2.

Measurements & Errors

Dr Charisse Griffith-Charles
Department of Geomatics Engineering & Land Management
At the end of this lesson the student will be able
to define, describe, apply, and evaluate:
 Measurement

 Difference between Precision and Accuracy

 Uncertainty

 Sources of measurement errors

 Classification of errors

2
 1. Bell, S. 1999. A Beginner's Guide to
Uncertainty of Measurement. National Physical
Laboratory. UK

 2. Schofield, W., and Mark Breach. 2007.

Engineering surveying. Oxford; Burlington, MA:
Butterworth-Heinemann.

Kavanagh, B. F. 2010. Surveying with

Construction Applications. Pearson
Chapter 3 – Distance Measurement

3
 Measurement tells us about a property of
something
 Measurement gives a number to that property
 Measurements are done using instruments
 Two elements: a number and units
 Counting is usually not considered a
measurement
 Tests are not measurements
 Measurements may be used as part of the
testing process

4
 the assignation of a numerical value to
represent that quantity measured
 comparison with some standard unit or
quantity

5
 2 types of measurements – Direct & Indirect
 Direct measurements involve applying the
instrumentation directly to the phenomenon to
be measured e.g. applying a tape to a line
 Indirect observations take place when the
quantity required cannot be directly observed.
The answer is obtained by determining a
relationship between the required quantity and
some other values that can be measured.

6
 Define the units of measurement
 Construct a mathematical model to
represent and simplify the reality of the
components of measurement
 Adopt a procedure for the observation

Observation Mathematical Model Measurement

INPUT PROCESS OUTPUT

7
All measurements generally involve the
combination of:
 Human skill (variable)
 Instrumentation/ mechanical equipment
 Natural conditions (variable)
 All measurements (observations) are never
exact i.e. they always contain errors
(uncertainties)
 Repeated measurements of the same
quantity may not give the same values
WHICH ONE IS CORRECT? 8
Dr. C. Griffith-Charles 9
Which is more Accurate: A, B, or C?
Which is more Precise: A, B, or C?

10
• Accuracy and Precision are different
• More scatter means less precision
• More scatter means less likely repeatable
• Precision only accurate if no biases exist
• Accuracy is defined within specific bounds
• Position fixes represent more probable
values, as supported by statistics

11
 Uncertainty of measurements is the doubt
that exists about the result of any
measurement.
 It tells us something about the quality of
the measurement.
 This is an expression of how big the margin
of doubt and how bad is the doubt, as
expressed by uncertainty interval and
confidence level

12
e = M-T
Error = measured value – true value

Dr. C. Griffith-Charles 13
The accuracy of a measurement is the nearness
of M to T, so a small error means a high
accuracy and vice versa

Dr. C. Griffith-Charles 14
 the spread or dispersion of measured
values of a quantity.
◦ Several values grouped closely together are a
more precise set of measurements than one with
a broader spread of values

Dr. C. Griffith-Charles 15
 The evaluation of the measurement then
leads to:
• The MPV of the desired unknown quantity
• The determination of a figure or measure
of the accuracy of a single measurement
• The estimation of the accuracy of the MPV
and its confidence range

C. Griffith-Charles 16
One of the most commonly used estimates for
the single MPV is the arithmetic mean :

x i
x= i =1

n
C. Griffith-Charles 17
After obtaining the sample mean, the residuals
that remain are ν as follows:

 i = xi − x

C. Griffith-Charles 18
The standard deviation is given by:

 (x − x)
i
2

= i =1
n

C. Griffith-Charles 19
The standard error of a single measurement
(used for rejection criteria):

 (x − x)
i
2

sex = i =1
n −1

C. Griffith-Charles 20
Rejection criteria
 68% of observations fall within 1σ
 95% of observations fall within 1.96 (2) σ
 99% of observations fall within 2.96 (3) σ

C. Griffith-Charles 21
The standard error of the mean is (used for
confidence intervals):

 (x − x)i
2
sex
sex = i =1
=
n(n − 1) n

C. Griffith-Charles 22
Confidence intervals
 68% confidence the mean falls within 1σ
 95% confidence the mean falls within 1.96 (2)
σ
 99% confidence the mean falls within 2.96 (3)
σ

C. Griffith-Charles 23
 Errors sources fall into three
categories:
◦ Natural Errors
◦ Instrumental Errors
◦ Personal Errors

24
‘Measure thrice, cut once’. You can reduce the risk of
making a mistake by checking the measurement a
second or third time.

25
 Classifications of Errors

◦ Mistakes/Blunders/Gross Errors

◦ Systematic Errors (Biases)

◦ Random Errors

Which ones are true errors and which represent uncertainty?

26
Types of Errors
 BLUNDER

◦ Called a mistake or gross error

◦ Caused by misunderstanding the
problem/equipment, carelessness, fatigue,
miscommunication etc
◦ Very large – generally easy to detect
◦ Identified by repeating observations

27
Types of Errors
 SYSTEMATIC ERROR

◦ Errors which have a cumulative or constant

effect i.e. error occurs in the exact same
way for the same measurement conditions
◦ Also known as biases or cumulative errors
◦ Conditions producing these errors follow
physical laws which can be modelled
mathematically
◦ Eliminated using mathematical models

28
Types of Errors
 RANDOM ERROR
◦ All the remaining errors after gross and
systematic errors are eliminated
◦ Very small but random in nature i.e. any
size, any sign
◦ Caused by factors beyond observer’s control
◦ Obey laws of probability – use probability
theory to deal with their presence in
observations
◦ Present in ALL survey observations

29
At the end of this lesson the student will be able
to define, describe, apply, and evaluate:
 Measurement

 Difference between Precision and Accuracy

 Uncertainty

 Sources of measurement errors

 Classification of errors

30
 Linear Measurements

Dr Charisse Griffith-Charles
Department of Geomatics Engineering & Land Management
At the end of this lesson the student will be able to
define, describe, apply, and evaluate:
 Distance Measurement
◦ Methodology, Equipment, Observations, Mathematics,
Errors

 Taping

 Stadia

 Electronic Distance Meter (EDM)

32
33
34
 Horizontal
distance
Vertical B
distance Slope
distance
Terrain
distance
A

35
 Pacing is very useful especially in
reconnaissance and estimating distances
 The pace can be determined by counting
across a known distance
 Pedometer can also be used
 A pace is defined by one natural step
 A stride is considered to be two paces
 Generates terrain distances
 Length changes going up or downhill
 Accuracy of ~ 1:50 – 1:100

36
(a) Linen tape, (b) fibreglass, (c) steel, (d) steel
band, (e) spring balance
37
38
39
40
2:4.1
41
29.12

29.12

42
 Accuracy of Steel Tapes
◦ With all precautions accuracy can be as good
1:10000 (3mm for a distance of 30m)

 Accuracy of Fibreglass Tapes

◦ More prone to stretching with lower accuracy at
1:1000 (30mm for a distance of 30m)

43
 Zero Error

44
 Misreading of Tape
◦ Reduced:
 by reading at different zeros
 redundant readings
 measuring both directions

 Alignment errors of bays is generally

insignificant
 {an error of 15cm will only result in an error of 1mm
per 30m}

45
 Slope Measurements and Slope Corrections
 Standardisation
 Tension
 Temperature Variations
 Sag

 D = L- slope ± standardisation ± tension ±

temperature ± sag
◦ L is the measured slope distances, D is the
Horizontal Distance
46
Slope Correction
L'
H


L''
Corrected length L'' = L' cos

For Slope Correction

Ch = L' – L''

47
Standard Correction
Corrected length L' = L(l'/l)
Example 54.395m = 54.375 (30.011/30)
or the standardisation correction Cs can be
computed from:
Cs = L((l' – l)/l)
Example 0.019m = 54.375 ((30.011-30)/30)

Where:
l = standard length
l' = calibrated length
L = observed length
L' = corrected length
48
Temperature Correction

Ct = L'' (T0 – T)
Where:
L'' is the measured length and
α is the coefficient of thermal expansion
(0.00000645/1F or 0.0000116/1C for
steel)
T = standard temperature
To = observed temperature

49
Tension Correction

where P is the applied tension in lbs or kgs

(kg x 9.807 = N)
P0 is the standardisation tension in lbs or kgs
(units consistent)
L is the measured length
a is the cross-sectional area of the tape in in2 or
cm2
E is the modulus of elasticity of steel in lb/in2 or
kg/cm2 (E = 2,000,000 kg/cm2)
50
 The observed distance
from A to B was
318.00m using a tape
with cross sectional
area of 0.05cm,
standardised at a
tension of 5.5kg. The
tension applied during
the measurement was
9.46kg. What is the
correction and the  Correction = 0.013
corrected distance? (E  Corrected distance = 318.013
= 2,000,000 kg/cm2)


51
Sag (Catenary) Correction

where
w is the weight per unit length of the tape in
lb/ft or kg/m
W is the total weight of the tape between the
supports in lb or kg (= wL)
L is the distance between the supports
P is the tension applied in lb or kg

52
 A 30m length is
measured in one
catenary with a
30m tape with a
weight of 0.91kg.
A tension of 8kg is
applied. What is  Correction = 0.0162m
the correction and  Corrected distance =29.984m
the corrected
distance?

53
➢ There are two main techniques employed
for indirect distance measurement;
optical distance measurement (ODM),
and electronic distance measurement
(EDM)
➢ There are several different methods
involved in ODM, but main relevant ones
are stadia and subtense measurements

54
55
ODM
➢ Optical distance measurement is based on
the principles of the parallactic triangle,
where the distance d is derived from the
relationships between the parallactic angle 
of the triangle and it’s base s

56
 d = (s/2) cot(/2)

s
 d
➢ To obtain a value for d, one parameter (either
s or ) is held constant and the other is
measured.
➢ For stadia tacheometry, the base is measured
and the parallactic angle held constant, and
for subtense tacheometry the parallactic angle
is measured and the base held constant.

57
Stadia Tacheometry
➢ The term ‘tacheometry’ means rapid or fast
measurement
➢ the angle is kept fixed while the base is
measured
➢ This is achieved using two supplementary
horizontal lines (stadia) placed at equal
distances above and below the central
horizontal line in the telescope of an
instrument

58
Stadia Tacheometry
These lines serve both purposes
➢ Since they are fixed in the telescope, they
form a fixed angle with the optical centre
of the instrument, and
➢ Secondly they provide the lines for the
measurement of the base.

59
Since the parallactic angle  is constant,
the equation reduces to
d = sk
60
61
Stadia Tacheometry
➢ For most modern instruments, the
relationship between the stadia and the
parallactic angle is designed such that the
multiplication constant k = 100, therefore

d = 100s

62
Foundation of Tacheometry Surveying
63
D/S = d/s

Reticules in Theodolite and Levels d/s = 100

D = 100 S
64
Slope distance =100S cosα
Horizontal distance = 100S cos2α
Vertical distance = 100S cosα sinα = 100S (1/2 sin2α)
(Remember - sin 2A = 2 sin A cos A)
65
 Errors with angle observations (theodolite)
 Errors in reading the staff
 Errors in holding the staff vertical
 Uncertainty: Magnitude of ambiguity increases with
the distance between staff and instrument
 This system is sufficiently precise for locating
topographic details such as rivers, bridges,
buildings, and roads when an accuracy of 1/400 is
acceptable.

66
67
68
69
D = (nλ + d )/2
λ=v/f
D is the length of the line
n is the integral number of wavelengths in the double path of the light
λ is the wavelength of the modulated beam
d is the distance representing the phase difference
70
H = S cos(Va)

D = S sin (Va)

Va represents Vertical Angle

71
 Error Sources
◦ Instrument Errors
◦ Scale Errors
◦ Prism Constant / Zero Error
◦ Curvature and Refraction

* - raw observations can be

corrected after, before
 Precautions: reducing
◦ Prism Constants need to be entered correctly *
◦ Instrument set-up parameters need to be correctly entered *
◦ For long distances parameters need to be entered correctly *
◦ Instrument needs to be periodically checked against baseline
◦ Calibrated as determined from checks or required by law

72
Method Accuracy Usage

Pacing 1/50 to 1/200 Reconnaissance

Odometer 1/200 Reconnaissance

Taping 1/1,000 to 1/10,000 Land surveys

EDM ±0.04 to 1/300,000 All types of Surveying

73
At the end of this lesson the student will be able to
define, describe, apply, and evaluate:
 Distance Measurement
◦ Methodology, Equipment, Observations, Mathematics,
Errors

 Taping

 Stadia

 Electronic Distance Meter (EDM)

74
The End!!!!!

75
 Linear Measurements

Dr Charisse Griffith-Charles
Department of Geomatics Engineering & Land Management