Akwasi Afrifa Acheampong

149 Pages·2016·2.86 MB·English

by  Akwasi Afrifa Acheampong

Kwame Nkrumah University of Science and Technology
College of Engineering
Department of Geomatic Engineering
Retrieval of Integrated Water Vapour
from GNSS Signals for Numerical
Weather Predictions
by
Akwasi Afrifa Acheampong, MPhil, MGhIE
Thesis submitted in partial fulfilment for the award of the degree of
Doctor of Philosophy in GEOMATIC ENGINEERING
August 2015
Supervisors: Prof. Collins Fosu
Department of Geomatic Engineering, KNUST, Kumasi - GHANA
Dr. Leonard Kofitse Amekudzi
Department of Physics, KNUST, Kumasi - GHANA
Prof. Eigil Kaas
Niels Bohr Institute, Univ. of Copenhagen
Declaration of Authorship
I, Akwasi Afrifa Acheampong, declare that this thesis titled, ’Retrieval of Inte-
grated Water Vapour from GNSS Signals for Numerical Weather Predic-
tions’ and the work presented in it are my own. I confirm that:
(cid:4) This work was done wholly while in candidature for a research degree at this
University and no part of this thesis has previously been submitted for a degree
or any other qualification at this University or any other institution.
(cid:4) Where I have consulted the published work of others, due acknowledgements and
citations have been provided. With the exception of such quotations, this thesis
is entirely my own work.
Signed: .................................................................... Date: .......................................
Akwasi Afrifa Acheampong
(PhD Candidate)
Signed: .................................................................... Date: .......................................
Prof. Collins Fosu
(Supervisor)
Signed: .................................................................... Date: .......................................
Dr. Isaac Dadzie
(Internal Examiner)
ii
“Life is a stage, rise up early, prepare well and act your part with honesty and diligence
and leave the rest to Tweduanpon”
Kwasi Acheampong Atta-Badu
Abstract
Atmospheric Water vapour is an important greenhouse gas and contributes greatly in
maintaining the Earth’s energy balance. This critical meteorological parameter is not
sensed by any facility in Ghana contributing weather data to the Global Telecommu-
nication System of WMO. This thesis presents a highly precise tool for water vapour
sensing based on the concept of Global Navigation Satellite Systems (GNSS) mete-
orology and tests the computed results against global reanalysis data. Conventional
approaches used to sense the atmospheric water vapour or precipitable water (PW) or
Integrated Water Vapour such as radiosondes, hygrometers, microwave radiometers or
sun photometers are affected by meteorological conditions, expensive and have coverage
limitations. However, GNSS meteorological concept offers an easier, inexpensive and
all-weather technique to retrieve PW or IWV from Zenith Tropospheric Delays over
a reference station with very high temporal resolutions. This study employed precise
point positioning (PPP) techniques to quantify the extent of delays on the signal due
to the troposphere and stratosphere media where the atmospheric water vapour resides.
The KNUST GPS Base station was used to logdual-frequency signals for approximately
260days between the months of February 2013 to December 2014. Stringent process-
ing criteria were set using an elevation cut-off of 5o, precise orbital and clock products,
Antex files, nominal tropospheric correction and mapping functions. The delays which
were originally slantedare mapped untothe zenithdirectionandintegratedwithsurface
meteorological parameters to retrieve PW or IWV. This research work investigated the
applicationsofground-basedGNSStometeorologyandgivesallcorrectionmodelsimple-
mented in PPP and for Tropospheric delay estimation.The gLAB software was used for
ZTD computations. PW values obtained were compared with ERA-Interim, Japanese
Meteorological Agency Reanalysis (JRA) and National Centres for Environmental Pre-
diction (NCEP) global reanalysis data. Correlation analysis were run on computed PW
from logged GNSS datasets and downscaled reanalysis data. The obtained results show
stronger correlation between the retrieved PW values and those provided by the ERA-
interim. The computed amount of ZTDs varies perfectly with weather pattern in the
country. Again, a linear-model was derived that could predict PW based on ZTD with
standard errors of 6.01mm for JRA, 5.40mm for ERA-Interim and 6.34mm for NCEP
reanalysis data. Finally, the study results indicate that with a more densified network
of GNSS base stations the retrieved PW or IWV will greatly improve numerical weather
predictions and more specifically precipitation forecasting in Ghana.
Keywords: GNSS, PPP, gLAB software, Reanalysis, Integrated Water Vapour, Pre-
cipitable Water, ERA-Interim, NCEP, JRA
Acknowledgements
My greatest appreciation goes to the Lord Almighty for bringing me this far throughout
my course. Next, I wish to convey sincere gratitudes to Profs. Collins Fosu, Leonard
K. Amekudzi and Eigil Kaas for accepting me as their student. Your support, guidance
and valuable suggestions were of immense importance to this dissertation.
Special thanks go to all the staff members of the Geomatic Engineering Department,
KNUST, Niels Bohr Institute and Danida Fellowship Center, Copenhagen - Denmark.
This thesis benefited greatly from funds, seminars and equipment from the Building
Stronger Universities’ Environment and Climate Platform under Danish International
Development Agency (DANIDA). To the Platform Chairman, Coordinators (both local
and abroad) and officers at the International Programs Office of KNUST and DANIDA
Fellowship Centre, Copenhagen, I am most grateful.
Last but not the least, I would like to extend special thanks to my family and all my
friends. Your company, friendship, support and encouragement in times of need were
invaluable. May God richly bless you all.
v
Contents
Declaration of Authorship ii
Acknowledgements v
Contents vi
List of Figures ix
List of Tables xi
Abbreviations xii
Physical Constants xiii
Symbols xiv
1 Introduction 1
1.1 GNSS Meteorology and Applications . . . . . . . . . . . . . . . . . . . . . 3
1.2 Atmospheric Water Vapour . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Tropospheric Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Research Problem and Justification . . . . . . . . . . . . . . . . . . . . . . 7
1.5 Aim and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 GNSS, Positioning and Mathematical Models 10
2.1 GNSS Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.1.1 Applications of GNSS . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.2 GNSS Constellation Status . . . . . . . . . . . . . . . . . . . . . . 14
2.1.3 GNSS Signal and Observables . . . . . . . . . . . . . . . . . . . . . 14
2.2 Positioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.1 Reference Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.2 Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.3 PPP Correction Models . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.3.1 Antenna Phase Center Variations . . . . . . . . . . . . . 25
2.2.3.2 Phase Wind-up Effects . . . . . . . . . . . . . . . . . . . 26
2.2.3.3 Relativistic Effects . . . . . . . . . . . . . . . . . . . . . . 27
2.2.3.4 Ionospheric Delays . . . . . . . . . . . . . . . . . . . . . . 28
2.2.3.5 Tropospheric Delays . . . . . . . . . . . . . . . . . . . . . 29
vi
Contents vii
2.2.3.6 Mapping Functions . . . . . . . . . . . . . . . . . . . . . 31
2.3 Height Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3 Atmosphere, Water Vapour and Reanalysis Data 35
3.1 Composition of Earth’s atmosphere . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Water Vapour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2.1 Relationship between Humidity and Water Vapour . . . . . . . . . 38
3.2.2 Integrated Water Vapour . . . . . . . . . . . . . . . . . . . . . . . 40
3.3 Meteorological Observations and Techniques . . . . . . . . . . . . . . . . . 40
3.4 Weather Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4.1 Importance of Weather Data . . . . . . . . . . . . . . . . . . . . . 45
3.5 Ghana and Its Climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.6 Reanalysis Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4 Tropospheric Delays from PPP 51
4.1 Tropospheric Delay Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.2 Precipitable Water Computation . . . . . . . . . . . . . . . . . . . . . . . 54
4.3 Observation Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.4 Processing Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.5 Coordinates of Antenna Position . . . . . . . . . . . . . . . . . . . . . . . 58
4.6 Tests on Computed ZTD. . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
5 Results and Discussions 62
5.1 PW Comparisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1.1 Curve Fitting and Model for Prediction . . . . . . . . . . . . . . . 65
5.1.2 Seasonal Variations of PW . . . . . . . . . . . . . . . . . . . . . . 68
5.2 GNSS Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.2.1 GNSS Tomography Formulation . . . . . . . . . . . . . . . . . . . 73
5.3 Proposed GNSS Meteorological Set-up in Ghana . . . . . . . . . . . . . . 76
6 Conclusions 79
6.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
6.2 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.3 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
A Plots and Charts 108
A.1 Curve Fitting and Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
A.2 Linear Model and Reanalysis Data Comparison . . . . . . . . . . . . . . . 112
A.3 Seasonal Variations of PW . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
B Processed Coordinates for Antenna Position 118
B.1 Results from GLAB using PPP . . . . . . . . . . . . . . . . . . . . . . . . 118
B.2 Results from CSRS-PPP online PPP . . . . . . . . . . . . . . . . . . . . . 119
B.3 Results from GAPS online PPP . . . . . . . . . . . . . . . . . . . . . . . . 120
B.4 Results from APPS online PPP . . . . . . . . . . . . . . . . . . . . . . . . 121
B.5 Results from AUSPOS online PPP . . . . . . . . . . . . . . . . . . . . . . 122
C The gLAB GNSS Processing Software 124
C.1 Interfaces of the gLAB Software. . . . . . . . . . . . . . . . . . . . . . . . 124
C.2 RINEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Contents viii
C.3 APPS Final Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
C.4 CSRS-PPP Final Output . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
C.5 gLAB Final Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
D Climate Data Operators 131
D.1 Installing CDO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
List of Figures
1.1 Earth’s Atmospheric Layers . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 Locations of Ground-Based GPS Stations (shown in blue dots) in GUAN.
[ref : www.ecmwf.int] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1 Offsets in satellite’s centre of mass and antenna phase center, where, PCO and
PCV are phase centre offsets and variations (Karabati´c, 2011) . . . . . . . . . . 25
3.1 Atmospheric structure and subdivision (Pottiaux, 2010) . . . . . . . . . . 36
4.1 Ray bending effects on a radio signal . . . . . . . . . . . . . . . . . . . . . . . 52
4.2 KNUST Base Station and Sokkia(cid:114) GSR 2600 used for data logging . . . . . . . 56
4.3 Kumasi, study area for this study (Google, 2015) . . . . . . . . . . . . . . . . 56
4.4 ZTD plot against Day-of-Year for online PPP server values and gLAB
computations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.5 Box plot showing data distribution of computed ZTD from the 3 process-
ing approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.6 Correlation plot APPS and CSRS values against gLAB. . . . . . . . . . . 61
5.1 Plot of Computed PW from Weather-free Model and Retrieved from Reanalysis
Models against DoY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.2 Linear-fitplotsforCorrelationvaluesforcomputedPWfromWeather-freemodel
and Reanalysis models resulted in 0.541 for JRA, 0.598 for ERA-Interim and
0.458 for NCEP-R1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
5.3 Plot of Computed PW using Weather data and PW Retrieved from Reanalysis
Models against DoY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
5.4 Correlation Plots using Bootstrapping and 95% CI tests resulted in r
values of 0.7729 for JRA, 0.8345 for ERA-Interim and 0.6491 for NCEP-R1 67
5.5 Linear-fitting models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.6 Linear model trendline with 95% C.I. bounds and Reanalysis Data com-
parison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.7 PW values against Days of the Year grouped according to weather seasons 70
5.8 GNSS Tomographic concepts showing rays through vertical layers called
voxels (Bosy et al., 2010) . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.9 Conceptual view of signals through voxels . . . . . . . . . . . . . . . . . . 74
5.10 Number of Observed Satellites and visibility times above the horizon . . . 76
5.11 GNSS Meteorological Set-up in Ghana . . . . . . . . . . . . . . . . . . . . . . 77
A.1 Linear-fitting models: Linear fit . . . . . . . . . . . . . . . . . . . . . . . . 109
A.2 Linear-fitting models: Logarithmic fit . . . . . . . . . . . . . . . . . . . . 110
A.3 Linear-fitting models: Normalized Linear fit . . . . . . . . . . . . . . . . . 111
A.4 Linear model Trendline and JRA comparison . . . . . . . . . . . . . . . . 112
ix