.Velımský, I. Ziyad, A.Jackson, A.Kuvshinov, F. Samrock, J. Riddick, K. Arora, K.C.S. Rao, L. Pedersen, C. Finlay, Recent upgrades of the Gan geomagnetic observatory IAGA XXVI th Scientific Assembly, Merida, Velimský, J., Muslim, A., Kuvshinov, A. Detecting tidal signals at Gan magnetic observatory. Observations versus predictions. XVII th IAGA Workshop on Geomagnetic Observatory Instruments, Data Acquisition and Processing, Dourbes References Turbitt, C.W., GDASView Users' Guide, Seismology & Geomagnetism Programme Internal Report (In press) 70

75 14 Papers Produced using Gan Data Thebault, E., Finlay, C.C., Beggan, C.D., Alken, P. Aubert, J., Barrois, O., Bertrand, F., Bondar, T., Boness, A., Brocco, L., Canet, E. Chambodut, A., Chulliat, A., Coisson, P., Civet, F., Du, A., Fournier, A., Fratter, I., Gillet, N., Hamilton, B., Hamoudi, M., Hulot, G., Jager, T., Korte, M., Kuang, W., Lalanne, X., Langlais, B., Leger, J.M., Lesur, V., Lowes, F.J., Macmillan, S., Mandea, M., Manoj, C., Maus, S., Olsen, N., Petrov, V., Ridley, V., Rother, M., Sabaka, T.J., Saturnino, D., Schachtschneider, R., Sirol, O., Tangborn, A., Thomson, A., Toeffner-Clausen, L., Vigneron, P., Wardinski, I., and Zvereva, T., (2016) International Geomagnetic Reference Field: the 12th generation, Earth, Planets and Space, Vol 67:79, doi: /s Finlay, C.C., Olsen, N. and Toeffner-Clausen, L. (2016) DTU candidate field models for IGRF-12 and the CHAOS-5 geomagnetic field model, Earth, Planets and Space, Vol 67:114 doi: /s Gillet, N., Barrois, O., and Finlay, C.C. (2016) Stochastic forecasting of the geomagnetic field from the COV-OBS.x1 geomagnetic field model, and candidate models for IGRF-12,Earth, Planets and Space, Vol 67:71 doi: /s z Hamilton, B., Ridley, V.A., Beggan, C.D., Macmillan, S., (2016) The BGS magnetic field candidate models for the 12th generation IGRF. Earth, Planets and Space, Vol 67:69, doi: /s x Samrock F. and Kuvshinov A., Tippers at island observatories: Can we use them to probe electrical conductivity of the Earth's crust and upper mantle? Geophys. Res. Lett., doi: /grl