Azimuth GPS

Comparative Analysis of GPS Azimuth and Derived Azimuth for the
Establishment of Project Controls

Matthew Oliver Ralp L. DIMAL and Louie P. BALICANTA, Philippines
Keywords: Azimuth, Astronomic Observations, GPS
SUMMARY
The azimuths obtained from astronomic observation and GPS survey were evaluated to show
significant differences. Astronomic observation was used in azimuth determination for old
surveys in the Philippines, but is gradually being replaced by the use of the GPS, because of
its rapid measuring capability and weather independency. However, GPS provides a different
type of azimuth as compared to astronomic observation. The relationship and precision of
both techniques were tested on various baselines located in different parts of the country.
Nine baselines in five cities from around the Philippine Archipelago were chosen as test sites
to compare the difference between the measured GPS azimuth and astronomic azimuth. Dual
frequency GPS receivers were used to measure geodetic azimuth using static GPS technique
while the astronomic observations made use of series of solar observations using a Wilde T2
theodolite. The differences in measured azimuth were analyzed and evaluated using the
standards of project control accuracy in the Philippines. The results derived from the two
methods yielded comparative difference of amounts less than 15 seconds in all test sites, the
minimum requirement of PE for tertiary project controls. This means that in the establishment
of project controls of at least tertiary control requirements, azimuth from GPS and astronomic
observations may be used or interchanged.
TS 6D Engineering Application of GNSS

Dimal, Matthew Oliver Ralp L. and Balicanta, Louie P.
Comparative Analysis of GPS Azimuth and Astronomic Azimuth for the Establishment of Project Controls
7th FIG Regional Conference
Spatial Data Serving People: Land Governance and the Environment Building the Capacity
Hanoi, Vietnam, 19-22 October 2009
Page 1 of 18
Comparative Analysis of GPS Azimuth and Derived Azimuth for the

Establishment of Project Controls
Matthew Oliver Ralp L. DIMAL and Louie P. BALICANTA, Philippines
1.
INTRODUCTION
Azimuth is one of the basic and important information in field surveying, which form the
basis for orienting different navigation systems (Davis, Foote, Anderson, & Mikhail, 1981).
For surveyors and engineers, the information provides important data required for all kinds of
land surveys from cadastral-property survey, mapping, and various construction and
development projects.
Prior to the advent of electronic and sophisticated surveying instruments, azimuths were
traditionally obtained from the observation of different celestial bodies. The most popular
celestial bodies being observed were the sun during the day, and Polaris or the North Star
during the night (Mueller, 1969). Other celestial bodies have also been used; however, the
unique characteristics of the sun and Polaris for places on the northern hemisphere made them
the more popular choice in astronomic observation. With astronomic observation, an optical
surveying instrument is utilized to observe the position of the celestial body, which may be
use to determine the lines direction called astronomic azimuth. This method was widely
used in the establishment of old Philippine geodetic control networks established through
triangulation and trilateration (Allman, 1991). However, in recent years, the use of
astronomic observations in surveying became less, due partly to the multitude of problems in
observing celestial objects, and due to the apparent shift of the industry towards GPS
(Lennartz-Johansen & Ellegaard, 2002).
Global Positioning System or GPS is a breakthrough in the field of surveying. This
technology is independent from the method of sighting, as compared to a theodolite or total
station, which are highly dependent on intervisibility between stations. Using information
transmitted form orbiting satellites, geographical information like coordinates and azimuth
may be derived by an observer using a GPS receiver. Modern geodetic control networks are
now using this technology.
In the Philippines, the National Mapping and Resource Information Authority (NAMRIA) has
used GPS technology in establishing control stations for the Philippine Reference System of
1992 or PRS92 (Department of Environment and Natural Resources, 1994). Since PRS92
uses an ellipsoidal model such as the Clarke Spheroid of 1866, the derived azimuth from the
use of GPS is geodetic azimuth. Survey grade GPS units are very expensive, yet the
technology is very popular among local surveyors because of its efficiency and of its ability to
give fast accurate results.
Since the astronomic observation-derived azimuth and GPS derived-geodetic azimuth are
from different systems and references, the question among surveyors arises as to the
equivalence or similarity of the two azimuths. The need for a comparative analysis between
the azimuth obtained from GPS and astronomic observation has risen for various reasons.
Page 2 of 18
Most survey information databases prior to the introduction of PRS92 are based from azimuth
obtained from astronomic observation. Confusion arises when GPS-derived information and
old survey records, like the bearing of tie line of a property, are compared. Some sectors in
the field of surveying in the Philippines are claiming that they cannot use GPS to connect or to
reproduce old surveys. There is also a need to integrate and provide conformity in use of
astronomic observation and GPS survey in the country, which may be addressed by
structuring guidelines and procedures to local surveyors for handling both types of azimuths.
Cadastral maps and approved survey plans prior to the implementation of PRS92 were based
from astronomic azimuth. However, due mainly to technical and financial limitations, most
local surveyors are still unable to use GPS technology and become highly reliant on the old
surveys not in PRS92, contributing to the systems slow and poor implementation
(Department of Environment and Natural Resources, 1994).
The research aims to provide the applicability and the resulting consequences of interchanging
the two types of azimuth in the conduct of establishing project controls for isolated surveys.
This will analyze if the use of the GPS significantly affects the measurements in direction of
property boundaries that were outlined using astronomic observations.
2.
THEORETICAL FRAMEWORK
2.1
Azimuths
By definition, the azimuth of a line is the direction given by the angle between the meridian
and the line measured in a clockwise direction either from the north or south branch of the
meridian (Anderson & Mikhail, 1998).
There are several types of azimuth, including astronomic, geodetic, and grid azimuth. The
astronomic azimuth often renamed as true azimuth is the direction given by the horizontal
angle between the astronomic meridian of the observer and the line measured in a clockwise
direction either from the north or south of the astronomic meridian. The geodetic azimuth
is the direction given by the horizontal angle between the ellipsoidal meridian of the observer
and the line measured in a clockwise direction from the north or south of the ellipsoidal
meridian. The grid azimuth is the direction given by the horizontal angle between the grid
meridian (a meridian made parallel to the central meridian of a plane rectangular system) of
the observer and the line measured in a clockwise direction from the north or south of the grid
meridian (Wolfgang, 1980).
The astronomic azimuth is related to the astronomic meridian, astronomic latitude and
astronomic longitude. All these information are obtained when using the geoid as the model
to represent the shape of the earth. The geoid is a physical model of the earth obtained from
gravity measurements and astronomic observations (Heiskanen, 1967). The astronomic
meridian plane is the plane containing the astronomic normal at a certain point and is parallel
to the instantaneous rotation of the earth (Mueller, 1969). The astronomic latitude ( ) is
the angle between the astronomic normal and the plane of the instantaneous equator (Mueller,
Page 3 of 18
1969). The astronomic longitude () is the angle between the designated zero meridian, the
Greenwich mean astronomic meridian, and that of the meridian of a certain point measured in
the plane of the instantaneous equator (Mueller 1969).
Figure 2.1: The Geoid and the Ellipsoid
The geodetic azimuth is related to the geodetic meridian, geodetic latitude and geodetic
longitude. With these parameters, the earth is represented in a mathematical model as an
ellipsoid. There are several ellipsoidal models including the World Geodetic System of 1984
(WGS84) used by the GPS, and the Clarke Spheroid of 1866 used in the Philippines.
The geodetic meridian plane is the plane containing the geodetic normal at a certain point
and is parallel to the rotational axis of the earth. The geodetic latitude ( ) is the angle
between the geodetic normal and the geodetic equator. The geodetic longitude ( ) is the
angle between the designated zero meridian, or the Greenwich meridian, and the meridian
containing a certain point measured in the plane of the geodetic equator. (Mueller, 1969)
As it is more convenient to do positioning on a rectangular XY coordinate system, an earth
model is projected on a plane that fits the location being mapped. This process is called map
projection (Wolfgang, 1980). There are different map projections, and in the Philippines the
Transverse Mercator projection is used. In doing the projection, a central meridian is being
chosen and is considered as the true meridian. All other meridians are made parallel to this
central meridian and are called grid meridians. As discussed, grid azimuths are referred
from the grid meridian containing the observer.
2.2
Astronomic Observation
The primary method of obtaining the astronomic azimuth used in this research is the Method
of Azimuth Determination at any Hour Angle or MADHA. The celestial sphere is used as the
primary model in astronomic observation, where the earth is made the center of the sphere and
other celestial objects are of the infinite distance from the earth. Position and direction of
celestial objects are defined by different celestial coordinate systems. MADHA is derived
Page 4 of 18
from the relationship of two celestial coordinate systems namely the hour angle system and
the horizon system.
The relationship of the parameters of the hour angle system and the horizon system can be
derived using spherical trigonometry or by using matrix transformation. The derived formula
to obtain the astronomic azimuth is:
Eq. 1
This same formula is also given in the Almanac for Geodetic Engineers published by the
Philippine Atmospheric Geophysical and Astronomical Services Administration. Equation 2
is an alternative formula which is also provided in the Almanac.
1/2
Where:
Eq. 2
(Eq. 2b)
In astronomic observation using MADHA, the observer measures the vertical angle and the
instant of observation (time) of a celestial body. The horizontal angle is also measured to
correspond to the subtended angle between the baseline and the celestial body measured
clockwise. The North Polar Distance is obtained from the annually published almanac.
is
the astronomic latitude which may be obtained through astronomic observation.

Systematic errors were also considered in the measurement and calculation of azimuth. Since
the methodology made use of the tangency method, or sighting the sun in the four limbs with
the telescope at direct and reverse, the effects of the pointing and collimation errors became
negligible. Corrections for index, diurnal aberration, parallactic displacement and astronomic
refraction were computed and applied in the measurements for the determination of azimuth.
Other natural errors associated to astronomic observations were not considered due to their
very small effect to the measured altitude of celestial objects.
The value of the positions latitude can be determined using a variety of methods including
the use of the geodetic almanac for the different provinces, and the use of a topographic map,
or by control connection from known geographic positions.
If more accurate results are required, astronomic latitude of a point shall be determined by the
Method of Latitude Determination by Observation of a Circumpolar Star at Upper
Culmination. With this method, the celestial body is exactly at the meridian of the observer.
The parameters azimuth and hour angle are therefore eliminated, thus simplifying the formula
to:

Eq.3
Page 5 of 18
2.3
GPS Observation
The basic principle of the GPS is the use of positioning satellites to determine the position of a
receiver located on the surface of the earth. There are more than twenty four NAVSTAR
satellites orbiting around the earth. At least four (4) satellites are needed to obtain a relatively
accurate receiver position similar to the principle of resection. Accuracy is measured in terms
of the Dilution of Precision. This dilution of precision has several levels; the highest is the
Geometric Dilution of Precision (GDOP) which considers the errors in the x, y, z and time.
The lower the DOP or GDOP the higher the accuracy of the obtained measurement. This is
similar to the concept of the strength of figure in triangulation networks. The accuracy of the
GPS measurement is further improved by performing differential GPS survey wherein two or
more survey grade GPS receivers simultaneously observe the same set of satellites. (Leick,
2004)
Relative or differential static GPS survey shall be used as the primary method in the
determination of the geodetic azimuth ( ). Two GPS receivers located on stations where
is needed shall be observed simultaneously. Appropriate processing software should obtain
the WGS84 geographic coordinates: geocentric latitude and longitude (
) of
both stations. Since these coordinates are based from the WGS84 datum, there is a need to
convert these to PRS92 horizontal datum having geographic coordinates: geodetic latitude and
longitude (
). These are the same geodetic latitude and geodetic longitude
based from a reference ellipsoid. With PRS92, the reference ellipsoid used is Clarke Spheroid
of 1866, which is the same ellipsoid used in the earlier Philippine datum.
To obtain the geodetic azimuth directly from the geodetic latitude and longitude of the two
points being observed, the Gauss Mid-latitude method (Rapp, 1979) can be used. The formula
to obtain the geodetic azimuth is expressed as:
-
where:
Eq. 4a
Eq.4b
Eq. 4c
Eq. 4d
(Prime Vertical Radius of Curvature)
Eq. 4e
(Meridional Radius of Curvature)
Eq. 4f

Page 6 of 18
Eq. 4g
Eq. 4h
Eq. 4i
Another way of solving the geodetic azimuth is by obtaining the grid coordinates of the two
stations, based on the local horizontal datum. In the Philippines the local horizontal datum is
in the Philippine Reference System of 1992, and the grid coordinates is in terms of the
Philippine Plane Coordinate System-Transverse Mercator (PPCS-TM). The Bureau of Lands
Technical Bulletin no. 26 provides a solution for this conversion. Most GPS processing
software also provides a conversion command. The input of the conversion is the geographic
coordinates
and , and the output is in terms of grid coordinates northing (N) and Easting
(E). A grid azimuth can be computed by the expression:

=
Where:E
Eq. 5a
Eq.5b
N =
Eq 5c
The grid azimuth can be transformed to the geodetic azimuth by applying the t-T correction
and the convergence correction. Technical Bulletin no. 26 (TB 26) provides the solution for tT correction. It is also stated in TB26 that for baseline distances less than 1,500 meters the tT correction is negligible. The convergence correction is expressed as:
Eq. 5d
Where c = convergence correction in seconds, = difference in longitude between two points in seconds
= geodetic latitude of the first point
and
The geodetic azimuth is therefore expressed as:

=
(negligible for short distances)

Eq. 5e
Page 7 of 18
2.4 Relationship between the Astronomic and Geodetic Azimuths

From the definition of the astronomic azimuth, geodetic azimuth, and grid azimuth, each form
is similar to one another, but is neither the same nor equal. As to the comparison between the
astronomic azimuth and geodetic azimuth, the difference lies on the meridian (astronomic
meridian versus ellipsoidal meridian) and the normal (vertical reference affected by gravity
versus normal to the ellipsoid). The deflection of the vertical is the difference between the
astronomic meridian and the geodetic/ellipsoidal meridian (Heiskanen, 1967). The deflection
of the vertical may also be viewed as the angle between the directions of the plumb line and
the ellipsoidal normal at the same point. It has two components and which depend
directly on the shape of the geoid in the region. These are expressed as:
northsouthcomponent
eastwestcomponent
Eq.6
Eq.7
The discrepancy between the astronomic coordinates and geodetic coordinates vary from
around 0 to 10 in moderately plain areas and from 10 to 30 in a rugged topography
(Bhattacharji, 1958). It is for this same reason that the astronomic azimuth and the geodetic
azimuth also vary from one another. The difference between the two azimuths is expressed in
the form of the Laplace correction (Leick, 2004) as shown by Equation 11a.
Eq. 8a
cos
sin
Where: A = astronomic azimuth,

angle
Eq.8b
= geodetic azimuth, = ellipsoidal zenith angle, = observed zenith
Ideally to convert the geodetic azimuth to astronomic/true azimuth, it is therefore necessary to

obtain the two components of the deflection of the vertical which in turn is dependent on both
the astronomic and ellipsoidal/geodetic positions ( , and , respectively).
3.
METHODOLOGY
Five key cities distributed throughout the country were used as study areas for the research.
These were Laoag City in Region I, Baguio City in the Cordillera Autonomous Region,
Quezon City in the National Capital Region, Puerto Princesa City in Region IV-B, and Cebu
Page 8 of 18
City in Region VI. The test sites were chosen based on the availability of existing controls,
scattered distribution on a national scale, security of the survey personnel, and accessibility to
these sites. PRS92 stations were obtained from NAMRIA, and were selected as points of the
baseline where the astronomic observation and differential GPS survey were conducted.
From the obtained WGS84/PRS92 geographic coordinates and the sketches obtained from
NAMRIA, the group was able to recover control points from the study areas. Additional
requirements on stationing included the following criteria:
1. Intervisibility of the two points
2. Applicability of stations for astronomical observation purposes; and
3. Clearance from obstruction such as buildings, electrical posts and trees of around 15o
from the horizon.
Astronomic observation was conducted to obtain the astronomic/true azimuth of the baseline
located in the study areas. Solar observations were conducted for azimuth and latitude
determination in these baselines. In the azimuth determination using solar observations, the
tangency method was conducted with multiple repetitions in the morning and in the afternoon
over several days. The latitude determination using solar observations was conducted during
the suns upper culmination in midday with at least two sets of observation in each test site.
GPS surveys were also conducted on the same baselines, using the rapid static technique in
GPS. The receivers were set with 15 seconds logging interval, 15 elevation mask, WGS84
horizontal datum, and with a maximum GDOP of 3. The Static GPS survey should be done
on both in the morning and in the afternoon. Two dual frequency GPS receivers were used to
observe at least 4 similar satellites simultaneously for at least one hour.
After set-up, antenna height should be measured and recorded on the GPS observation field
sheet. Height, station name, serial number, antenna model, start and stop time, number of
available satellites per 15 minutes and GDOP should also be recorded. After each fieldwork,
the data were downloaded immediately and were converted to RINEX format.
The baseline processing was done using available commercial GPS processing software. For
the observation to be considered acceptable, the baseline must have been considered fixed
by the processing software, before the corrective measures were applied. The resulting
WGS84 geographic coordinates were then transformed to PRS92 geographic coordinates.
The geodetic coordinates of the baseline points were then converted to PRS92 grid
coordinates using the processing software, which then were used in the computations of the
grid azimuth. Using the convergency correction, the geodetic azimuth was then computed
based from the calculated grid azimuth. The resulting astronomic azimuth and geodetic
azimuth of the baseline were tabulated and analyzed.
The resulting astronomic azimuths were compared with the other sets within a series of
observations, and other series of the same baseline observation. The rejection of a set of
observation was in accordance to section 116 of DAO 98-12, stating the astronomic
observation should not exceed 15, 10 and 5 respectively for a tertiary, secondary and
primary project control survey. All sets within the said limits were considered in the
computation of the mean astronomic azimuth which were then compared to the geodetic
azimuth of the same baseline.
Section 146 of DAO 98-12 states the formula for the computation of the probable error of the
astronomic azimuth from two sets and is expressed as:
Page 9 of 18
ProbableError 0.33725D
Eq.9
Where: D= the difference in azimuth between the first and second sets expressed in seconds.
The resulting value was compared to the required P.E. stated in section 116. Any set
exceeding the prescribed P.E was eliminated. Section 147 on the other hand states that the
probable error of the mean azimuth from several series of observations shall be determined
using the formula:
ProbableError 0.6745
v2/ n n1
1/2
Eq.10
Where: v = are the residuals and is equal to the difference between the mean and an observation
n = the number of series of observation
Similar to Equation 11, the computed P.E. was compared to the maximum allowable P.E.
stated in Section 116, with all values exceeding the allowable amount rejected. All computed
astronomic observations passing the rejection procedure were then be used in the computation
of the mean astronomic/true azimuth simply expressed as:
Eq. 11
Where: An = computed astronomic azimuth for a set n
n = number of sets
The mean astronomic azimuths were then compared to the computed geodetic azimuth and
were evaluated using Section 116 of DAO 98-12 and appropriate formulas.
4.
RESULTS
Actual GPS observation and astronomic observation were done on the following date: Cebu
City - August 15 to 19, 2008; Laoag City - September 19 to 21, 2008; Puerto Princesa October 11 to 15, 2008; Quezon City October 24, 2008; and Baguio City October 25 to
27, 2008. The geodetic coordinates of the control points of the baselines are tabulated in
Table 4.1. The values obtained from NAMRIA were used on pre-established control points
from the said agency. The rest of the control points which were established for this research
were referenced from the NAMRIA established control points and were derived after GPS
post-processing.

Page 10 of 18
Table 4.1 Geodetic Coordinates of the Baseline Control Points
4.1
Azimuth Results
Solar observations for azimuth determination were employed for the nine baselines, with the
exception of MMA39-MMA39A which utilized stellar observations for more accurate
azimuth results. Geodetic latitudes from GPS observation was used in the computation of the
astronomic azimuths. Table 4.2 shows the tabulation of the mean astronomic azimuth of each
baseline, standard deviation and computed probable error.
Table 4.1: Mean astronomic azimuth obtained using the geodetic latitude of the occupied control point
From the computation, the maximum and minimum computed probable error is 8.13 seconds
and 0.25 seconds for baseline ILN7-ILN7B and ILN6-ILN6B respectively. Two sets of
computations were done to obtain the geodetic azimuths from the same GPS observations.
Both sets were processed using the WGS84 coordinates of the control points from NAMRIA
Page 11 of 18
similar. One set of geodetic azimuths were obtained using Equation 4a (Method A) and the
second set using Equation 5e (Method B). As expected there were no significant differences
between the geodetic azimuth computations between Method A and Method B. Table 4.3
shows the comparison of the computations.
Table 4.3: Geodetic Azimuth from Equation 4a (Method A) vs. Geodetic Azimuth from Equation 5e
The mean astronomic azimuths shown in Table 4.2 and the geodetic azimuths shown in Table
4.3 are then compared to each other. The difference between these two azimuths and probable
error computation were obtained. The probable errors were using the criteria for project
control establishment. Table 4.4 shows the result of the computations.
Table 4.4: Astronomic Azimuth (computed using geodetic latitude) vs. Geodetic Azimuth
The difference between the azimuths from astronomic observation and GPS observation
ranges from 7 seconds to 40 seconds. Computed probable errors of the baselines are within
the minimum tertiary accuracy standard of 15 seconds. The baseline MMA39-MMA39A has
the least difference (7 seconds) and probable error (2 seconds). The astronomic azimuth of
Page 12 of 18
this baseline came from Polaris observation and the astronomic azimuths were obtained using
the geodetic latitude from GPS observations.
4.2
Latitude versus Azimuth
Since the geodetic latitude was used in the computation of the astronomic azimuth. The
researchers felt that there is a need to assess the effects of the inputted values of latitude to the
computed astronomic azimuth using the astronomic computation formula. This was done by
having the independent variables constant except for latitude whose values ranges from 2
minutes below to 2 minutes above the inputted latitude and an interval of 1 second. This was
done on baselines having morning and afternoon observations. Figure 4.5 shows the results of
the computation in graphical format.
Laoag City: ILN6-ILN6B
Cebu City: CBU76-CBU76B
Puerto Princesa: PWL35B-PWL35
Baguio: BPS01-BPS02
Figure 4.5: Latitude versus Azimuth
The graphs in Figure 4.5 show a linear relationship between the latitude and the azimuth for
values near the latitude (the geodetic latitude) used in the computation of the astronomic
azimuth. From the linear curve fit of each baseline, the slopes were computed. The geodetic
latitude ( ) used in the computations for morning and afternoon observations produced two
different astronomic azimuths.
5.
ANALYSIS OF RESULTS
5.1
Baseline and Azimuth Relationship
The baseline lengths for all the samples range from 98 to 300 meters. Table 4.2 and Table 4.3
show both the length and the probable error for each baseline. The probable errors shown in
the table is the researchs criteria for accuracy or precision. From the said tables, there is no
Page 13 of 18
indication that the length of the baseline affect the value of the probable error as longer
baseline length that not always have smaller value for the computed probable error when
compared to shorter baselines. The result of Chang and Tsais research (2006) also shows
similar observation. In their research three baselines of distance: 300 meters, 6,900 meters
and 18,300 meters were used as samples. It was also observed that no increase in accuracy
was obtained for longer baselines. The precision of measurement for both astronomic and
GPS survey does not depend on the length of the baselines.
5.2
Analysis of the Latitude
The graphs in Figure 4.5 show that the latitude and azimuth baselines exhibited a linear
relationship for latitude values ranging from 2 minutes above and below the inputted latitude.
All the graphs show a similar pattern that is: an increase in latitude constitutes an increase in
azimuth for morning observation and an increase in latitude for afternoon observation
constitute a decrease in azimuth. There is also a one is to one relationship between the two as
one value of latitude provides a single resulting azimuth within the given range of values.
The initial latitudes used in the solar computation are the geodetic latitudes obtained from
GPS observations. These latitudes were used in the computation of astronomic azimuths from
morning and afternoon solar observation. The resulting azimuth obtained shows a difference
between the computed azimuths. The differences are minimal and within the criteria as shown
in Table 4.2 but these are differences nonetheless. Figure 4.5 shows the two linear curves that
represent morning and afternoon data intersecting at a specific point. This intersection point, if
one would interpret the graphs, describe a single value of latitude for both morning and
afternoon observation and would result to a single value of resulting azimuth. Given that the
other parameters expressed in Equation 2 (altitude, North Polar Distance and horizontal angle)
which are being observed using the geoid as the physical model, one could assume that the
latitude obtained from the intersection is the astronomic latitude. This hypothesis should be
validated in future researches.
5.3
Comparison and Analysis between Geodetic and Astronomic Azimuths
5.3.1. Astronomic Azimuths

The result of the astronomic observation from Table 4.2 shows that that standard deviation
ranges from 7.18 to 30.24 seconds. The computed probable errors passed the criteria for at
least tertiary project control accuracy. It was expected that results from Polaris observation
would yield a probable error passing the requirement for primary project control and this was
shown by the result of MMA39-MMA39A baseline. However, it is important to note that
several baselines that were observed using solar observation also passed the primary project
control accuracy requirements.
5.3.2
Geodetic Azimuths
Geodetic azimuths from GPS observation was obtained by using the Gauss Mid-latitude
method expressed by Equation 4a and by converting the grid azimuth to geodetic azimuth
expressed in Equation 5e. The resulting geodetic azimuths using the two methods did not
vary.
Page 14 of 18
5.3.3. Astronomic Azimuths versus Geodetic Azimuths

It is expected that the astronomic azimuth and the geodetic azimuths would have a difference.
The discussion from Equation 8a indicates that this difference is inherent because of the effect
of the deflection of the vertical.
Table 4.4 shows that the difference between the two azimuths. Two baselines fall within the
tertiary project control accuracy, three baselines within secondary and four baselines within
the primary. Precision in the astronomic observation is not indicative to the closeness of the
astronomic azimuth to GPS azimuth as CBU76-CBU76B which has a probable error of 2.19
seconds in the astronomic computation yields a difference within tertiary when compared to
the geodetic azimuth.
The obtained difference between the astronomic and geodetic azimuths is larger compared to
the results obtained from a related research that showed a difference better than 1 second
(Chang & Tsai, 2006). The disparity can be attributed to the type of equipment and the
baselines used in the two researches. Chang and Tsai used established baselines whose
astronomic azimuths were observed and measured periodically using high precision
gyroscope. This research used baselines established measured only during the scheduled time
of the fieldwork and the equipment used was a Wild T2 theodolite.
However, the result showed that for project control network establishment, the resulting
azimuth obtained using GPS survey or traditional astronomic observation can be used
interchangeably.
6.
CONCLUSIONS
This research aimed to determine the effect and significance of interchanging astronomic
azimuth from solar observation with geodetic azimuth measured from the GPS. Variables
that may cause variations in the computed azimuth were also evaluated to show any
consequence in the computations. The variability in the length of the baselines for project
controls was checked. The results showed the length to have no detectable effect on the
precision of the observations on either method. The computed and measured latitude used in
the computations from the astronomic observations was found to have a significant effect.
Any error in the derived or assumed latitude of the place of observation will cause a linear
increase of inaccuracy of the computed azimuth. An alternative method to minimize the error
introduced by assumed latitude had been suggested with the use of a pair of solar observations
conducted one in the morning and one in the afternoon. Preliminary results from this research
showed positive reduction of errors, though this conclusion needs further validation.
In assessing the applicability of solar observations in contemporary usage, the research also
found some interesting results. The Philippine geodetic engineers manual, DAO 98-12, had
only considered solar observation to be used in tertiary project control surveys. The probable
errors computed from the results of the fieldwork have shown that the azimuths from solar
observations may consistently attain primary project control accuracy with carefully and
systematically executed fieldwork and application of appropriate corrections. The research
also noted some factors that may minimize the usability of solar observations in surveying
applications, which included weather conditions, forest canopy, and urban built-up. To
counter some of these concerns, a revised manual in conducting solar observations had been
forwarded to the authorities for evaluation and comments.
Page 15 of 18
In the determination of geodetic azimuth from GPS observation, a comparison of results,

between those of logged coordinates with unreferenced solution and those of predetermined
positions with referenced solutions, had been analyzed. It was observed that no significant
effect in measured geodetic azimuth exists, and that unreferenced solutions may be used in the
determination of geodetic azimuth with no apparent difference from referenced solutions. But
for position or grid coordinates, GPS derived positions from the referenced and from the
unreferenced solutions had very different outcome.
In the computation of geodetic azimuth, a difference of 2 was observed on one baseline, 1
for two baselines, and the rest with no apparent difference. For project controls of primary,
secondary and tertiary precision requirements the use of the geodetic latitude in the
determination of the astronomic azimuth was determined to be sufficient at least for areas
with flat topography. The geodetic azimuths and astronomic azimuths as shown in the study
proved to be within the criteria of project controls. Results from related studies from other
countries have also yielded similar results (Chang & Tsai, 2006; Yilmaz & Karaali, 2006).
This concludes that for practical purposes or at least for project controls, the two azimuths can
be used interchangeably.
An analysis on the methodologies of astronomic observation and GPS survey showed several
advantages and disadvantages of each of the methods:
a. The disadvantage of astronomic observation lies on its too much dependency on
weather condition. A clear sky in daytime observation for solar observation and clear
sky in night time for Polaris observation is needed. This is true if the parameter
needed is astronomic latitude, astronomic longitude or astronomic azimuth.
Consideration of the terrestrial lighting should be considered when performing
astronomic observation during night time. Minimal lighting within the vicinity is
required so that stars and star pattern would be visible.
b. Expertise on the use of the instrument is also a major consideration and since the
astronomic observation uses an optical instrument where majority of the task is
performed by the instrument man and with limited time consideration this again
becomes the weakness of astronomic observation compared to GPS observation.
c. GPS observation on the other hand provides a faster way of obtaining azimuth. The
use of GPS seems more advantageous being weather independent and with minimal
user intervention.
7.
RECOMMENDATIONS
More local studies analyzing the consequences of the use of the GPS in property surveys
should be conducted to increase local understanding and awareness of the technology, and its
proper use in the practice of surveying. Other factors that were not considered but may
contribute to the difference the derived azimuths, like the ruggedness of terrain, variability in
geologic attributes and proximity of the location to the coastlines, should also be analyzed.
Its application in geodetic surveys should also be checked. Due to the extensiveness and
complexities of requirements needed in geodetic applications, further research should be
conducted to ensure the seamless transition of the national positional framework from mainly
using astronomic observations towards using GPS techniques. The determination of the
Laplace correction from an extensive research would be required in fulfilling this requirement.
Page 16 of 18
The proposed field procedures inscribed in the generated revised manual should be reviewed
and evaluated further, to update current practices in land surveying. New techniques in solar
observations and GPS surveys as consequence of the findings in this research should also be
validated by further studies.

Page 17 of 18
8.
REFERENCES
Allman, J. (1991). An Analysis of the Primary Triangulation Network of the Philippines
Internal Report. Quezon City: SAGRIC Intl Pty Ltd.
Anderson, J., & Mikhail, E. (1998). Surveying Theory and Practice Seventh Edition. New
York: McGraw-Hill Companies, Inc.
Bhattacharji, J. C. (1958). Comments on `Azimuth Determination without Circle
Readings' by R. K. C. Johns. Bulletin Godsique , 32 (4), 52-55.
Chang, C., & Tsai, W. (2006). Evaluation of a GPS-Based Approach for Rapid and
Precise Determination of Geodetic/Astronomical Azimuth. Journal of Surveying
Engineering , 149-154.
Davis, R., Foote, F., Anderson, J., & Mikhail, E. (1981). Surveying Theory and Practice
(6th ed.). New York: McGraw-Hill International.
Department of Environment and Natural Resources. (1994). Implementing Guidelines in
the Adoption of the Philippine Reference System 1992 for Land Surveys in the
Philippines. DENR Administrative Order.
Heiskanen, W. H. (1967). Physical Geodesy. San Francisco: WH Freeman and Company.
Leick, A. (2004). GPS Satellite Surveying. John Wiley and Sons.
Lennartz-Johansen, H., & Ellegaard, S. (2002). Analyzing Europe's Largest Suspension
Bridge. FIG XXII International Congress. Washington D.C.
Mueller, I. I. (1969). Spherical and Practical Astronomy as Applied to Geodesy. New
York, New York, United States of America: Frederick Ungar Publishing Co.
Rapp, R. H. (1979). Geometric Geodesy. Columbus, Ohio: Ohio State University.
Wolfgang, T. (1980). Geodesy: An Introduction. Berlin: Walter de Gruyter & Co.
Yilmaz, N., & Karaali, C. (2006). Positioning with Astronomic and Geodetic Method.
XXIII FIG Congrees. Munich, Germany.
CONTACTS
Mr. Matthew Oliver Ralp L. Dimal
University of the Philippines
Dept. of Geodetic Engineering, University of the Philippines, Diliman
Quezon City
PHILIPPINES
Tel. + 632 981 8500 3126
Fax + 632 920 8924
Email: mldimal@up.edu.ph
Web site: http://dge.upd.edu.ph/
Mr. Louie P. Balicanta
University of the Philippines
Dept. of Geodetic Engineering, University of the Philippines, Diliman
Quezon City
PHILIPPINES
Tel. + 632 981 8500 3126
Fax + 632 920 8924
Email: lpbalicanta@up.edu.ph
Web site: http://dge.upd.edu.ph/
Page 18 of 18

Azimuth GPS

Uploaded by

Copyright:

Available Formats

Azimuth GPS

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Azimuth GPS

Uploaded by

Copyright:

Available Formats

Comparative Analysis of GPS Azimuth and Derived Azimuth for the

Establishment of Project Controls

TS 6D Engineering Application of GNSS

Comparative Analysis of GPS Azimuth and Derived Azimuth for the

Figure 2.1: The Geoid and the Ellipsoid

the astronomic latitude which may be obtained through astronomic observation.

TS 6D Engineering Application of GNSS

(Meridional Radius of Curvature)

TS 6D Engineering Application of GNSS

(E). A grid azimuth can be computed by the expression:

The geodetic azimuth is therefore expressed as:

(negligible for short distances)

TS 6D Engineering Application of GNSS

2.4 Relationship between the Astronomic and Geodetic Azimuths

Where: A = astronomic azimuth,

= geodetic azimuth, = ellipsoidal zenith angle, = observed zenith

Ideally to convert the geodetic azimuth to astronomic/true azimuth, it is therefore necessary to

TS 6D Engineering Application of GNSS

Table 4.1 Geodetic Coordinates of the Baseline Control Points

Latitude versus Azimuth

Cebu City: CBU76-CBU76B

Puerto Princesa: PWL35B-PWL35

Figure 4.5: Latitude versus Azimuth

Baseline and Azimuth Relationship

Analysis of the Latitude

Comparison and Analysis between Geodetic and Astronomic Azimuths

5.3.1. Astronomic Azimuths

5.3.3. Astronomic Azimuths versus Geodetic Azimuths

In the determination of geodetic azimuth from GPS observation, a comparison of results,

TS 6D Engineering Application of GNSS

You might also like