ICID 21st International Congress on Irrigation and Drainage,

15-23
ICID 21stOctober 2011,
Congress, Tehran,
Tehran, Iran
October 2011 R.56.5.35

Improved management and minimising

uncertainty - discharge measurement
solutions for irrigation channels in Iran
AMELIORATION DE LA GESTION ET DIMINUTION DE
LINCERTITUDE SOLUTIONS DE MESURE DE DEBIT
POUR LES CANAUX DIRRIGATION EN IRAN

Stewart Child1, Martin Dibley1 and Mohammad Hamzehdoost1

ABSTRACT
The accurate measurement of flow in irrigation network is of fundamental importance to
the assessment and optimum management of Irans irrigation water requirements. The
Government of Iran through its Department of Water Resources Management (WRM), has
embarked on a programme to improve the monitoring of flow in irrigation channels throughout
the country. The Abtin Co, with technical support from Hydro-Logic Ltd. (UK), has been
commissioned to assist with the design and implementation of flow measurement solutions.
This paper outlines the flow measurement methods used, including technical background,
the issues encountered with their implementation and an assessment of uncertainties in the
computed discharge values.

In view of the range and variety of channel sizes and site conditions no single solution was
applicable for all sites, even though the importance of standardisation was recognised. The
recommended solutions consist of several techniques and technologies including rated
sections (stage-discharge relationships), acoustic technologies, including bed-mounted
ultrasonic Doppler and transit time systems, and the use of both existing and new flow
measurement structures.

The potential uncertainties in discharge data were considered at the investigation and the
design stage. It is generally believed that the network is capable of obtaining discharge
data to uncertainty levels of within 10% at the 95% confidence level for the majority of the
gauging stations. The development of the stage-discharge relationships and velocity index
ratings will be required at rated section and acoustic installations. A range of gaugings will
be undertaken at each of the discharge measurement sites for performance checking or

1 Abtin Co. Iran, E-mail : [email protected]

429
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

calibration purposes. Calibration will be undertaken by conventional current meter gauging

measurement techniques and supported by the use of Acoustic Doppler technology where
appropriate. This gauging data can then be used to provide an estimate of the uncertainty
in continuous discharge measurement for each site. The gauging station calibration and the
determination of uncertainties are described in this paper.

Discharge data from each of the sites will be sent via GPRS telemetry, this data feed provides
a means of remotely monitoring. Also, live data will be displayed at each station. The data
hosting capabilities allow multiple users to access the data ensuring it can be quality checked
and managed effectively.

Key words: Irrigation canals, Discharge measurement, Uncertainty, Current meter, Doppler
technology.

RESUME
La mesure prcise de dbit dans le rseau dirrigation est dune importance fondamentale
lvaluation et la gestion optimale des besoins en eau dirrigation en Iran. Le gouvernement
de lIran travers son Dpartement des Ressources en Eau (WRM), a entrepris un programme
visant amliorer le suivi des flux dans les canaux dirrigation travers le pays. La compagnie
Abtin, avec lappui technique dHydro-Logic Ltd. (Royaume Uni), ont t commands pour
assister la conception et la mise en uvre de solutions de mesure de dbit. Ce document
dcrit les mthodes de mesure du dbit utilise, y compris les connaissances techniques de
base, les problmes rencontrs pendant leur mise en uvre et une valuation des incertitudes
dans les valeurs de dbit calcules.

Vu ltendue et la varit de tailles de canaux et les conditions du terrain dchantillon,

aucune solution na t trouve qui aurait t valable pour tous les terrains, cependant
limportance de la normalisation a t reconnue. Les solutions prconises comprennent
de plusieurs techniques et technologies y compris les sections estimes (stade-dbit),
les technologies acoustiques, comme Doppler ultrasonique mont sur le billon et des
systmes du temps de transit, et lutilisation des structures de mesure de dbit, existantes
et nouvelles.

Les incertitudes potentielles des donnes de dcharge ont t examines pendant lenqute
et la phase de conception. Il est gnralement admis que le rseau est capable dobtenir des
donnes de dcharge lincertitude de 10% contre la certitude de 95% pour la majorit des
stations de jaugeage. Le dveloppement des relations niveau-dbit et les cotes de lindice
de vitesse seront exigs la section nominale et les installations sonores. Une gamme de
jaugeages sera mene sur chaque site de mesure de dcharge pour vrifier les performances
ou des fins dtalonnage. Ltalonnage sera entrepris par les techniques classiques de
la mesure de jaugeage dampremtre et il sera soutenu par lutilisation de la technologie
Acoustique de Doppler, le cas chant. Ces donnes mesures peuvent ensuite tre utilises
pour fournir une estimation de lincertitude dans la mesure de dcharge continue pour
chaque site. Ce document dcrit la calibration de la station dtalonnage et la dtermination
des incertitudes.

430
ICID 21st Congress, Tehran, October 2011 R.56.5.35

Les donnes sur la dcharge de chacun des sites seront envoyes via tlmtrie GPRS,
ce flux de donnes permet de surveiller distance. De plus, les donnes en direct seront
affiches chaque station. Les capacits dhbergement de donnes permettent plusieurs
utilisateurs daccder aux donnes. Cest pour assurer que la qualit des donnes peut tre
contrle et gre de faon efficace.

Mots cls: Canaux dirrigation, mesure de dbit, incertitude, ampremtre, technologie

Doppler.

1. INTRODUCTION

Only a small proportion of Iran has sufficient rainfall for the cultivation of crops. The rest of the
country has arid or semi arid climate. Therefore, it is necessary to establish a large irrigation
network to provide valuable water resources. Iran has undertaken a large investment in water
resources and irrigation infrastructure with a series of dams and channels to maximise its
utilisation. Significant works in this regard have been accomplished since the middle of the
20th century that has helped deliver water management and developed agribusiness.

Effective monitoring of flow in the irrigation channels is important to enable timely and
proactive management of the available water resource. The Iranian Department of Water
Resources Management (WRM) have identified locations throughout the country where
reliable, good quality flow data is required. Abtin Co in association with Hydro-Logic Ltd has
been involved with the provision of flow monitoring facilities at a number of these sites. The
work undertaken has included site surveys and the selection of the most appropriate flow
measurement methodology for each location.

Many of the flow measurement sites identified are in newly constructed channels, while others
have been used for irrigation purposes for some time. At some of the sites, flow monitoring
infrastructure already existed, while at other locations no monitoring had previously been
undertaken. Earlier monitoring work had included spot flow measurements from cableways
and bridges using conventional rotating element current meters and the derivation of some
stage (water level) - discharge relationships.

Several different flow monitoring techniques have been identified and applied based on
specific site conditions. The use of a range of flow measurement techniques means that a
wide range of conditions can be measured in the most accurate and cost effective manner.

2. SITE CONDITIONS
2.1 Background

The conditions of both the physical properties of the channel and the flowing body of water can
have a significant impact on the ability to undertake a reliable and accurate flow measurement.
These impacts can include factors such as skewed flow, entrained air, silt and weed. Some
of the key issues faced and their potential impacts are indicated below.

431
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

2.2 Channel conditions

Conditions in the irrigation channels in Iran presented a range of different flow measurement
challenges. Reported flows at the sites under consideration ranged from prolonged dry
conditions to 90 m3s-1. As some channels were only used for limited periods the flow
measurement methods selected had to be sustainable throughout prolonged dry periods so
there are no problems in data quality when channels start to flow. Stage (water level) ranges at
some sites were up to 7 m presenting further challenges for accurate measurement techniques
throughout the range of flows. An example of a large irrigation channel is shown in Fig. 1

Fig. 1. Large irrigation Channel

The majority of the flow monitoring stations were on channels constructed for the purpose
of irrigation and formed from concrete with a graduated longitudinal fall. Generally the sites
were rectangular or trapezoidal concrete channels and in excellent repair, providing good
flow measurement conditions.

2.3 Velocity distribution

Poorly distributed velocities or highly turbulent flow can introduce uncertainty into flow
measurements. The use of field surveys prior to site and method selection is important in
ensuring that reliable flow measurement can be achieved throughout the full flow range. To
ensure a good velocity distribution at the point of measurement most flow measurement
techniques require an optimum straight length of approach, ideally a minimum of 10 times
the width of the channel.

432
ICID 21st Congress, Tehran, October 2011 R.56.5.35

2.4 Entrained air

Sluice gate systems particularly at dam discharge points cause a large volume of entrained
air in the discharged body of water (Fig. 2), this can also occur at natural falls or at discharge
points. The presence of too much air within the body of water can cause errors in velocity
estimation, or complete failure when using some acoustic flow measurement technologies.
In particular this applies to transit time/time of flight systems. Conversely, Doppler devices
will operate well with some air in the water as long as this is not excessive.

Fig. 2. Entrained air

2.5 Silt and suspended solids

A build up of silt can impact flow measurement. In high energy systems the silt particles are
held in suspension in the water and carried along with the flowing body of water. If the water
is slowed down by a backwater effect then the suspended material can drop out and form
a layer of silt, affecting flow measurement. Upstream of one of the measurement points a
number of settlement tanks had been constructed to remove sediment from suspension in the
water body before passing onto the network (Fig. 3). A build up of silt may block bed mounted
ultrasonic Doppler systems and cause a change in the area or even obscure lower paths
in a transit time (time of flight) system. Build up of silt at gauging structures with discharge
controls can also significantly impact on the reliability of a defined rating. The occurrence of
high levels of suspended solids at transit time gauging stations can cause system failure as
the sound pulses may not reach their target destination.

433
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

Fig. 3. Silt settlement tanks

2.6 Non Modularity and variable backwater

If using a structure to create a stage discharge relationship or derive a theoretical flow it is

important to ensure that modular flow is maintained. Modular flow means there is a change
from sub-critical to super-critical flow at the structure, as the flow moves over the control.
This means that the upstream water level is independent of the downstream conditions.

Sluices and controls located downstream of a measurement section may create a variable
backwater effect. This in turn means that relationships between stage and flow are unstable.
This issue can also occur with vegetation, as during seasons of growth there will be an
increase in head for the same flow compromising any relationships between stage and flow.

2.7 Temperature

In Iran a considerable range of temperatures is experienced both in air and water. Air
temperatures can vary from -20 to 60 oC. Instrumentation specifications should match the
temperature ranges experienced. Cabinets with good ventilation and suitable colour coat are
designed to minimise the impact of temperature extremes and to provide a safe and secure
housing for equipment.

2.8 Impact of pumps

Pump systems are placed in many of the irrigation channels for water abstraction. Flow
measurement should take place away from the turbulence and eddy impact zones of pumping
to ensure flow measurement accuracy.

434
ICID 21st Congress, Tehran, October 2011 R.56.5.35

2.9 Gauging and location of suitable infrastructure

It is important to ensure that calibration and check gauging can be undertaken throughout
the full range of flows. When possible flow monitoring stations should be located close to
existing bridges or cableways to assist with current meter gauging. Where flow measurement
structures are located in channels they should be placed close to channel crossing to allow
access to both banks and help with the maintenance of stations and calibration.

3. FLOW MONITORING TECHNIQUES

3.1 Background

The range of site conditions has meant that to ensure cost effective and accurate flow
measurement at the sites several different flow measurement techniques have to be employed.
The number of selected techniques have been kept to a minimum so that operators are not
required to be proficient with a large and diverse range of flow measurement techniques.
The techniques used include:
1. Purpose built flow measurement structures that comply with the relevant International
Standard: Both rectangular thin plate weirs and existing Parshall flumes presented
suitable flow measurement techniques. Provided these are well maintained, constructed
in accordance with the relevant standard and operated within their design range, flow
measurement structures are capable of producing good quality flow data. However, the
design process requires that the modular limit is maximised while the amount of afflux
is minimised.
2. Rated Sections i.e. stage (water level) discharge relationships derived by current meter
gauging. These are low cost and low impact and can provide excellent means of continuous
data. The problem with rated sections is that they require considerable gauging effort and
the stage-discharge relationships can be unstable due to seasonal changes in vegetation
growth and other variable backwater affects such as the operation of sluices.
3. Acoustic techniques: In recent years acoustic stream flow monitoring technologies have
been used more extensively throughout the world as alternatives to the more conventional
techniques. These can represent a cost efficient method of accurate flow measurement.
They have been utilised for a range of sites and can operate in variable backwater
conditions. Key methods employed in the irrigation project include both bed mounted
ultrasonic Doppler devices and time of flight/transit time ultrasonic technology. The time
of flight or transit time method also has benefits in terms of reduced calibration effort.

For all flow measurement solutions good hydrometric practice must be employed to ensure
recording good quality data. A key element is the accurate measurement of stage. A staff
gauge should be correctly installed at all the measurement sites, these should be manufactured
according to the guidelines contained in Hydrometry Water level measuring devices (ISO
4373: 2008). As a number of the measurement locations are in trapezoidal channels sloping
staff gauges can be used to provide an accurate reference check water level measurement.
Sloping staff gauges need to be manufactured to suit each individual site i.e. the manufacture
is dependent on an accurate survey of the slope of the channel sides.

435
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

3.2 Flow Measurement Structures

Flow measurement structures are not so widely used for new sites as they used to be
since there are now more effective lower cost solutions available. However where purpose
built structures are already in existence, or where existing structures can be modified, flow
measurement structures can provide a robust and effective means of flow measurement at a
site. The uncertainty in flow determinations for a well located and maintained site should be
of the order of 5% at the 95% confidence level. Due to the nature of a structure, calibration
is not usually necessary within the design flow range. In part of the project area Parshall
flumes were already in place. A further thin plate weir modification was also recommended
for improving an existing channel control structure.

Both of these types of structure have appropriate International Standards describing their
design application and use, namely:
1. Rectangular thin plate weirs (ISO 1438:2008);
2. Parshall flumes (ISO 9826: 1992)

Rectangular thin plate weirs: The rectangular thin plate weir is a commonly used structure
for providing accurate flow measurement in small open channels. The key parameters and
installation and maintenance requirements are outlined in the International Standard ISO
1438: 2008. One of the main requirements for a thin plate weir is that there is good aeration
of the nappe.

Fig. 4. Example of thin plate weir gauging structure

436
ICID 21st Congress, Tehran, October 2011 R.56.5.35

Parshall flumes: The Parshall flume is one of a large class of open channel primary structures
known as critical flow Venturi flumes. It has a rectangular cross section and a unique rating
based on the size of the flume. A distinguishing characteristic of the Parshall flume is the
downward sloping invert of the throat. This feature gives the Parshall flume its ability to operate
at higher ratios of downstream to upstream head than any other such device. The three
stages of the flume include a level approach with a downward sloping throat and a diverging
outlet that slopes upwards. The Parshall flume acts a short throated flume with the control
located near the end of the level floor in the converging section. A Parshall flume is shown
in Fig. 5.

Fig. 5. Parshall flume

Unlike long throated flumes (ISO 4359: 1983 {revised edition to be published in 2012}),
the formula/stage-discharge relationships for which are based on hydraulic theory, Parshall
flumes are based on laboratory testing. They come in a number of sizes which have been the
subject of detailed hydrometric assessment under laboratory conditions. Therefore, a new
construction has to be built in exact accordance with a specified size. The head measuring
device is located in the converging throat of the flume at the exact position specified in the
ISO standard. Check gauging is recommended to ensure that the structure is performing
correctly.

437
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

3.3 Stage-Discharge Relationships

The stage discharge technique relies on a natural or artificial control that provides a fixed
and stable relationship between the discharge and the upstream water level (stage). By
undertaking a range of calibration gaugings the relationship between an observed stage and
its corresponding flow can be defined. A good fall was noted at a range of the sites providing
ideal conditions for this solution.

The principles of the stage discharge technique are outlined in the International Standard ISO
1100-2 :1998 Measurement of liquid flow in open channels- Part 2 : Determination of the
stage discharge relationship. A new version is due to be published in 2011. This will outline
the minimum requirements for accurate flow measurement using this technique. ISO 1100-2
recommends that a minimum of 15-20 gaugings are undertaken to define the stage-discharge
relationship. The stage-discharge relationship is often defined by one or more shifted power
law equations of the form:

Q = (h a ) Equation (1)

Where, Q = discharge (m3s-1), h = stage (m) and C, a, are constants.

In order to undertake the stage measurement, the most cost effective and robust means
was to install a pressure sensor appropriate for the design stage range, paired with a data
logging device.

3.4 Acoustic Flow Measurement Techniques

3.4.1 General

In challenging hydrometric conditions there can be significant advantages in utilising acoustic

flow monitoring techniques. There are four main categories of acoustic technologies that are
currently used for velocity, and thus flow, determination.

Acoustic flow monitoring techniques include:

1. Bed mounted devices that are based on the Doppler shift principle these only sample
a portion of the channel. The portion sampled is usually unknown.
2. More sophisticated range gated devices based on acoustic Doppler current profiler
(ADCP) technology
3. Echo correlation devices - these look like bed mounted Dopplers but are based on a
somewhat different principle.
4. Time of flight/transit time ultrasonic.

Following the investigations of site details it was decided that the use of both time of flight
/ transit-time ultrasonic and bed mounted Doppler devices would be appropriate for some
of the sites.

438
ICID 21st Congress, Tehran, October 2011 R.56.5.35

3.4.2 Transit time / time of flight ultrasonic

Transit time ultrasonic river gauging is based upon the principle that the time taken for sound
pulses to be transmitted through flowing water over a known distance, is different to the time
taken for the pulse to travel the same distance when that water is stationary (ISO6416:2004).
When pulses are transmitted at an angle to the direction of flow the speed of the pulse in the
downstream direction will be enhanced by the flow of water whilst, returning in the upstream
direction, the speed of the pulse will be impeded by the flow. The difference in the travel times
in each direction between the same points will be proportional to the component of water
velocity along the flight path taken by the signal. The timing difference referred to is very
small (of the order of millionths of a second) but can be measured accurately electronically
and the resolved component of mean velocity of water along the flight path calculated. A level
sensor is also incorporated into the system. This technology has been successfully used in
stream flow measurement since the nineteen seventies.

Using such systems velocities can be measured at different levels or paths. If there are sufficient
paths operating within the cross-section and these are well distributed throughout the water
column these devices are capable of returning accurate determinations of flow with little
calibration effort. They work particularly well in clear water which is a problem with some of
the Doppler technologies which require particles or reflectors in the water to function. These
systems are ideally suited to installation in the concrete channel sections seen throughout
the majority of the irrigation sites. These provide excellent fixing points and for the trapezoidal
channels a bespoke fixing bracket allows the paths to be configured accurately. An example
of the probes used is shown in Fig. 6.

Where flow measurement is required in sections with skewed flow, downstream of bends in
the irrigation system, a cross path system has been specified. This measures velocity in both
directions across the channel width and therefore provides a more accurate determination of
flow by removing bias. The paths are offset to measure a range of levels through the vertical
section, maximising measurement accuracy.

A further application was used installing the time of flight system in much smaller sites than
usually applied. This involved the use of a single pair of transducers measuring velocity along
a single path. As water level varies what this single path represents in the vertical section
will change and therefore it is necessary to undertake a velocity index calculation based on
a series of flow gaugings under different conditions (see Section 3.5).

439
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

Fig. 6. Time of flight probe

3.4.3 Doppler devices

The Doppler method is based on the principle named after Christian Doppler 1843 (ISO
15769:2010). Sound is transmitted from the device and reflected from particles in the water
column, the reflected sound is measured by a receiver on the instrument. The transmitter
and receiver have overlapping fields and the angle of this enables a shift in frequency to be
measured that is used to infer the velocity of the particle and therefore the water body carrying
the particle. The ultrasonic Doppler flow meters were originally developed for use in sewer
systems where they have been widely and successfully used for a number of years. It is only
more recently that they have been used in larger open channels and natural watercourses.
Most ultrasonic Doppler systems will only sample velocity in a portion of the open channel.
The size of the portion sampled is dependent on several factors. These include the spread and
angle of the acoustic beam (a torch beam is a good analogy), the other design characteristics
of the sensor e.g. how the sound is transmitted, received and processed and the physical
characteristics of the channel e.g. the dimensions of the channel and the amount of suspended
solids. Some simpler systems have failed when used in clear streams during periods when
there are very few suspended particles in the flowing water.

One of the main problems in using these systems is ascertaining what portion of the channel
has been sampled. Therefore, it is often necessary to derive a relationship between the
measured and the mean velocity in the cross-section. This requires current meter gauging,

440
ICID 21st Congress, Tehran, October 2011 R.56.5.35

which can be used as an index velocity to determine the mean velocity in the measuring
section. This is sometimes referred to as the velocity index technique. These systems also
generally have an in-built water level sensor. The profiling range of the velocity sensor is limited
and may be reduced by changes in suspended material. For example if there is an increase
in sediment the acoustic beam will penetrate less of the vertical velocity profile as the signal
is restricted by the particle matter in the water.

These bed mounted devices were used on sites within the irrigation project where there was
insufficient depth for a transit time system or a suitable stable control to develop a stage-
discharge relationship. In order to achieve a critical minimum depth water level can be raised
by means of a downstream control structure.

3.4.4 The use of velocity index rating techniques

Velocity index techniques, such as those required when using the Doppler technology, are
velocity area methods. Discharge is therefore determined using the continuity equation,

Q = V A Equation (2)

Where

Q = discharge (m3s-1)
= mean velocity at instrument measuring section (ms-1)
A = cross-sectional area at instrument measuring section (m2)

For many installations the velocity determined by the instrument will not be the same as
the mean velocity in the measuring section since in most channels the instrument will not
sample the whole of the flowing cross-section. The mean velocity is determined by using a
relationship between the mean velocity and the instrument velocity, usually referred to as the
index velocity. Velocity index relationships can take the following general forms:

V = fn( Vi ) Equation (3)

V = fn( Vi , h ) Equation (4)

Where

Vi = instrument/index velocity (ms-1)

h = stage (m)

The cross-sectional area is determined using a relationship between stage and cross-sectional
area. This can be determined by cross-sectional survey.

The computation process is summarised as follows:

441
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

1. Mean velocity is determined from the velocity index rating, using the measured velocity
and, for more complex ratings, also the stage;
2. The cross-sectional area is determined from the stage via a stage area relationship,
look-up table or direct calculation from cross-sectional survey data directly input into the
instrument;
3. The corresponding mean velocity and cross-sectional are multiplied to obtain the
discharge.

4. DATA MANAGEMENT
4.1 Background

The effective management of the irrigation network across Iran is supported by the timely
feed of flow data from the monitoring points to the operational control centre. This has been
achieved by using a suite of Isodaq GPRS data enabled loggers. These transfer data from
the sites on a daily basis to a central holding database system. The system is web based.
Therefore, data can be accessed by any user who has the unique sign in name and password.
This data handling site allows graphical display and download of the sites to ensure that the
irrigation network can be measured effectively. Live displays through the processing device
or the logger are also made available at each site for data quality assessment.

4.2 Logging Systems

Two types of logger have been utilised at the sites depending on the flow measurement
methodologies used. Where an alternative display is available on a primary system e.g. a
transit time processing unit a GPRS logger with no display is used to transfer data. At stage
discharge and structure sites where no display is available from the sensor system a logger
with integrated display has been used Fig. 7. This has a display screen and the facility to
incorporate a stage-discharge relationship table into the logger so that discharge can be
displayed on site. Where possible these logging systems are located in a control room to
improve security and functionality.

442
ICID 21st Congress, Tehran, October 2011 R.56.5.35

Fig. 7. Logging devices at a Parshall Flume gauging station

All data is logged at 15 minute intervals, this has been globally adopted as an appropriate
time interval in obtaining a representative flow in surface water applications. Groundwater
systems are generally logged at an hourly interval. It is possible to alter all these systems
logging intervals to ensure an appropriate degree of data is issued at all times.

4.3 Data Interface

The data interface used allows users to interact with the data and display graphs and data
from the gauging station network. This can be shown in a map format for easy identification
and management. The interface with the loggers allows remote configuration, helping to
effectively manage field sites remotely. The operation of the irrigation network can be assisted
by a series of alarm thresholds that can trigger text and email alarms when key parameters
rise or fall through a defined threshold.

443
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

Data can be viewed on time series graphs that can be scaled over a range of time periods to
help identify trends in data and ensure that the water resource is being monitored effectively.
Data can be exported from the data management software in a range of formats for further
data interpretation and archiving on operational databases.

A typical plot of the daily feed of flow data from the sites is shown below. This data is from
the two Parshall flume structures and shows head (upstream water level), tail (downstream
water level) and calculated flow.

Fig. 8. Typical data presentation

5. UNCERTAINTIES
5.1 Background

All hydrometric measurements are subject to errors. Error is the difference between the true
value and the measured value. As the true value is unknown, it is not possible to determine
the error. Therefore, the uncertainty of a measurement is estimated which is the range in
which the true value is expected to lie. Normally the uncertainty is estimated at the 95%
confidence level (two standard deviations). This means that there is a 95% probability that
the true value will lie within the estimated uncertainty limits.

As stated in the EN ISO 25377:2007 Hydrometric Uncertainty Guidance

444
ICID 21st Congress, Tehran, October 2011 R.56.5.35

All measurements of a physical quantity are subject to uncertainties.....The result of a

measurement is only an estimate of the true value of the measured quantity and therefore is
only complete when accompanied by a statement of its uncertainty.

And

The discrepancy between the true and measured values is the measurement error. The error,
which cannot be known, causes uncertainty about the correctness of the result.

Hydrometry is not a precise science and it is important that the users of hydrometric data
are made aware of the uncertainties in the flow determinations. The uncertainty in a stream
flow measurement is a combination of a number of individual measurement and assumption
uncertainties. The individual uncertainties are normally combined in quadrature.

The ISO recommends that uncertainties should be estimated in the first instance, at the
68% confidence level (1 standard deviation). However, for most hydrometric applications
uncertainties are expressed at the 95 % confidence level (2 standard deviations).

5.2 Uncertainties in Flows Derived using Measurement Structures

The uncertainty (U*c) for a flow measurement structure is a combination of the combined
uncertainties in the discharge coefficient (Ce), the structure geometry and the stage. The
uncertainty in the geometry for a rectangular notch is the uncertainty in the width (b). For a
v-notch the uncertainty in the geometry is the uncertainty in tan(a/2) where a equals the notch
angle. The stage uncertainty (he) is a combination of instrument and gauge zero uncertainties
and can be estimated as follows:

*
u (he ) 2 + u ( E ) 2
u he = 100 Equation (5)
he

Where u(he) is the absolute uncertainty in the instrument measurement and u(E) is the absolute
uncertainty in the stage zero for a given stage he.

The overall uncertainty for rectangular notch is given by the following equation:

U c* (Q) = u * (C e ) 2 + u *b 2 + (1.5u * (he )) 2 Equation (6)

An uncertainty curve for a full width rectangular notch 3 m wide and 0.5 m high, operating
within its design range, is shown in Figure 9. It has been assumed that the uncertainty in the
stage measuring instrument is 2 mm and the gauge zero uncertainty is 1 mm at the 68%
confidence level. The minimum recommended operating stage for a rectangular notch is 30
mm which corresponds to a discharge of 0.029 m3s-1 for which the estimated uncertainty at
the 95% confidence level is about 12.3 %.

445
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

Fig. 9. Example of an uncertainty curve for a thin plate weir

By far the most significant source of overall uncertainty is the uncertainty in the stage
measurement. At the majority of the sites good quality low range, pressure transducers/
transmitters are used. If the level range of the pressure transmitter is not greater than 3 m,
these should be capable of measuring water level to 3 mm when set-up correctly.

Generally a well constructed and maintained flow measurement structure operating within
its design range should be capable of delivering flow data to within 5 10% at the 95%
confidence level (two standard deviations).

5.3 Uncertainties using Other Techniques

5.3.1 Velocity index techniques

The overall uncertainty can be obtained by combining the component uncertainties in

quadrature. The component uncertainties are those related to the velocity index rating,
the index/instrument/measured velocity, the cross-sectional area/stage area relationship
and the overall uncertainty in the stage. The largest source of uncertainty is generally in the
velocity index rating. Provided this is relatively stable and is defined by a reasonable amount
of gaugings (min. 10 15) over the target flow range, it should be possible to determine
discharge to within 10%.

446
ICID 21st Congress, Tehran, October 2011 R.56.5.35

5.3.2 Time of flight/transit time ultrasonic

The overall uncertainty is a function of uncertainties due to the limited number of paths (pairs
of transducers) and their distribution in the vertical, the instrument calculated velocity, the width
measurement and the stage or depth measurement. The largest source of uncertainty is likely
to be the former. However, with a four path system with a well distributed array of transducers
it should be possible to determine discharge at the 95% confidence level to within 5 10%.

5.3.3 Rated sections

The uncertainties associated with rated sections are a function of the stability and sensitivity
of the control, the number of gaugings that have been used to derive the stage-discharge
relationship and the overall uncertainty in the stage measurement. For stable controls a
minimum of 15 gaugings is usually recommended (ISO1100-2:1998) over the target flow
range. For stable controls it should be possible to determine discharges to within 10% but
where variable backwater occurs due to variations in seasonal weed growth the uncertainties
could be significantly larger. However, as most of the sites in the Iranian irrigation channels
where stage-discharge relationships have been used are very good controls, uncertainties
of 5 10% are readily achievable.

6. CALIBRATION AND MAINTENANCE

6.1 Background

To achieve good quality reliable data attention must be given to high quality calibration and
maintenance in the field. If a solution requires calibration this must be done over the full flow
range to minimise errors through extrapolation. Maintenance is recommended on a routine
basis, with monthly visits considered good practice.

6.2 Current Meter Gauging

The key method of calibrating or checking a station is by gauging the flow with a current
meter. A number of different methods of flow gauging have been adopted in hydrometry. The
most applicable methods for the channels encountered in Iran are rotating element current
meters or Acoustic Doppler Current Profilers (ADCPs).

Current meter gauging has been, and still is, undertaken using rotating element current
meters (cup/bucket and impeller meters). These determine the velocity at the point where the
instrument is placed in the measuring section. In recent years, electromagnetic and ultrasonic
Doppler meters have been increasingly used by some organisations. Spot flow gauging is
undertaken using the velocity area method, where a cross-sectional area is defined and
velocities are measured at a number of verticals (positions across the measuring section).
Flow is a product of velocity multiplied by area. Under low flow gaugings in smaller channels
current meters are deployed by wading. In larger channels and for higher discharges, bridges
and cableways (see Figure 10) are used.

447
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

Fig. 10. Gauging cableway

The procedures for current meter gauging are outlined in the International Standard ISO:748:
2007. The recommended number of verticals for channel widths is shown in Table 1.

448
ICID 21st Congress, Tehran, October 2011 R.56.5.35

Table 1 Recommended number of verticals for current meter Gauging (Source: ISO 748:
2007)

Channel width (m) ISO 748: 2007

>0 & <0.3 5 to 6
>0.3 & <0.5 5 to 6
>0.5 & < 1 6 to 7
>1 & < 3 7 to 12
>3&<5 13 to 16
>5 & < 10 22
> 10 22

6.3 Acoustic Doppler Current Profilers (ADCPs)

A more recent method of gauging open channels is the Acoustic Doppler Current Profiler
(ADCP). The ADCP uses a similar principle to a bed mounted device by reflecting acoustic
signals from particles carried in the water column. The instruments are mounted on a small
boat or floating platform and are moved slowly across the surface of the water between the
two banks. Through a phased array of four transducers the instrument will measure velocity
in a conical sphere below the sensor face. The instrument will also measure direction and
depth providing a cross-sectional area. ADCP technology remains relatively expensive and
limits are in place regarding its distribution. It does however represent the most, robust and
cost efficient means of flow gauging in many of the larger channels seen in the irrigation
network, particularly those carrying significant volumes of water where health and safety
becomes an issue.

6.4 Flow Validation

A key element of ongoing data quality assessment is confirming that the flow measurement
method is still providing accurate measurement. This involves undertaking a flow gauging and
confirming that the reported flow from the station is within acceptable limits. If measurements lie
outside the acceptable limits then possible causes such as changed controls or maintenance
issues must be examined.

6.5 Ongoing Maintenance

A key factor in the quality of hydrometric data is the maintenance of a gauging station. This
includes checking for sensor drift against a fixed datum and ensuring that the station is clean
and well maintained.

The presence of silt has been noted at a number of sites in Iran. If this is dropped from
suspension in the water column and deposited on the bed then measurement accuracy
can be affected. This can change the area of a site, obscure sensors and make gauging
more challenging thereby increasing uncertainties. High quality routine maintenance must

449
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

be undertaken at all flow measurement sites to ensure that they are operating within their
hydrometric limits.

Fig. 11. The problem of silt accumulation and vegetation growth

Algal build up on a flow measuring structure can introduce significant errors in calculated flow.
This impact is particularly critical on the sites where a concrete structure is used to provide
a control for stage discharge rating assessment. In addition seasonal vegetation growth can
make stage-discharge and velocity index ratings very unstable. Crests of control structures
and gaugeboards should be cleaned on a routine bases to ensure that they do not impact
on measurement accuracy and introduce errors or uncertainties to a data set.

7. STANDARDS
It is essential that both continuous and spot flow calibration measurements are undertaken in
accordance with sound hydrometric practice. This can be achieved if the design, installation,
operation and maintenance of the gauging stations and undertaking of calibration gaugings
are in accordance with the relevant International Standards. There are appropriate standards
for all the techniques described in this paper and used on the Iran irrigation network flow
monitoring project with the exception of the application of ADCPs. However, a draft standard
on the use of ADCPs will hopefully be published in 2012.

450
ICID 21st Congress, Tehran, October 2011 R.56.5.35

The relevant standards are summarised in Table 2:

Table 2: Summary of relevant International standards

ISO Standard No. Subject matter Reference No.

748 Current meter gauging 3
1100-2 Stage-discharge relationships 4
1438 Thin plate weirs 5
4373 Water level measuring devices 6
6416 Time of flight/transit time ultrasonic gauges 7
9826 Parshall Flumes 8
15769 Continuous flow measurement using acoustic 9
technologies other than time of flight and velocity
index ratings
24578 (draft) Application of ADCPs Pub. 2012
25377 Hydrometric uncertainty 11

ISO hydrometric standards have been drafted by committees with expert representation from
a number of countries and provide good guidance on hydrometric practice. However, the
authors are of the opinion that standards are a compromise between practices in different
countries. As such they should generally be used as the norm or the minimum requirement
and not as rigid guides.

8. SUMMARY
The rehabilitated irrigation channel flow monitoring network consists of a mixture of flow
measurements structures, rated sections, time of flight ultrasonic and other acoustic gauging
stations. This network should be capable of providing flow data with uncertainties of within
5 10% at the 95% confidence level at all but the most extreme flows provided the stations
are operated in accordance with good hydrometric practice. However, in order to achieve
this level of uncertainty it is essential as for all hydrometric networks a regular maintenance
programme is agreed and implemented

ACKNOWLEDGEMENTS
The irrigation channel monitoring described in this short paper is based on work undertaken
by Abtin working in association with Hydro-Logic, for the Government of Iran through its
Department of Water Resources Management (WRM). The authors gratefully acknowledge the
support and encouragement of WRM. However, it should be noted that the views expressed
in this paper are those of the authors and as such should not be interpreted as Government
of Iran or department of WRM views or policy. In addition the authors would also like to
gratefully acknowledge the guidance, support and council provided their other colleagues

451
ICID 21st Congress, Tehran, October 2011 International Commission on Irrigation and Drainage

involved with the work including Michel Yonnet, Distec Hassan Nasuti, Abtin, Distec and Rod
Hawnt and Nik Whalley of Hydro-Logic Ltd.

REFERENCES

Ackers P, White WR, Perkins JA & Harrison AJM, (1978) Weirs and flumes for flow
measurement, John Wiley and Sons.
BSI BS 3680-7: 2000 EN-Measurement of Liquid Flow in Open Channels - Part 7: Specification
of Equipment for the Measurement of Water Level
ISO 748:2007 Hydrometry Measurement of liquid flow in open channels using current-
meters or floats
ISO1100-2: 1998 Measurement of liquid flow in open channels- part 2: determination of the
stage-discharge relationship
ISO 1438:2008 Hydrometry Open channel flow measurement using thin plate weirs
ISO 4373:2008 Hydrometry Water level measuring devices
ISO 6416:2004 Hydrometry Measurement of discharge by the ultrasonic (acoustic) method
ISO 9826: 1992 Measurement of liquid flow in open channels -- Parshall and SANIIRI flumes
ISO 15769 2010: Hydrometry Guidelines for the application of acoustic velocity meters
using the Doppler and echo correlation methods
EN ISO TS 25377:2007 Hydrometric Uncertainty Guidance

452