UNIVERSITY OF NAIROBI
USE OF SINGLE FREQUENCY DIFFERENTIAL GPS
RECEIVERS IN GENERAL BOUNDARY SURVEYS
STUDY AREA: MATHUNTHINI REGISTRATION SECTION
BY
KIRERA LINDA KATHAMBI
F19/2558/2008
A project report submitted to the department of geospatial and space technology in partial
fulfillment of the requirements for the award of the degree of:
Bachelor of Science in Geospatial Engineering
APRIL, 2013
1
ABSTRACT
The general boundaries are mainly found in the Kenyan rural areas where most agricultural
activities are located. In addition to having land parcel boundaries which are not georeferenced
most of the land in the rural areas has not yet been surveyed. As a result problems related to land
disputes, cadastral inconsistencies as well as little or no loans for investments from financial
institutions generally affects the rural life.
The research work aims at determining the most effective method of georeferencing the PIDs as
well as on what can be done to improve their accuracy and that of the general boundary surveys
generally. Land parcels are geo-referenced using X, Y (2D) Coordinate System.
Research methodology involved; the identification of the study area, obtaining of the PID,
selection of the land parcels to be surveyed, measuring of the parcel corner coordinates using the
handheld and the Differential GPS (DGPS) receivers, scanning and georeferencing of the PID
using both the DGPS and handheld coordinates and digitization of the land parcels to obtain the
areas and perimeters. A google earth image of the same study areas was obtained to help in the
analysis.
The two sets of coordinates were different with a percentage mean difference of 15.4% in the
northings and -10.93% in the eastings respectively. The position difference, that is, the difference
between the positions fixed using the handheld GPS receiver from that fixed using the DGPS
was as high as 8m and only two positions were less than 3m out of thirteen. The percentage
difference in the original PID areas and perimeters from the DGPS ones ranged from 23% to
49% and 0.19% to 88.89% respectively.
It was thus concluded that the handheld coordinates did not meet the accuracy requirements for
georeferencing the land parcel boundaries. The analysis carried out using the google earth image
showed that the PID did not represent what was on the ground.
Based on the findings the research recommends the use of DGPS receivers for georeferencing of
the land parcel boundaries. It also recommends its use in carrying out of the general boundary
surveys for better accuracies in comparison to the handheld GPS receivers or the total stations.
i
DEDICATION
The research is dedicated to my parents Susan and Davis, my sisters and brothers as well as my
dear friends and Rick who through the days and nights they prayed for me and encouraged me
when the various challenges seemed to wear me down. Thanks so much for your support God
bless you so much.
ii
ACKNOWLEDGEMENT
I thank the almighty God for granting me good health throughout the research period. I wish to
extend my special thanks to my supervisors Dr. Musyoka and Mr. Matara, for the guidance and
support they offered to see this work being accomplished. Not forgetting Mr. Benson Makau for
the GPS services during the field work. I also thank my fellow students and any other person not
mentioned but in one way or the other contributed to the success of this work. God bless you all.
iii
TABLE OF CONTENTS
ABSTRACT................................................................................................................................................... I
DEDICATION .............................................................................................................................................. II
ACKNOWLEDGEMENT ........................................................................................................................... III
TABLE OF CONTENTS .............................................................................................................................IV
LIST OF TABLES .......................................................................................................................................VI
LIST OF FIGURES .................................................................................................................................... VII
ABBREVIATIONS .................................................................................................................................. VIII
REPORT ORGANIZATION .......................................................................................................................IX
CHAPTER ONE ........................................................................................................................... 1
1.0. INTRODUCTION............................................................................................................................. 1
1.1. BACKGROUND ................................................................................................................................ 1
1.2. PROBLEM STATEMENT AND JUSTIFICATION .......................................................................... 5
1.2.1.
1.2.2.
PROBLEM STATEMENT ..............................................................................................................................5
JUSTIFICATION .............................................................................................................................................6
1.3. OBJECTIVES ..................................................................................................................................... 6
1.3.1.
1.3.2.
OVERALL OBJECTIVE .................................................................................................................................6
SPECIFIC OBJECTIVES ................................................................................................................................6
1.4. RESEARCH HYPOTHESIS .............................................................................................................. 7
CHAPTER TWO .......................................................................................................................... 8
2.0. LITERATURE REVIEW ................................................................................................................. 8
2.1. INTRODUCTION TO GPS ................................................................................................................ 8
2.2. OVERVIEW OF GPS ......................................................................................................................... 8
2.2.1.
GPS SEGMENTS ............................................................................................................................................9
2.3. HOW THE GPS WORKS ................................................................................................................. 10
2.4. DIFFERENTIAL GPS ...................................................................................................................... 11
2.4.1.
ERRORS IN GPS MEASUREMENTS ......................................................................................................... 12
2.5. SATELLITE GEOMETRY MEASURES ........................................................................................ 13
2.5.1.
DILUTION OF PRECISION (DOP).............................................................................................................. 14
2.6. SATELLITE BASED AUGMENTATION SYSTEM (SBAS) ........................................................ 15
2.7. CADASTRAL SURVEYING FOR LAND REGISTRATION IN KENYA..................................... 16
2.7.1.
2.7.2.
2.7.3.
GENERAL BOUNDARIES ........................................................................................................................... 16
PRELIMINARY INDEX DIAGRAMS (PIDs) ............................................................................................. 17
FIXATION OF GENERAL BOUNDARIES ................................................................................................. 18
2.8.
2.9.
2.10.
2.11.
THE REFLY PROCESS ................................................................................................................... 18
CADASTRE 2014............................................................................................................................. 19
CADASTRE 2034........................................................................................................................... 19
GEOREFERENCING ..................................................................................................................... 20
2.11.1.
2.11.2.
NEED FOR GEOREFERNCING ................................................................................................................ 21
METHODS .................................................................................................................................................. 21
iv
CHAPTER THREE .................................................................................................................... 23
3.0. METHODOLOGY............................................................................................................................. 23
3.1. STUDY AREA ..................................................................................................................................... 23
3.2. DATA, EQUIPMENTS AND THEIR SOURCES ............................................................................... 23
3.2.1.
3.2.2.
DATA ............................................................................................................................................................. 24
EQUIPMENTS............................................................................................................................................... 24
3.3. DESCRIPTION OF THE DGPS SURVEY TECHNIQUE ............................................................... 25
3.4. OFFICE PLANNING AND PREPARATION .................................................................................. 25
3.4.1.
3.4.2.
3.4.3.
3.4.4.
SATELLITE COVERAGE ............................................................................................................................ 25
CREATION OF A DATA DICTIONARY .................................................................................................... 26
BASE RECEIVER CONFIGURATION........................................................................................................ 26
ROVER RECEIVER CONFIGURATION .................................................................................................... 26
3.5. FIELD PROCEDURES .................................................................................................................... 26
3.5.1.
3.5.2.
3.5.3.
3.5.4.
RECONNAISSANCE .................................................................................................................................... 26
BASE RECEIVER SET-UP ........................................................................................................................... 27
DATA COLLECTION WITH ROVER RECEIVER..................................................................................... 27
FIELD SKETCHES ....................................................................................................................................... 28
3.6. POST-PROCESSING OPERATIONS ............................................................................................. 29
3.6.1.
3.6.2.
DOWNLOADING AND ARCHIVING ........................................................................................................ 29
PROCESSING ............................................................................................................................................... 29
3.7. MAP PRODUCTION ....................................................................................................................... 31
3.8. GEOREFERENCING AND DIGITIZING IN ARC GIS.................................................................. 32
3.8.1.
3.8.2.
GEOREFERENCING .................................................................................................................................... 32
DIGITIZING .................................................................................................................................................. 33
CHAPTER FOUR ....................................................................................................................... 35
4.0. RESULTS, ANALYSIS AND DISCUSSIONS................................................................................. 35
4.1. PROCESSING RESULTS ................................................................................................................... 35
4.1.1. GEOREFERENCING RESULTS ..................................................................................................................... 36
4.2. ANALYSIS AND DISCUSSIONS ...................................................................................................... 39
4.2.1. IMAGE ANALYSIS ......................................................................................................................................... 39
4.2.2. COORDINATE ANALYSIS ............................................................................................................................. 44
4.2.3. ORIGINAL PID / DGPS AREAS AND PERIMETERS .................................................................................. 45
CHAPTER FIVE ........................................................................................................................ 47
5.0. CONCLUSIONS AND RECOMMENDATIONS ........................................................................... 47
5.1. CONCLUSIONS .................................................................................................................................. 47
5.2. RECOMMENDATIONS ..................................................................................................................... 50
REFERENCES .......................................................................................................................................... 52
APPENDIX 1 ............................................................................................................................................. 54
v
LIST OF TABLES
TABLE 3.1: LINK TABLE INDICATING THE RESIDUALS ................................................................ 33
TABLE 4.1: COORDINATES USING THE SINGLE FREQUENCY DGPS .......................................... 35
TABLE 4.2: COORDINATES FROM THE HANDHELD GPS RECEIVER .......................................... 36
TABLE 4.3: COORDINATES ANALYSIS ............................................................................................... 44
TABLE 4.4: PID/DGPS AREAS AND PERIMETERS COMPARISON .................................................. 45
vi
LIST OF FIGURES
FIGURE 1.1: A SCANNED PID FOR MATHUNTHINI REGISTRATION SECTION, DIAGRAM NO.
10 .......................................................................................................................................................... 2
FIGURE 1.2: A CASE OF CADASTRAL INCONSISTENCY (YONCHON 2000) .................................. 4
FIGURE 2.1: THE GLOBAL POSITIONING SYSTEM (GPS), CONSTELLATION (SOURCE:
WWW.GEOCITIES.COM/MIKECRAMER/GPS.HTM/#1MAGES) ................................................. 8
FIGURE 2.2: THE GPS SEGMENTS .......................................................................................................... 9
FIGURE 2.3: HOW GPS WORKS ............................................................................................................. 10
FIGURE 2.4: GEOMETRY OF DIFFERENTIAL GPS ............................................................................ 11
FIGURE 2.5: GPS RANGE MEASUREMENT ERRORS ........................................................................ 12
FIGURE 2.6: THE SATELLITE BASED AUGMENTATION SYSTEM (CASA, GNSS OVERVIEW,
2010) ................................................................................................................................................... 15
FIGURE 3.1: STUDY AREA ..................................................................................................................... 23
FIGURE 3.2: METHODOLOGY OVERVIEW ......................................................................................... 24
FIGURE 3.3: OVERVIEW OF DGPS SURVEY TECHNIQUE ............................................................... 25
FIGURE 3.4: BASE RECEIVER ............................................................................................................... 27
FIGURE 3.5: DATA COLLECTION WITH THE ROVER RECEIVER .................................................. 28
FIGURE 3.6: PROCESSING...................................................................................................................... 29
FIGURE 3.7: PROCESSING SCREEN ..................................................................................................... 30
FIGURE 3.8: PROCESSING SCREEN ..................................................................................................... 31
FIGURE 4.1: GEOREFERENCED PID ..................................................................................................... 37
FIGURE 4.2: POINT SCENARIOS ........................................................................................................... 38
FIGURE 4.3: DISPUTED BOUNDARY ................................................................................................... 39
FIGURE 4.4: HANDHELD AND DGPS OVERLAY ............................................................................... 40
FIGURE 4.5: INCONSISTENCY .............................................................................................................. 41
FIGURE 4.6: UPPER SIDE ........................................................................................................................ 42
FIGURE 4.7: LOWER SIDE ...................................................................................................................... 42
FIGURE 4.8: UPPER SIDE WITH MORE CONTROLS .......................................................................... 43
FIGURE 4.9: LOWER SIDE WITH LESS CONTROLS .......................................................................... 43
FIGURE 5.1: GEOREFERENCED DIGITAL PID ................................................................................... 49
vii
ABBREVIATIONS
GPS
- Global Positioning System
DOD
- Department of Defense
IOC
- Initial Operational Capability
NAVSTAR
- NAVigation Satellite Timing And Ranging system.
DOP
- Dilution of Precision
PDOP
- Position Dilution of Precision
HDOP
- Horizontal Dilution of Precision
VDOP
- Vertical Dilution of Precision
GDOP
- Geometric Dilution of precision
TDOP
- Time Dilution of Precision
SBAS
- Satellite Based Augmentation System
WADGPS
-Wide Area Differential Global Positioning System
WAAS
-Wide Area Augmentation System
FAA
-Federal Aviation Administration
EGNOS
-European Geostationary Navigation Overlay System.
GAGAN
- GPS Aided GEO Augmented Navigation
SA
- Selective Availability
PID
- Preliminary Index Diagram
LIS
- Land Information System
viii
REPORT ORGANIZATION
CHAPTER ONE:
INTRODUCTION
CHAPTER TWO:
LITERATURE REVIEW
CHAPTER THREE:
METHODOLOGY
CHAPTER FOUR:
RESULTS ANALYSIS
AND DISCUSSION
CHAPTER FIVE:
CONCLUSIONS AND
RECOMMENDATIONS
• Background information
• Problem statement and Justification
• Objectives
• Research hypothesis
• GPS
• Cadastral
• Georeferencing
• Study area
• Data and Equipments
• Description of the DGPS survey technique
• Georeferencing and Digitizing in ArcGIS
• Processing results
• Analysis and Discussion
• Conclusions
• Recommendations
ix
CHAPTER ONE
1.0.
INTRODUCTION
For land registration to be carried out in Kenya, accurate large scale cadastral mapping is
mandatory and an effective survey technique has to be chosen. According to the National Land
Policy, the processes of land surveying and mapping are integral to an efficient land
administration and management system. In addition to preparing the maps and plans to support
land registration, the earth is also mapped for land use planning. The traditional mapping
processes used have been hampered by slow, cumbersome and outdated modes of operation, and
failure to regulate non-title surveys leading to the development of incompatible maps. The
measures that the Government has proposed for the mentioned problem are as follows:
A. Amendment of the Survey Act (Cap 299) to allow:
i.
For the use of modern technology such as Global Navigation Satellite Systems
(GNSS) and Geographical Information Systems (GIS) and streamline survey
authentication procedures.
ii.
Regulation of non-title surveys.
B. Establish a unitary and homogeneous network of control points of adequate density,
preferably using dynamic technology such as GNSS; and
C. Improve mapping standards in general boundary areas so that they fit into a computerized
system. (National land policy, 2009)
Boundaries are often marked by linear features meaning they are surveyed topographically by
ground or photogrammetric techniques. Ground survey methods are normally slow and very
expensive. Currently photogrammetric mapping techniques are still favoured for achievement of
faster land registration progress. Well, this can even be made faster and cheaper through the use
of accurate instantaneous positioning with GPS.
1.1.BACKGROUND INFORMATION
The Land Act 2012 provides that in managing public land on behalf of the national and county
governments, the National Land commission shall identify public land, prepare and keep a
database of all public land which shall be geo-referenced and authenticated by the statutory body
1
responsible for survey. Also, all land to be involved in the compulsory acquisition of interest in
land is to be geo-referenced and authenticated by the office or authority responsible for survey at
both the national and county government.
Again with reference to the Land Registration Act, 2012, all plans kept in the land registry are to
be geo-referenced after a date appointed by the commission. According to section 15 (1) of the
same act, the office or authority responsible for the survey of land shall prepare and thereafter
maintain a map or series of maps to be known as cadastral map(s) for every registration unit. Part
2 of the same section states that the parcel boundaries on such maps shall be geo-referenced and
surveyed to such standards as to ensure compatibility with other documents required under this
act or any other law. In simple terms according to the current National Land Policy no ungeoreferenced cadastral maps will be used for land registration.
Figure 1.1: A scanned PID for Mathunthini registration section, diagram no. 10
2
The main issue of this research is to show how to georeference the PIDs used for land
registration in the general boundary areas. As evident in figure 1 above the PIDs is only a sketch
diagram.
Georeferencing the land parcel boundaries means defining their location in terms of a defined
coordinates systems. Land parcels are geo-referenced using X, Y (2D) coordinate system. Most
rural and suburban land parcels are not geo-referenced and are on general boundaries with some
at accuracy level of meters. Geo-referencing is about accurate spatial information that is critical
in decision making on real estate investment. It will be the cornerstone of the National Spatial
Data Infrastructure (NSDI), an informational infrastructure to provide a platform for economic
development and growth by facilitating data sharing hence optimal use of natural and built
resources
The changing scenario in governance especially for survey and land records department demands
advanced information technologies and tools. The need of the day is creation of a comprehensive
Land Information System (LIS), which has great potential in the field of land records
management and can be used as multi-purpose administrative tool. This contribution could in
turn facilitate the move towards a multi-purpose cadastre. In brief, the current situation of LIS in
Kenya is that with the specific boundary LIS, cadastral plans are being scanned and screen
digitized by survey of Kenya. The general boundary LIS is still all in paper form (Mulaku,
2012).
Maps help determine the location of property, indicate the size and shape of each parcel and
reveal geographical relationships that affect property value. Computerization of map and parcel
data can enhance the capability to manage, analyze, summarize, display and disseminate
geographically referenced information. Over the past few years GPS has emerged as a major tool
for undertaking precise surveys. More recently, it has made inroads in those applications
requiring lower precision surveys and is fast becoming a primary technology for acquiring data
for input into geographical and land information systems (GIS/LIS). GPS receivers can provide
coordinates, which are sufficiently accurate for cadastral purposes.
The precise line of delimitation in the general boundary areas is usually left undetermined.
General boundary surveys also lack controls. Due to this cadastral inconsistency results over
3
time. This means that the property (occupational) boundaries and the cadastral boundaries are
usually in disagreement.
Figure 1.2: A case of cadastral inconsistency (Yonchon 2000)
Cadastral surveys establish property corners, boundaries and areas of land parcels. Conventional
surveying methods have been used, and are still being used, for this purpose. Conventional
methods however have the drawbacks that extensive traversing is required. Extensive clear
cutting and intervening private properties might be required as well. GPS overcomes these
conventional-methods drawbacks. There are several advantages of using GPS for cadastral
surveying. The most important one is that intervisibility between the points is not required with
GPS. This means that extensive traversing is eliminated, clear cutting is not required and
intervening private properties is avoided. Other advantage is that GPS provides user-defined
coordinates in a digital format, which can be easily exported to any GIS system for further
4
analysis. The accuracy obtained using GPS is consistent over the entire network thus making it
the most appropriate tool for data collection towards the creation of a Survey Accurate Cadastral
Map. Such accuracy is not possible with conventional surveying methods. Also, with GPS, one
reference station can support a number of rover receivers. A number of private and governmental
organizations have reported that the use of GPS in cadastral surveying is cost-effective.
To further improve GPS positioning accuracy, differential or relative positioning method, which
employs two receivers simultaneously tracking the same GPS satellites, is used. In this case
positioning accuracy at the level of a few meters increasing to millimeter can be obtained. A
single frequency GPS receiver is basically a GPS receiver that is capable of tracking only the L1
carrier wave signal. However when using single frequency GPS receivers the accuracy of the
positioning decreases, particularly in the height component. One main factor that leads to
degradation in accuracy is the un-modeled ionospheric error. But with the use of differential
technology (DGPS) the overall accuracy usually increases. The discovery of the differential
mode has improved the accuracy of GPS surveys and has gained the capacity to improve the
accuracy of General Boundary surveys and fasten the Fixed Boundary surveys thus reducing the
cost of surveying and enhancing access to land and value and security of tenure.
1.2.PROBLEM STATEMENT AND JUSTIFICATION
1.2.1.
PROBLEM STATEMENT
According to the National land policy all the maps to be used for land registration are first to be
geo-referenced, this means that the PIDs have to be geo-referenced prior to their use in land
registration. The problem is that these PIDs lack coordinates and are generally inaccurate due to
the errors inherent in the enlarged unrectified aerial photos from which they were produced
through tracing. There are also lots of other land parcels that are yet to be surveyed.
With the expensive refly programme long abandoned, and other ground survey methods
considered inacurrate, time consuming and expensive, it is clear that the GPS survey techinique
proves to be the most appropriate technology for the improvement of the accuracy of the general
boundary surveys in addition to surveying the remaining land parcels.
5
1.2.2.
JUSTIFICATION
No land is to be registered using un-georeferenced documents hence the need to have all such
documents georeferenced. Land price is continuously rising and thus the land owners or users
want to occupy the exact amount of land for which they pay for. Old cadastral maps cannot
provide the accuracy people want. The parcel boundaries registered on such cadastral maps are
inconsistent with the real occupation of land even though these maps have been updated
continuously by the subdivision and / or merging of parcels. A parcel area measured on the map
is usually different from what is written in the land register partly because the paper maps shrunk
or stretched in most cases. The lines near map edges hardly meet their counterparts in the
adjoining map sheets at a point.
There is no doubt Kenyans deserve a valuable and secure land tenure system and there is general
agreement that continuous efforts should be invested towards making cadastral surveys cheap,
thereby enhancing access to land. GPS provides the opportunity to fix all forms of general
boundaries thus enhancing the security and value of tenure besides reducing the cost of fixed
boundary surveys. GPS technique is not only accurate but also cheap and faster compared to
other ground survey methods.
1.3.OBJECTIVES
1.3.1.
OVERALL OBJECTIVE
To determine the most effective and efficient approach of georeferencing the PIDs.
1.3.2.
SPECIFIC OBJECTIVES
1) To obtain a PID of the selected research area, georeference and digitize it and then
determine the areas and perimeters of the parcels of interest.
2) Creating an overlay of the PID with a satellite imagery of the same area to help visually
determine the position difference of the parcel boundaries if any.
3) To clearly outline the DGPS survey methodology that can be applied for production of
more accurate digital cadastral maps compared to the PIDs.
4) To compare and contrast the DGPS and handheld coordinates and determine their
difference in terms of the accuracy requirements.
6
1.4.RESEARCH HYPOTHESIS
This study shows how the general boundary surveys can be more accurately done and how
accurate digital maps can be produced more economically compared to the traditional survey
methods. Generally it shows how the general boundaries mapping standards can be improved to
fit into a computerized system.
There is sufficient evidence that the GPS approach offers significant advantages in terms of
productivity. Other advantages of this approach are as follows;

All data are collected on a unified system, thereby ensuring compatibility between
surveys conducted by different parties.

The methodology generates digital data, which will facilitate the transition to a
Multipurpose LIS/GIS in the future.

Even if the data collected at present is not directly utilized in this transition, the
introduction of digital technology into Kenya will prepare the foundation of expertise
required for this transition.

The digital cadastral maps are easily reproducible in the event of loss or damage of the
original and also it’s much easier to update such maps compared to the paper maps.

Different cadastral map scales may be generated with minimum effort to support
alternative applications.
The introduction of modern technology, and the gradual transfer of this technology in Kenya,
will prepare the way for future surveying and mapping activities such as for environmental
assessment studies and AM/FM (Automated Mapping/Facilities Management) applications.
A coordinate based cadastre will create uniformity and eliminate the time and labour consuming
process of data integration and adjustment of dimensions and measurements. Technologically
GIS advanced tools will be used to store, manage and analyze this cadastral data system along
with the GPS technology which is used for the identification of the coordinate based corners in
the field.
7
CHAPTER TWO
2.0. LITERATURE REVIEW
2.1.INTRODUCTION TO GPS
GPS meaning Global Positioning System refers to a satellite-based navigation system developed
by the U.S Department of Defense (DOD) back in the early 1970s. It replaced the Transit
system. GPS was initially developed as a military system but was made available to civilians
later on. With GPS it’s to carry out continuous positioning and obtain timing information under
all weather conditions anywhere in the world.
2.2.OVERVIEW OF GPS
The GPS constellation completed in July 1993 consisted of a nominal of 24 operational satellites
(21 satellites plus 3 active spares). This was known as the initial operational capability (IOC).
Figure 2.1: The Global Positioning System (GPS), constellation (Source:
www.geocities.com/mikecramer/gps.htm/#1mages)
Four GPS satellites are placed in each of the six orbital planes so as to ensure continuous
worldwide coverage. This makes it possible for four to ten GPS satellites to be visible anywhere
8
in the world as long as an elevation mask of 10 degrees is considered. The following are some
characteristics of a GPS satellite orbit.
a) Are elliptical in shape
b) Have a maximum eccentricity of about 0.01
c) Have an inclination of about 55 degrees to the equator.
d) The semi major axis of a GPS orbit is about 26,560 km
e) The orbital period is approximately 12 sidereal hours (approximately 11 hours 58
minutes).
2.2.1.
GPS SEGMENTS
Space segment
1
2
3
1- Data to GPS receiver
2- Data from GPS receiver
3- Navigation data from
GPS receiver
2
User segment
Control segment
Figure 2.2: The GPS segments
Space segment – The GPS satellites helps locate the position being determined by broadcasting
the signal used by the receiver. The signals can be blocked by features such as buildings,
mountains, and even people. The position is usually calculated by using signals from atleast four
satellites and there is usually the need to have a station in an open space for clear reception.
User segment –It comprises of a sensitive receiver which can detect signals (power of the signal
is less than a quadrillionth power of a light bulb) and a computer to convert the data into useful
9
information. GPS receiver helps one locate his or her own position. The users include both the
military and civilians.
Control segment – This ensures that the entire system operates efficiently. Updating of the
satellite signals and keeping of the signals in their appropriate orbits is essential for system
effectiveness.
2.3.HOW THE GPS WORKS
The location of a point is usually determined by applying the concept of resection as long as
distances from a point on the earth to three GPS satellites are known along with the satellite
locations. Theoretically, only three distances to three simultaneously tracked satellites are needed
to obtain a position. However practically, a fourth satellite is usually needed so as to account for
the receiver clock offset.
satellite
satellite
distance
satellite
distance
distance
satellite
distance
Receiver
position
Earth
surface
Figure 2.3: How GPS works
GPS satellites continuously transmit microwave radio signals composed of:
a) Two carriers
b) Two codes (or more for modernized satellites)
c) A navigation message.
10
On switching on a GPS receiver, it picks up the GPS signal through its antenna. It then
processes the signal using its built-in processing software. The result of the signal processing
is usually:
a) The distance to the GPS satellites through the digital codes (known as pseudoranges)
b) The satellite coordinates through the navigation message.
2.4.DIFFERENTIAL GPS
Differential GPS technology is one in which two GPS receivers are usually used to track a single
satellite simultaneously. There is usually the control or reference receiver usually located at a
known position and one or more rover receiver(s). The reference receiver at the known control
point measures the errors in the GPS signals and transmits the corrections to the rover receivers.
The corrections can be real-time or can be computed later on during post-processing. For
effective transmission of the errors the rover receivers have to be within the same geographical
area as the control receiver as the corrections cannot be universal but only useful over a
significant area. The corrections are normally sent every few seconds when using real time
transmission
Corrections
Control receiver
Rover
receiver(s)
Figure 2.4: Geometry of Differential GPS
The DGPS technology is usually capable of improving the accuracy of GPS measurements as it
models the ionospheric errors.
11
2.4.1.
ERRORS IN GPS MEASUREMENTS
As seen earlier on differential GPS works by measuring the errors in GPS signals at a reference
station and sends the corrections to users at the rover station(s). The errors are usually similar for
rover receivers within close range with reference to the control station. The definition of "close"
depends on the specific error.
Figure 2.5: GPS range measurement errors
The errors marked with a diagonal bar in Figure 2.5 above, are the same for close receivers.
These are the errors that are removed in DGPS systems.
Top most is the true range which is usually between 20,000 and 40,000 is required for
navigation. The receiver clock error is estimated each time (commonly every second) when
performing a solution and can be thousands of kilometers in some receivers.
The Selective Availability (SA), had a standard deviation of about 30 meters when it was initially
turned on. It was the dominant source of error for the civilian GPS user but It’s now is zero as it
was turned off. However, when it was on, it was totally removed by DGPS systems.
The ionosphere error varies greatly with time of the day, the location, and the solar cycle. Being
a function of elevation angle, it can be as low as 1m for high elevation angles at night and 50m in
12
the late afternoon, in the tropics, at solar maximum for a 2 degree elevation angle. Low elevation
angle lines of sight have a longer path length within the ionosphere than vertical paths.
The atmospheric error is about 2.5 m for a vertical line of sight. It varies in a very predictably
and is well modeled in most receivers. Only at angles below 5 degrees do complex bending
effects come into play.
The largest error will be the satellite clock error. For better navigation the satellite clocks have
to be synchronized otherwise the navigation will be degraded. This is usually accomplished by
setting all the GPS satellite clocks to a form of Universal Time Coordinated (UTC). The time
differs from UTC by some integer number of seconds. For this reason it is called GPS Time.
Even though extremely good atomic clocks are on each satellite, there is a wander in the clocks.
This is a random process and cannot be modeled. There may also be some residual systematic
error in the predicted clock state.
Multipath errors and errors due to thermal noise inside the receivers are specific to
individual receivers.
The multipath error results when the GPS signals gets reflected by obstructing objects near the
GPS receiver. Through appropriate receiver locations the DGPS technology is usually able to
minimize or totally eliminate this error.
Thermal noise inside the receiver depends on the individual receiver design. It is usually lower
in more expensive receivers such as the dual frequency receivers. It’s however also usually lower
for the newly manufactured receivers. Today the receiver noise varies from 2 m to 10 cm for the
civilian receivers.
2.5.SATELLITE GEOMETRY MEASURES
Satellite geometry represents the geometric locations of the GPS satellites as seen by the
receiver(s). It plays an important role in the total positioning accuracy. It’s usually good for the
well spread satellites and poor for those concentrated together.
13
2.5.1.
DILUTION OF PRECISION (DOP)
This refers to a single dimensionless number used to measure the satellite geometry. The lower
the value of the DOP is the better the geometric strength and vice versa. The DOP number is
computed based on the relative receiver-satellite geometry at any instance i.e., it requires the
availability of both the receiver and the satellite coordinates. Approximate values for the
coordinates are generally sufficient though, which means that the DOP value can be determined
without making any measurements. As a result of the relative motion of the satellite and the
receivers, the value of the DOP will change over time. The changes in the values will however
be generally slow except in the following cases;

A satellite is rising or falling as seen by the user’s receiver

There is an obstruction between the receiver and the satellite
In practice, various DOP forms are used depending on the user needs. For the general GPS
positioning purposes, a user may be interested in examining the effect of the satellite geometry
on the quality of the resulting three-dimensional position (latitude, longitude, and height). This
could be done by examining the value of the position dilution of precision (PDOP). That is,
PDOP represents the contribution of the satellite geometry to the three-dimensional positioning
precision. It can be broken down into two components; horizontal dilution of precision (HDOP)
and vertical dilution of precision (VDOP). The former represents the satellite geometry effect on
the horizontal component of the positioning accuracy, while the latter represents the satellite
geometry effect on the vertical component of the positioning accuracy. Because a GPS user can
track only those satellites above the horizon, VDOP will always be larger than HDOP. As a
result the GPS height solution is expected to be less precise than the horizontal solution. Other
commonly used DOP forms include the time dilution of precision (TDOP) and the geometric
dilution of precision (GDOP), which represents the combined effect of the PDOD and the TDOP.
To ensure high precision GPS positioning, a PDOP value of 5 or less is usually recommended. In
fact, the actual PDOP value is usually much less than 5, with the typical average value in the
neighborhood of 2.
14
2.6.SATELLITE BASED AUGMENTATION SYSTEM (SBAS)
SBAS deliver error corrections, extra ranging signals (from the geostationary satellite) and
integrity information for each GPS satellite being monitored. A government-operated state-space
domain wide area differential global positioning system (WADGPS) includes four satellite based
augmentation systems (SBAS), namely;

Wide area augmentation system (WAAS), which was developed by the U.S. Federal
Aviation Administration (FAA)

European geostationary navigation overlay system (EGNOS), which relies on both GPS
and the Russian GLONASS.

India’s GPS and GEO augmented navigation (GAGAN) system.

Japan’s multifunction transportation satellite (MTSAS) base satellite augmentation
system (MSAS).
Figure 2.6: The satellite based augmentation system (CASA, GNSS Overview, 2010)
15
As shown in Figure 8, in addition to onboard GPS navigation equipment, SBAS comprise:

A network of ground reference stations to monitor GPS signals;

Master stations that collect and process reference station data and generate SBAS
messages;

Uplink stations that send the messages to the geostationary satellites; and

Transponders in the geostationary satellites that broadcast the SBAS messages to the
aircraft.
2.7.CADASTRAL SURVEYING FOR LAND REGISTRATION IN KENYA
Cadastral Surveying and Mapping is the cornerstone of any Cadastral System. The surveys
usually result in the preparation of land registration documents. The process of land registration
and issuing of titles is usually a large scale project and therefore requires survey techniques that
are simple, quick and affordable, thus speeds up the official access to secure land tenure by many
citizens. In Kenya, there are two principal methods of defining boundaries on the ground for
administration of titles; the establishment of marks to define corners and the creation or adoption
of continuous physical features which become recognized as boundaries of the land parcel. To
hasten the registration process, both ground and photogrammetric techniques were adopted.
Cadastral surveying refers to that branch of surveying directly related with provision of title to
land. It is an aspect of land administration and its primary objective is to determine for each
parcel of land its location, extent of boundaries, surface area and separate identity. In its wider
perception cadastral surveying includes valuation and assessment of land and property. They
usually result in the preparation of the land registration documents and they conform to different
requirements depending on the registration legislation in place. In Kenya cadastral surveys are of
two main types:
Fixed boundary surveys

General boundary surveys
2.7.1.
GENERAL BOUNDARIES
In general boundary areas relaxed (approximate) surveys are carried out. General boundary
surveys were introduced along with the land reforms for the individualization of land tenure. The
16
aim was to speed up the issuance of individual title to land owners and to realize as much
cadastral coverage as possible in the former naïve reserves. Under this system, the precise line
of boundaries between parcels is left undefined – it could be the side of a hedge or a fence, or at
the middle. This system relies heavily on demarcation of boundaries in a clear manner using
physical features such as fences, hedges, ditches, etc. The output for these surveys is registry
index maps to support registration. Mutation sheets are produced on subdivision, almagation and
other procedures which require amendments to the registry map. Registry maps differ in terms of
production techniques, content and accuracy.
General boundaries are generally less demanding in standards of survey and tend to be used in
areas with relatively lower land values than fixed boundaries. General boundaries are mainly
found in the rural areas while fixed boundaries mark the extent of properties in the urban areas.
2.7.2.
PRELIMINARY INDEX DIAGRAMS (PIDs)
These maps are produced from enlarged, un-rectified aerial photographs and serve as interim
registry index maps (interim RIM). They are deemed to be interim registry maps because they do
not satisfy the conditions of a registry index map. They are produced from identification and
markings of boundaries on the photograph. Boundaries, which are not identifiable, are marked by
estimation. In a few cases, boundary surveys were done on base maps by plane table methods.
Boundary identification is done by Land Adjudication Department Junior survey Assistants
whose
knowledge of and training in mapping, resulting in very inaccurate documents, but
technical supervision is provided by Survey of Kenya (after 1967). Some of the problems
associated with the use of PIDs include:
Registered areas in most cases differ from the physical ground areas

Sometimes a single land parcel may get registered to two different people

The PIDs are usually not in any coordinate system and the positions of the land parcels as
contained in the diagrams at times differs from their physical locations on the ground

A parcel area may be very clearly available on the PIDs but in truth there is no such
parcel on the ground as at times the boundaries get lost and since these PIDs lack the use
of coordinates then it is impossible to use them to recover the boundaries.
17
2.7.3.
FIXATION OF GENERAL BOUNDARIES
General boundaries as has been introduced, define most of land property in Kenya. They are
found in all consolidation, enclosure and rangeland areas. The registration in these areas is
supported by either the Registry Index Maps (RIM) or the Interim RIMs. Except where it is
noted in the register that the boundaries are “fixed” the registry index map is deemed to indicate
only the approximate position of the boundaries. Section 22 of the Registered Land Act provides
for “fixing” of general boundaries and to cause the positions of the boundaries to be defined
precisely. This may be done if:
The Registrar considers it desirable.

An interested party makes an application to the Registrar
If an approval for fixation of boundary is obtained, how and by whom such fixation may be done
shifts to the Survey Act Cap 299, and considering that most general boundaries exist in the rural
areas of Kenya where there is lack of survey control, GPS proves most useful in such
circumstances.
2.8.THE REFLY PROCESS
Refly is the process associated with the production of Registry Index Maps (RIMs). Here the
land owners are usually encouraged to grow hedges to show their parcel boundaries and once the
hedges are “air visible” aerial photography is usually done and accurate maps (RIMs) showing
the parcel boundaries are prepared from rectified aerial photos by tracing. This can be referred to
as aerial survey. The missing boundaries are then marked by the use of ground survey methods.
RIMs have been produced for Kiambu district, most of Nyeri and Kirinyaga districts and parts of
Meru and Murang’a districts. The RIMs are also currently being produced from satellite
imagery.
Even though the RIMs produced are more accurate compared to the PIDs, the re-fly is a time
consuming and expensive process and as a result was long abandoned by the government as soon
as it was implemented. The ‘push button’ GPS technology which requires relatively low-skilled
personnel can be used to cater for the inaccuracies which in return improve cadastral surveys.
18
2.9.CADASTRE 2014
In 1998, the FIG working group submitted the booklet "Cadastre 2014 – A Vision for a Future
Cadastral System" to the XXII FIG Congress in Brighton. The booklet identified the cadastral
trends at that time, and looked into the developments of the cadastre in the future (Kaufmann
and Steudler, 1998). The publication is known since as "Cadastre 2014" and basically gives six
vision statements for a future cadastral system and several recommendations related to those:
1) "The cadastre of the future will show the complete legal situation of land, including
public rights and restrictions!" As land becomes a scarce resource and more and more
public rights and restrictions influence the private landownership, the cadastral system of
the future needs to show the complete legal situation in order to provide the required land
tenure security.
2) "Separation between maps and registers will be abolished!" The separation was
historically necessary because of the available technology at the time, but this can
nowadays be overcome, at least technically if not institutionally as well.
3) "Cadastral mapping will be dead! Long live modelling!" The production of plans and
maps has always been the main objective and raison d'être of surveyors; modern concepts
and technology provide different and much more advanced opportunities, which
surveyors need to acknowledge by adopting principles from information technology.
4) "'Paper and pencil'-cadastre will have gone!" Digital technology will be a prerequisite
for efficient and adequate service.
5) "Cadastre of the future will be highly privatized! Public and private sectors are working
closely together!" Public systems tend to be less flexible and customer oriented than
private organizations; the private sector can help to improve the efficiency, flexibility and
introduce innovative solutions while the public sector can concentrate on supervision and
control.
6) "Cadastre will be cost recovering." Cost/benefit analysis will become an important aspect
of cadastral reform projects and the considerable investments need to be justified.
(Daniel Staudler,2006)
2.10. CADASTRE 2034
Cadastre 2034 is an ongoing dialogue within the global cadastral fraternity, started at the FIG
2010 Congress in Sydney, Australia. The objective of the dialogue is to establish to what extent
19
the vision of cadastre 2014 has been realized and to set out a vision for how cadastres might be
20 years after 2014, that is, 2034.
Towards a vision for cadastre 2034, the discussion revolves around six points.

Moving from approximate boundary representation towards surveying accurate boundary
representations.

Shift focus from parcel based systems to systems of layered property objects, that is,
shifting towards 3-D cadastre.

Expansion from 2-D cadastres to include the 3rd (height) and 4th (time) dimensions.

Access and updating of cadastral information in real time.

Making national cadastres interoperable at regional and global levels.

Representation of organic and fuzzy boundaries.
The use of the RTK GPS surveys can easily enable the accessing and updating of cadastral
information in real time.
2.11. GEOREFERENCING
To Geo-reference means reference to an object by a specific location either on, above or below
the earth’s surface. Geo-referenced boundary means reference to boundaries of a parcel of land
to a specific or unique location on, above or below the earth’s surface (survey act cap 299)
This term is also defined in the Land Registration Act, 2012 as referring to an object using a
specific location on, above or below the earth’s surface. In geo-referencing a location is defined
using a defined coordinate system of the country. Note that the Act also provides that the
Registrar shall register long term leases and issue certificates of leases over apartments, flats,
maisonettes, townhouses or offices having the effect of conferring ownership, if the property
comprised is geo-referenced and approved by the statutory body responsible for the survey of
land.
20
In other words to georeference means to define something in-terms of its existence in physical
space. That is, establishing its location in terms of map projections or coordinate systems. The
term is used both when establishing the relation between raster or vector images and coordinates,
and when determining the spatial location of other geographical features. Examples would
include establishing the correct position of an aerial photograph within a map or finding the
geographical coordinates of a place name or street address. This procedure is thus imperative to
data modeling in the field of geographic information systems (GIS) and other cartographic
methods. When data from different sources need to be combined and then used in a GIS
application, it becomes essential to have a common referencing system. This is brought about by
using various geo-referencing techniques. Most geo-referencing tasks are undertaken either
because the user wants to produce a new map or because they want to link two or more different
datasets together by virtue of the fact that they relate to the same geographic locations.
2.11.1. NEED FOR GEOREFERNCING

Geo-referencing is crucial to making aerial and satellite imagery, usually raster images,
useful for mapping as it explains how other data, such as the above GPS points, relate to
the imagery.

Very essential information may be contained in data or images that were produced at a
different point of time. It may be desired either to combine or compare this data with that
currently available. The latter can be used to analyze the changes in the features under
study over a period of time.

Different maps may use different projection systems. Georeferencing tools contain
methods to combine and overlay these maps with minimum distortion.

Using georeferencing methods, data obtained from surveying tools like total stations may
be given a point of reference from topographic maps already available.
2.11.2. METHODS
There are various GIS tools available that can transform image data to some geographic control
framework, like Arc Map, PCI Geomatica, or ERDAS Imagine. One is able to georeference a set
of points, lines, polygons, images, or 3D structures. For instance, a GPS device will record
latitude and longitude coordinates for a given point of interest, effectively georeferencing this
21
point. A georeference must be a unique identifier. In other words, there must be only one
location for which a geo-reference acts as the reference. Images may be encoded using special
GIS file formats or be accompanied by a world file.
To geo-reference an image, one first needs to establish control points, input the known
geographic coordinates of these control points, choose the coordinate system and other
projection parameters and then minimize residuals. Residuals are the difference between the
actual coordinates of the control points and the coordinates predicted by the geographic model
created using the control points. They provide a method of determining the level of accuracy of
the geo-referencing process.
22
CHAPTER THREE
3.0. METHODOLOGY
3.1. STUDY AREA
The research area Mathunthini registration section is in Kenya, Machakos County, Mwala
District. Machakos stretches from latitudes 0º 45’S to1º 31’S and longitudes 36° 45’E to 37° 45’
E. It covers an area of 6,281.4 km2 most of which is semi-arid. High and medium potential areas
where rain fed agriculture is carried out consist of 1,574km2 or 26% of the total area.
Figure 3.1: Study Area
3.2. DATA, EQUIPMENTS AND THEIR SOURCES
The data and materials required for the planning and implementation of the research were as
follows.
23
3.2.1.

DATA
A preliminarily index diagram of Mathunthini registration section from the survey of
Kenya
3.2.2.
EQUIPMENTS

Single frequency differential GPS receivers.

Handheld GPS receiver

Computer

Flash disk

Compact disc

Tape measure

Tripod stand.
Figure 3.2: methodology overview
24
3.3. DESCRIPTION OF THE DGPS SURVEY TECHNIQUE
The survey technique was divided into three main steps:

Office planning and preparation,

field procedures

Post-processing operations.
Figure 3.3: overview of DGPS survey technique
3.4. OFFICE PLANNING AND PREPARATION
3.4.1.
SATELLITE COVERAGE
Comprehensive planning and organization make up an essential part of the DGPS survey
technique. Prior to the survey, the satellite coverage of the area of study was determined where at
any instant 11 to 13 satellites could be acquired by the receivers.
25
3.4.2.
CREATION OF A DATA DICTIONARY
A data dictionary is a catalog of information about the definition, structure, and usage of the
data. The use of a data dictionary was recommended. It helped structure and guided the data
collection process and provided additional meaning to the resulting data files.
A data dictionary was created for the parcel corners and included the following attributes; Parcel
ID, Date, and Time.
3.4.3.
BASE RECEIVER CONFIGURATION
Achieving good accuracy with DGPS is possible only under specific operational conditions.
These conditions relate to the circumstances under which data are collected. The reference and
rover receivers should be configured to ensure compliance with certain requirements. The
following parameters were set for the base receiver:

Logging interval: 1 second

PDOP mask: 1.9
The logging interval specifies the regularity with which positions are stored within the receiver.
The PDOP mask set ensured that data was collected only when favorable satellite constellation
existed.
3.4.4.
ROVER RECEIVER CONFIGURATION
The rover receiver was configured in a similar way. In order to achieve good accuracy with the
given hardware configuration, the logging interval at the base and rover receivers were identical.
In addition, the following parameters were set for the rover receiver:

Antenna height: 2m

Datum: WGS 84
3.5. FIELD PROCEDURES
3.5.1.
RECONNAISSANCE
Reconnaissance involves procedure that brings to bear the general overview of the working area.
This applies to all surveying techniques but is especially true for GPS surveying. During the
recce the appropriate location for the base station was identified putting into consideration
security issues. The location was also at an open horizon and away from obstructing surfaces.
The size of the areas that were being surveyed was also estimated to help determine the
26
occupation period for the rover station. These procedures enable the measurement process to be
optimized and organized in a logical fashion.
3.5.2.
BASE RECEIVER SET-UP
The base receiver was set up and the configuration checked once more. The receiver was
centered and leveled over the mark as shown in the figure 3.4 below.
Figure 3.4: Base Receiver
It was then powered up and set to begin collecting data. The height of the antenna was measured
(1.375m) and recorded in a field book, together with the station name, the date and time. Care
was taken to ensure that the base station was collecting data prior to the rover receiver, since
only simultaneous data between base and rover receivers may subsequently be differentially
corrected.
The issue of security was addressed by setting up the base receiver in the land of a friendly land
owner.
3.5.3.
DATA COLLECTION WITH ROVER RECEIVER
The antenna was mounted on a range pole, and the data collector was firmly clipped on the pole.
This configuration was well suited for the mobility that was required during the fieldwork. On
27
powering up the rover receiver the configuration was once more checked to ensure that nothing
had been altered.
Figure 3.5: Data Collection with the Rover Receiver
Before collecting the data the name of the file in which the data was to be stored was specified.
The default filename was selected then edited, this was so as to have the date and time of file
creation reflected. The measurement process involved the identification of each parcel corner,
the collection of DGPS data at each point, and the annotation of a field sketch. Occupation time
was 7 minutes in some stations and 5 minutes in others depending on how far or close the rover
station was from the base station.
3.5.4.
FIELD SKETCHES
Field sketches are an essential part of the data-collection procedure. Since DGPS measures point
positions, it is necessary in the post-processing stage to connect these points in order to depict
the boundary lines to be shown on the cadastral map. Due to this adequate notes were taken in
the field. The PID of Mathunthini registration section was also being used to identify the
boundaries which could be labeled on the diagram once identified on the ground. The same name
28
given to the points during storage in the DGPS receiver was labeled against the specific
boundary corners. In addition field pictures were taken.
3.6. POST-PROCESSING OPERATIONS
3.6.1.
DOWNLOADING AND ARCHIVING
Later on after the field work the data was downloaded into a laptop prior to processing. Backups
were also created on a flash disk.
3.6.2.
PROCESSING
All data collected and stored in the receiver relate to the World Geodetic System of 1984 (WGS
84) reference ellipsoid. It is recommended that the data be processed on WGS 84 and converted
to the local datum once processing has been completed. The differential correction process was
conceptually divided into the following steps:
Figure 3.6: processing
These were executed as functions within the GIS data Pro software and are discussed in more
detail as follows.
i.
The base data file containing all data collected for the observing session was entered.
29
ii.
The reference position of the base station entered related to the WGS 84 ellipsoid there
was thus no need for position transformation
iii.
When executing the differential correction procedures two files were created. The first
was the difference file, which contained the actual correction values for the satellite
measurements on an epoch-by-epoch basis. This file was used to differentially correct the
rover file collected during the same time period. The second file was the correction file,
which contained the differentially corrected rover positions.
iv.
The differentially corrected rover positions were then exported from the GIS data pro
software in an usable format. The software automatically averaged all positions collected
for each of the parcel corner. An ASCII output format was utilized. This file then served
as input into the transformation software.
v.
The final results were then transformed to the local datum in this case arc 1960 for the
georeferencing of the land parcels
The following are some of the post processing screens using the GIS data pro software.
Figure 3.7: Processing Screen
The diagram indicates the various epochs.
30
Figure 3.8: Processing Screen
3.7. MAP PRODUCTION
The data collected and processed as described above are of little value to the property and land
registration system unless they are presented in graphical format. This formatting represents the
final step, and one of the most critical, in the DGPS survey technique, since it ensures that the
GPS data are compatible with the traditional measurement process currently utilized for parcel
mapping in Kenya. The ARC GIS 1.1 software was used for the production of the final
“cadastral map.”
The data collection process was also simultaneously done using the handheld GPS receiver
where it involved ensuring that the PDOP value was good enough and then picking of the point
coordinates.
During the field work 11-13 satellites were acquired by the receivers and 7-12 satellites were
used at the different data collection points. At some points all the acquired satellites could be
utilized and this indicated good PDOP values such as 1.5 with the range of the PDOP values
31
obtained being 1.5-3.5, the cycle slip which helps determine the number of times that the signals
got interrupted ranged from 0-4.
3.8. GEOREFERENCING AND DIGITIZING IN ARC GIS
3.8.1.
GEOREFERENCING
Georeferencing a raster dataset by definition means to define its location in terms of map
projections or coordinate systems. The following steps were my basis for georeferencing the PID
as well as the google earth image.
1) A new blank arc map was opened and the raster image (pid.jpeg) added
, this can
also be added from arc catalog.
2) Making sure that pid.jpeg was the selected layer in the georeferencing toolbar.
3) Georeferencing>fit to display was selected to bring the pid.jpeg image into the extent
of data view.
4) Add Control Points
Tool, from the Georeferencing toolbar was clicked on. Starting
at the first selected point of the pid.jpg image, the first control point was created by
clicking on the center of the intersection of the first point then right clicking and
selecting input XY data, typing in the coordinates and finally clicking OK.
5) All four control points at the corner points of the pid.jpg image were created as in steps
(1 to 4) above.
6) The View Link Table
button in the georeferencing toolbar was opened to view the
transformation data and the residual and RMS data for each link. All the numbers under
the residual column were all in agreement and none was too large the conclusion that all
the control points were properly positioned was drawn.
32
Table 3.1: link table indicating the residuals
7) Being satisfied with the results, the Update Georeferencing or Rectify from the
Georeferencing Toolbar was chosen thus storing the transformation information
internally within the raster dataset.
The above steps were repeated for the PID.jpeg and the google earth.jpeg images.
3.8.2.
DIGITIZING
This refers to the process of converting raster data to vector data. The guidelines that follow were
the ones utilized in the creation of the various features which included: line, polygon and point
features.
1) Creating an empty shape file. Arc Catalog was opened and the location of your
current mxd file browsed to. New shape file was created by right clicking on the folder
then going to new � Shape file… and opening the Create New Shape file window. The
polygon shape file was then named land parcels. Clicked on Edit… to see the
Coordinate System of the file and clicked on import in the Spatial Reference
Properties window to use the projection of the already existing layer. OK and OK again
were clicked on to create the shape file. Then new shape file was then added to arc map.
2) Adding a new field in the Attribute Table. The Attribute Table of the land parcel shape
which was empty was opened, clicked on the Options button then Add Field… this was
only possible in the stop editing mode.
3) Digitizing land parcels and entering tabular data. The Editor Toolbar was made
visible.
33
4) To begin editing the Editor Menu then Start Editing was clicked on. On being prompted
to choose the folder that contained the land parcel shape file I did so and then clicked
OK. The polygon images were then created by choosing the polygon tool and tracing
over the land parcels.
For the point and line features the above steps were repeated but now the point and polyline tools respectively were chosen instead of the polygon tool.
34
CHAPTER FOUR
4.0. RESULTS, ANALYSIS AND DISCUSSIONS
4.1. PROCESSING RESULTS
TABLE 4.1: Coordinates using the single frequency DGPS
POINT name
1
2
3
4
5
6
7
8
9
10
11
12
13
EASTING
(M)
330,333.08
330,637.12
330,765.96
330,886.86
330,192.01
330,520.24
330,534.70
331,038.60
330,957.35
330,766.01
330,515.41
330,585.89
330,624.52
NORTHING
(M)
9,845,867.53
9,845,113.30
9,844,645.76
9,844,775.61
9,846,128.50
9,846,417.24
9,846,430.87
9,846,078.22
9,845,961.19
9,845,801.94
9,845,925.21
9,846,042.17
9,846,283.41
HEIGHT
(M)
1,281.99
1,290.32
1,294.86
1,294.66
1,278.18
1,265.46
1,266.54
1,260.60
1,262.33
1,273.32
1,280.52
1,280.58
1,266.43
The table above indicates the processed set of the DGPS coordinates. All are in meters.
35
TABLE 4.2: Coordinates from the handheld GPS receiver
POINT NAME
1
2
3
4
5
6
7
8
9
10
11`
12
13
EASTING NORTHING
(M)
(M)
330,333
9,845,876
330,643
9,845,115
330,768
9,844,648
330,887
9,844,779
330,193
9,846,130
330,519
9,846,417
330,539
9,846,433
331,040
9,846,083
330,959
9,845,964
330,768
9,845,804
330,520
9,845,926
330,590
9,846,045
330,620
9,846,281
HEIGHT
(M)
1,302.70
1,310.87
1,315.92
1,315.43
1,298.86
1,284.92
1,281.31
1,284.20
1,285.16
1,294.29
1,302.46
1,306.79
1,282.76
The table above indicates the handheld GPS coordinates which are also in meters.
4.1.1. GEOREFERENCING RESULTS
The scanned PID was georeferenced using DGPS coordinates of four corner points in ArcGis
10.1. The PID was also georeferenced using the handheld coordinates and the Google earth
image was georeferenced using the DGPS coordinates. The georeferenced PID was tilted in the
clockwise direction with respect to the scanned PID as shown in figure 4.1. This is because it
was now defined in-terms of its existence in physical space. That is, its location in terms of map
projections or coordinate systems had been established.
36
Figure 4.1: Georeferenced PID
After georeferencing three point scenarios arose as indicated in figure 4.2 below.
37
2
3
1
Figure 4.2: Point scenarios
In scenario 1 the point is exactly at the middle of the intersection of the parcel corner. All the
four corner points which were used for georeferencing were as shown. This was however the
expected result for the control points. In scenario 2 the points seem to be just close to the
intersection but not really at the middle. This was the case with most of the other points which
were located within the area that was being georeferenced. Scenario three was rare but present.
38
The point was extremely off. The reason for this scenario is further explained in the analysis
section that follows.
4.2. ANALYSIS AND DISCUSSIONS
The data collected was reduced and analyzed using Microsoft Excel. The accuracy issues
concerning the PID have also been analyzed with the aid of a satellite image from Google earth
that was extracted using the DGPS coordinates.
4.2.1. IMAGE ANALYSIS
Scenario 3:
Boundary is
supposed to be
straight as shown by
the Google image
Figure 4.3: Disputed boundary
As informed by the parcel owner the boundary was disputed and was supposed to be straight and
not curved as indicated on the PID. The GPS point was not off but rather the boundary was
displaced. To support the argument, the satellite image helps indicate the right boundary.
39
The river seems to have changed its course and this helps emphasize on the need to have the
general boundaries fixed. Having such a river indicate the boundary can only lead to disputes.
Figure 4.4: Handheld and DGPS Overlay
The diagram above is an overlay of the digitized PID (the boundaries are colored) georeferenced
using DGPS coordinates with the one georeferenced using handheld coordinates (the boundaries
are dark). It helps visualize the difference in the coordinates. The visual difference might seem
to be small but it is significant as indicated by the position difference values in table 4.3 in page
44.
In the next diagram another case of having boundaries indicated on the PID yet not existing on
the ground is illustrated.
40
The road is totally off and so is
the boundary that it represents.
Figure 4.5: Inconsistency
The image above indicates a case of misrepresentation. The road is totally displaced from the
right route that is indicated on the google image. This helps prove right the point that the PIDs do
not at some point represent what is on the ground.
The overlays of the digitized PID with the google earth image shows how the boundaries are
slightly off in some parts and totally displaced in others. At this point it’s important to remember
that the PID is erroneous not only in terms of incorrect areas but also in boundary representation
and their continued use can only jeopardize cadastral surveys leave alone causing disputes.
The next set of diagrams helps indicate two different results obtained when georeferencing is
correctly done and when mistakes are done during the choice of controls respectively.
41
Figure 4.6: upper side
Figure 4.7: lower side
42
Figure 4.8: upper side with more controls
Figure 4.9: lower side with less controls
43
The last two sets indicate the problem that arises when controls are not evenly fixed. The
boundaries on the upper side with more controls are better off than those on the lower side with
lesser controls. This is because the PID was not firmly fixed in its position on the lower side due
to the inadequate controls.
4.2.2. COORDINATE ANALYSIS
Correlation is the attempt to measure the strength of such relationship between two variables by
means of a single number called sample correlation coefficient(r). In probability theory and
statistics, correlation coefficient indicates the strength and direction of linear relationship
between two random variables (Webster 1998). In this research observations for the different
receivers were the basis upon which relationship was obtained.
Table 4.3: Coordinates Analysis
EASTINGS (m)
POINT
1
2
3
4
5
7
8
9
10
12
13
14
15
Mean
Difference
%
MEAN
STD DEV
VARIANCE
95% C I
Correlation R
DGPS
330,333.08
330,637.12
330,765.96
330,886.86
330,192.01
330,520.24
330,534.70
331,038.60
330,957.35
330,766.01
330,515.41
330,585.89
330,624.52
Handheld
330,333
330,643
330,768
330,887
330,193
330,519
330,539
331,040
330,959
330,768
330,520
330,590
330,620
NORTHINGS (m)
Difference
(m)
DGPS
0.08
-5.88
-2.04
-0.14
-0.99
1.24
-4.30
-1.40
-1.65
-1.99
-4.59
-4.11
4.52
9,845,867.53
9,845,113.30
9,844,645.76
9,844,775.61
9,846,128.50
9,846,417.24
9,846,430.87
9,846,078.22
9,845,961.19
9,845,801.94
9,845,925.21
9,846,042.17
9,846,283.41
Handheld
9,845,876
9,845,115
9,844,648
9,844,779
9,846,130
9,846,417
9,846,433
9,846,083
9,845,964
9,845,804
9,845,926
9,846,045
9,846,281
-10.93%
330,642.90
239.9997368
68709.30264
154.9028894
Difference
(m)
Position
difference
(m)
-8.47
-1.70
-2.24
-3.39
-1.50
0.24
-2.13
-4.78
-2.81
-2.06
-0.79
-2.83
2.41
8.47
6.12
3.03
3.39
1.80
1.26
4.80
4.98
3.26
2.86
4.66
4.99
5.12
15.4%
330,645
240.0484727
57623.26923
130.4894384
0.999928
9,845,805.46
9,845,808
588.4776086 587.9556743
346305.8958
345691.875
319.8941938 319.6104723
0.999991
44
Values of r lies between -1 and +1, when r is equal to +1, it implies a perfect linear relationship
with a positive slope while r= -1 implies a perfect linear relationship with a negative slope.
Values near zero indicate little or no correlation (Al-Hassan, 2008). The calculated r by using
data for the handheld and differential GPS receivers’ was 0.999928 and 0.999991 for the
Eastings and Northings respectively. The two sets of coordinates thus showed strong positive
correlation.
The position difference, last column on table 3 is obtained from the function Pol (change in
eastings, change in northings) = Distance, using the scientific calculator. This can also be
determined adding the square of the difference in the northings and eastings and then finding the
square root. For example for the first point: (0.082 + (-8.47)2)1/2= 8.470377 which is
approximately 8.47. It refers to the difference between the position obtained using the DGPS
with that obtained using the handheld GPS receiver. The differences obtained were as high as 8m
with only two positions having less than 3m. The general boundaries are usually surveyed to an
accuracy of 3m. Due to the large values in the position differences, the handheld GPS receiver
did not meet the 3m accuracy. The low accuracy is mainly due to the effect of the un-modeled
ionospheric errors. The DGPS technology is capable of correcting this error.
4.2.3. ORIGINAL PID / DGPS AREAS AND PERIMETERS
The table that follows contains DGPS areas and perimeters computed by (Wangoshi 2001). The
PID areas and perimeters were obtained using Arc GIS 10.1 after digitizing.
TABLE 4.4: PID/DGPS AREAS AND PERIMETERS COMPARISON
Parcel
Number
603
618
572
552
515
607
Correlation
coefficient (r)
PID
AREA (m2)
DGPS
DIFF
1 041
3 409
1 234
10 226
2 615
8 729
9 185
-794
7 495
%
diff
29.17
-2.52
23.80
14 511
30 114
15 603
49.55 1,261
0.901486
PERIMETER(m)
PID DGPS DIFF
%
diff
437
438
1
-0.19
266
235
-31
5.74
516
486
-30
5.56
781
0.964248
45
-480
88.89
Parcels 552, 515 and 607 have combined areas and perimeters as they got to be owned by one
person who eventually merged them to a single larger parcel. Both the areas and perimeters
showed strong positive correlation. The calculated percentages ranged from 23% to 49% for the
areas and 0.19% to 88.89% for the perimeters. The allowed discrepancy in the areas is usually
around 10%. It has been found in other research works that the area discrepancies can be up to
more than 50% (Mulaku, 1996). The PIDs can therefore not be used for accurate parcel area
calculation.
46
CHAPTER FIVE
5.0. CONCLUSIONS AND RECOMMENDATIONS
5.1. CONCLUSIONS
The application area using DGPS has proven it to be a useful tool for general boundary
georeferencing and fixation if the needed precautions are adhered to. The DGPS technique
employed in the determination of the 3-D coordinates of the parcel corners yielded acceptable
results. The handheld results were not accurate enough.
From the analysis carried out, the following results were obtained, a percentage mean difference
of -10.93% and 15.4% in the Eastings and Northings respectively, for the DGPS and handheld
coordinates. The Northings showed a much higher difference compared to the Eastings. The two
sets of coordinates did show strong correlation but their position difference proved the difference
between them to be significant. From the calculated position differences between positions
obtained using the DGPS and those using handheld values as high as 8m were obtained. Among
the 13 points only two positions which were less than 3m which is usually the survey accuracy
for the general boundary surveys. This helps conclude that the handheld coordinates do not meet
the desired boundary tolerance distance for the fixation of the general boundaries. Thus if used
for georeferencing the results would not be accurate meaning that among the two sets only the
DGPS coordinates are acceptable.
The handheld GPS receiver does not only fail to meet the 3m tolerance but also makes it
impossible to improve the general boundary accuracy not forgetting that the aim is to make it
similar to that of the fixed boundaries that carries a 3cm tolerance. Handheld GPS is cheaper
compared to the DGPS receivers but their use would not yield acceptable results. DGPS
receivers are therefore the most efficient and effective. The single frequency DGPS receivers are
affordable compared to the double frequency ones hence the single frequency ones are
recommended.
From the analysis using the satellite imagery, the parcel boundaries on the PID seemed to not
totally coincide with those on the image completely. This is because the georeferencing was
47
obviously a bit different while selecting the control points. It is not possible to select the exact
same locations for the imagery and the PID. Some other boundaries were totally off and others
indicated where they didn’t even exist, this helps support the point that the PIDs are erroneous.
They represent the land parcels topologically and not topographically.
With the fact that the PIDs are erroneous computerizing them through digitizing doesn’t improve
their accuracy. The accuracy of the coordinates obtained thereafter will still be having the same
accuracy as that of the paper map. The use of such coordinates will not help meet the needs of
the users at all. This means that the georeferenced digital PID shown in figure 5.1 below is as
erroneous as the paper one.
48
Figure 5.1: Georeferenced Digital PID
49
5.2. RECOMMENDATIONS
Since the expensive refly programme was long abandoned and the other ground survey methods
considered time consuming and inaccurate, survey regulations should be revised to accommodate
alternative technologies to improve cadastral surveys. The only solution is to carry out a resurvey
of the land parcels using the accurate DGPS technology and create more accurate cadastral maps
compared to the PIDs. It’s not possible to do the resurvey in the whole country at once. It can be
divided into various blocks and the resurvey done block by block. This will help ensure that all
areas under general boundaries are resurveyed and those that were not initially surveyed are also
surveyed.
Another recommendation is that instead of having to resurvey all the land parcels, satellite
images in combination with the DGPS survey technique could be used to improve the PIDs. This
can be done by first fixing the controls using the DGPS followed by scanning, georeferencing
and digitizing of the PIDs after which they are overlayed with georeferenced satellite images and
the boundaries corrected by shifting or moving them. Where the boundaries are not clear on the
imagery the DGPS can then be used to locate them. This will help ensure that the PIDs represent
what’s on the ground which will lead to improved parcel areas hence the general accuracy.
The google earth images are not encouraged as there is always the need to georeference them and
this might result in more error creation instead of having solutions. High resolution satellite
images are more accurate than the google earth ones and there is usually no need to georeference
them as they are usually already georeferenced.
When it comes to georeferencing a minimum of four points should be used as the control points
and they should be well spread preferably at the corners of the blocks otherwise if concentrated
at a particular location other parts of the diagram will not be well fixed.
In conclusion the following are some recommended guidelines when it comes to the choice of
control points:
o They should be easy to confirm as representing the same geographic location e.g.
(road intersection boundary corners, etc.).
o They should be spread across the image to be registered, one suggestion is to
select a control point near each of the corners of the image and few others evenly
spread throughout the interior.
50
o
Good overlap is also important when working with more than one dataset.
o Make sure you are clicking as close as possible to the same geographic location,
zooming in and the use of snapping tool can help in this process.
51
REFERENCES
1.
Ahmed El-Rabbany, Introduction to GPS: the Global Positioning System, Second
Edition, Boston/London; Artech house.
2.
Bernard Agina Oluande, master’s thesis An Assessment of the Accuracies of
Preliminary Index Diagrams (PIDs) as used in Land Registration in Kenya master’s
program land management and land tenure Munich 26/03/2004 (Access time: 11/5/2012,
6:53 P:M)
3.
Daniel STEUDLER, Switzerland, Cadastre 2014 – Still a Vision. TS12 – Cadastre
2014 and Cadastral Modeling, shaping the change XXIII FIG Congress Munich,
Germany, October 8-13, 2006 (Access time: 11/15/2012, 4:38 P:M)
4.
Dr S.M Musyoka Adjustment Theory class lecture notes FGE 3 (unpublished)
5.
George Oner Ogalo.: GPS in Cadastres: A Case Study of Kenya, FIG XXII
International Congress, Washington, D.C. USA, 19-26 April, 2002 (Access time;
11/5/2012 7:05 P.M.)
6.
Grenville Barnes, Bruce Chaplin, D/ David Moyer, GPS methodology for cadastral
surveying in Albania. Working paper no.17 Albania series, Land Tenure Centre,
university of Wisconsin Madison August, 1998. ( Access time: 1/17/2013, 6:37 P.M)
7.
James OSUNDWA, Eric NYADIMO and David SIRIBA, Kenya. Challenges in the
establishment of a National cadastral data model for Kenya – Towards the creation of a
National Spatial Data Infrastructure (access time; 8/5/2012 3:37 P:M )
8.
Joe Sass, Low cost, High Accuracy, GNSS Survey and Mapping, Coastal Areas and
Land Administration- Building the Capacity 6th FIG Regional Conference San José,
Costa Rica 12-15 November 2007 (Access time: 11/15/2012, 4:21 P:M).
9.
Land Act 2012, laws of Kenya.
10.
Land Registration Act 2012, laws of Kenya.
11.
Maxwell Owusu Ansah, Understanding GPS Processing and Results, IPS5.4-GNSS
Processing and Applications shaping the change XXXIII FIG Congress Munich,
Germany, October 8-13,2006 (Access time11/15/2012, 4:27 P:M)
12.
Mulaku, G.C. and McLaughlin, J: Concepts for Improving Property Mapping in
Kenya, South African Journal of Surveying and Mapping, Vol.23, Part 4, April 1996, pp.
211 –216, 1996.
52
13.
National Land Commission Act 2012, laws of Kenya.
14.
Ondulo J.D. and Kalande W. Kenya, High Spatial Resolution Satellite Imagery for PID
Improvement in Kenya, TS 79 Cadastral Maps, Shaping the change, XXIII FIG
Congress, Munich, Germany, October 8-13, 2006 (Access time: 11/15/2012, 3:49 P:M)
15.
Prof G.C Mulaku Land Information System class lecture notes, FGE 5 (unpublished)
16.
Sectional paper on Kenya National Land Policy, August 2009.
17.
Survey Act Chapter 299, laws of Kenya.
53
APPENDIX 1
Why use single frequency GPS receivers and not dual frequency receivers? Dual frequency
receivers which measure both the L1 and L2 carrier phase measurements is mostly used in
modern RTK GPS kits. Using two GPS receivers, the baseline between the two station marks is
determined along with the unknown ambiguities. The baseline length determines the duration
necessary for the observation session as there is always the need for a change in satellite
geometry. Usually, the longer the baseline the longer the observation session required. For this
research work a static mode was used for data collection. With the fact that the dual frequency
receivers are usually required for longer baselines normally 15 km and above the single
frequency receivers which are applicable for shorter lengths are best fit for this study as the
length of the parcels measured are meters long. The single frequency receivers can be used for
baselines of length between 10 km to 15 km as well. In conclusion most precise GPS baseline
achievable is on L1 only fixed solution. Thus, single frequency receivers are most suitable for
short baselines but the results must be post-processed for good accuracies. In addition the single
frequency receivers are also cheaper than the dual frequency receivers.
54