Performance Assessment and Potential

Indicators for Drainage Systems

Editors: Evan Christen and Bernard Vincent

Authors: Vincent B., Vlotman W.F., Zimmer D., Hornbuckle J.W.

CSIRO Land and Water Science Report 62/07

September 2007
Report Title: Performance Assessment and Potential Indicators for Drainage Systems

Editors: Evan Christen1,2,3 and Bernard Vincent4

Authors: Vincent B. 4, Vlotman W.F.5, Zimmer D. 6, Hornbuckle J.W. 1,2,3

Affiliations:
The authors are members of the International Committee of Irrigation and Drainage Working
Group on Drainage, they represent the following organisations:
1
CSIRO Land and Water, Griffith, NSW 2680,
2
CRC for Irrigation Futures, PO Box 56, Darling Heights Qld 4350, Australia
3
University of New England, Armidale, NSW 2351, Australia
4
Cemagref, Parc de Tourvoie, Antony, 92164, France
5
Goulburn-Murray Water, 40 Casey Street, Tatura, Victoria 3616, Australia,
6
World Water Council, Espace Gaymard 2 Quater pl Arvieux, Marseille, 13002, France

CSIRO Land and Water Science Report 62/07

September 2007

ISSN: 1834-6618

Copyright and Disclaimer

© 2007 CSIRO To the extent permitted by law, all rights are reserved and no part of this
publication covered by copyright may be reproduced or copied in any form or by any means
except with the written permission of CSIRO Land and Water.

Important Disclaimer:
CSIRO advises that the information contained in this publication comprises general
statements based on scientific research. The reader is advised and needs to be aware that
such information may be incomplete or unable to be used in any specific situation. No
reliance or actions must therefore be made on that information without seeking prior expert
professional, scientific and technical advice. To the extent permitted by law, CSIRO
(including its employees and consultants) excludes all liability to any person for any
consequences, including but not limited to all losses, damages, costs, expenses and any
other compensation, arising directly or indirectly from using this publication (in part or in
whole) and any information or material contained in it.
Foreword
This document was initiated as a follow-up of a workshop jointly organised by the Working
Group on Drainage (WG-DRG) and the Working Group on Performance Assessment
(dissolved in 2003) in Cairo in 1996. A first draft was produced and presented at the ICID in
Montreal 2002 and both from Australia and the USA additional contribution were received. A
paper was presented in a special Benchmarking workshop held at the ICID meetings in
Montpellier 2003. In 2005 it was decided to prepare a final version of the document with the
assistance of CSIRO, Land and Water, Griffith, Australia. Case studies available up to this
point have been included in this final version of the report.
The authors wish to thank all members of the WG-DRG and those of Task Force 4 (2000–
2004) of ICID on Benchmarking of Irrigation and Drainage Projects for sharing their
observations and remarks.
Performance assessment is not common in the drainage sector mainly because no
comprehensive method has been developed so far. This document aims at partly filling this
gap.
It firstly explains the rationale and the basic concepts associated with performance
assessment and then propose methods for performance assessment. It finally shows the rare
case studies that are available, which are used to compile the inventory of the Performance
Assessment programmes installed throughout the world.
The document is a state of the art document, it is not definitive and might be considered as a
first step on the path of incorporating drainage benchmarking into irrigation benchmarking. At
the same time there will always be room for independent drainage assessment at farm and
regional levels. Unaffordable methods such as information management systems, readily
available geographical information systems and satellite remote sensing techniques will
become more and more applicable in the forthcoming years. Because of this expected
progress and because more case studies will become available, the task force who have
prepared this document would like to see it as a first volume of Drainage Performance
Assessment and Benchmarking. As new information becomes available, an additional
volume may be considered.
This report is divided into two parts; Part A describes performance assessment approaches,
assessment parameters, indicators and criteria and discusses the role of benchmarking,
whilst Part B provides practical case studies from Pakistan, Egypt, Morocco, France and
Australia.

Performance Assessment and Potential Indicators for Drainage Systems i

Contents
PART A ....................................................................................................................................1

1. Performance assessment.............................................................................................2
2. Parameters, Indicators, Criteria...................................................................................8
3. Benchmarking Drainage systems in IWRM ..............................................................26
4. Conclusions and Recommendations ........................................................................31
Annex A Potential parameters for Operational and Strategic drainage indicators.....32
Annex B Rationale and long-list of indicators for drainage system performance
assessment..................................................................................................................39

PART B ..................................................................................................................................44

1. Pakistan - Water Table Analysis ................................................................................46

2. Egypt - Water Table Analysis .....................................................................................48
3. Egypt - Complaint Assessment .................................................................................57
4. Morocco - Hydraulic System Indicators....................................................................66
5. Morocco - Crop yields in the Gharb Plain.................................................................71
6. France - Drainage Efficiency and Agronomic Outcomes ........................................74
7. Australia - Salinity Control in Vineyards...................................................................85

REFERENCES .......................................................................................................................92

Performance Assessment and Potential Indicators for Drainage Systems ii

PART A
This section of the report describes performance assessment approaches, assessment
parameters, indicators and criteria and discusses the role of benchmarking.

Part A Contents
PART A ....................................................................................................................................1

1. Performance assessment.............................................................................................2
1.1. Peculiarities of drainage performance assessment .........................................3
1.2. Definitions ............................................................................................................4
1.3. Scope of drainage performance assessment ...................................................4
1.4. Implementation of Performance Assessment Programmes............................5
1.4.1. Preliminary Investigations (First Step)........................................................7
1.4.2. Primary Investigation (Second Step)..........................................................7
1.4.3. Cause Analysis (Third Step).......................................................................7

2. Parameters, Indicators, Criteria...................................................................................8

2.1. Properties of Performance Indicators................................................................8
2.2. Potential parameters ...........................................................................................9
2.3. Selected indicators as function of rationale .....................................................9
2.4. Indicators............................................................................................................13
2.4.1. Water table as function of time (hydrograph). ..........................................14
2.4.2. Water table draw down rate. ....................................................................15
2.4.3. Water table as function of area. ...............................................................16
2.4.4. Q-h plots...................................................................................................17
2.4.5. Water level in canals ................................................................................19
2.4.6. Evolution of the Manning-Strickler Coefficient..........................................20
2.4.7. Crops responses ......................................................................................21
2.4.8. Salinity control ..........................................................................................22

2.5. Reliability and usability of the Performance assessment results. ................24

3. Benchmarking Drainage systems in IWRM ..............................................................26

4. Conclusions and Recommendations ........................................................................31

REFERENCES .......................................................................................................................92

Performance Assessment and Potential Indicators for Drainage Systems 1

1. Performance assessment
The purpose of this paper is to report progress of the ICID Working Group on Drainage with
Performance Assessment of drainage systems and to clarify the link of drainage performance
assessment with irrigation and integrated water management, as well as make the link to the
activity of benchmarking by the ICID Task Force 4. The paper aims at stimulating discussion
and enhances PA of drainage systems in general.
The state of the art of PA in the irrigation sector, after some ten years of development, was
reviewed at a workshop held at the 15th ICID Congress, the Hague 1993 (Bos, Abdel -
Dayem et al. 1994b). Although largely born and developed in the irrigation community, the
developed PA methods always, right from the beginning, covered irrigation as well as
drainage systems. It was not until the 16th ICID Congress held in 1996 in Cairo, when ICID
dedicated a workshop to this subject, that the drainage community itself became actively
involved in the PA debate in the sector. Since then the production of a special report on
Performance Assessment and Indicators for Drainage Systems has been on the agenda of
the ICID Working Group on Drainage and the current draft is the basis of this article (Vincent
2002). In 2000 performance assessment was expanded to benchmarking, initially
coordinated by IPTRID and presently by Task Force 4 of ICID, which held workshops at the
2001 and 2002 meetings of the International Executive Council (IEC) of ICID (Malano 2003).
The purpose of benchmarking is to be able to compare water management systems
worldwide. The purpose of performance assessment (PA) is to assess the functioning of a
system against set standards using certain specified indicators. Performance Assessment is
generally internal of the organisation, while benchmarking compares externally to an
organisation. Hence, benchmarking is one further step in the sequence of assessments that
take place during a water management system lifecycle; planning, design, construction,
operation and management and maintenance. Performance assessment is one step up from
monitoring and evaluation. Immediately the question arises-what is then quality control and
how does research relate to this all? The key to distinguishing between these activities is to
distinguish between technical, operational and strategic objectives and/or rationales.
Basically, all assessment activities up to benchmarking are internally oriented. Benchmarking
aims at using the same data available for performance assessment externally, such that one
can compare among systems.
In assessing the role of benchmarking performance assessment in the drainage sub-sector, it
should be duly recognised that although irrigation and drainage are closely related sub-
sectors and disciplines, they differ in several aspects. This is especially true for the large-
scale irrigated schemes in the arid zone on which most the past PA work has focussed,
covering, as mentioned, both the irrigation and the drainage systems. The drainage sub-
sector also extends to large areas of rain-fed land in the humid zones, both temperate and
tropical, which have so far not been covered by the ongoing PA work. Generally, it may also
be observed that the past and present PA work in drainage focussed almost exclusively on
the performance of subsurface drainage systems and paid hardly any attention to surface
drainage and flood control systems. The present paper primarily concentrates on drainage
under irrigated conditions.
The most pertinent difference with respect to PA between the irrigation and drainage sub-
sectors is that the provision of the irrigation services generally requires extensive operation of
the system and interaction with the farmers. In comparison, and traditionally, drainage is
generally more a gravity driven process, which operates with minimal management and
engineering control. In view of the fewer operational variables and options, performance
assessment in the drainage is often felt to be of a lower need and value than in the irrigation
sub-sector.

Performance Assessment and Potential Indicators for Drainage Systems 2

It should, however, be noted that trends of change are in the air which may well make the
need for benchmarking PA in the drainage sub-sector in the future more pervasive. More and
more drainage is seen as a critical component in Integrated Water Resources Management
(IWRM). Firstly, there is the high importance attached to the protection of environmental
values, many of which, most notably the water quality and biodiversity, are more affected by
the performance of the drainage than of the irrigation systems. Secondly, drainage is slowly
catching up from past neglect as it becomes more recognised that adequate drainage is
critical to the sustainability of irrigated agriculture in the arid zone and also to the continued
agricultural development of much rain-fed agriculture. Thirdly, controlled drainage is strongly
recommended for many situations, which will mean more involvement of agencies and
farmers in the operation of the drainage system (Vlotman 2002). In view of these trends, the
performance of the installed drainage systems may well be expected to become, in future,
more subjected to public scrutiny and debate than in the past. Drainage managers may well
need better instruments to deal with the expected increased demands for information and
accountability.

1.1. Peculiarities of drainage performance assessment

The way drainage performance assessment applies to the IWRM sector is not yet fully
defined. This may be due to the different levels of drainage development or to the different
technical and socio-economic contexts in which drainage programs are established. It is
certainly also due to the fact that different organisations are involved with the same water
management system. Interrelations between the irrigation and drainage performance
assessment are clear (Smedema 1996); this is especially the case for criteria related to
salinity. But several differences may also be outlined.
● The importance of drainage to combat salinity or waterlogging is basically not evident for
users in irrigated areas. Subsurface drainage functioning is not so easy to perceive by the
farmers. The image of drainage is even negative as "it removes water from the soil" and
its effects appear opposite to those of irrigation.
● Failures of irrigation services are rather easy to detect whereas those of drainage are not
obvious. In addition, the consequences of drainage under-performance must be assessed
on the medium-long term while those of irrigation are immediate and clearly observed in
the crop development.
● Drainage performance assessment programmes may have different objectives: validate
design or construction methods; assess the needs for rehabilitation or the relevance of an
investment. These objectives may require different methods and indicators.
● It is impossible to build a drainage performance assessment program without the active
involvement of farmers, because farmers can provide reliable indicators for the
performance of their drainage systems.
In countries with established drainage, usual in the temperate climate zone, the impression is
that monitoring drainage systems is enough to assess their performances; in these countries
farmers are generally those who invest in drainage. In countries with drainage systems
installed in the last two decades, usually in the semi-arid and arid regions, performance
assessment of their drainage systems have in general not started. In these countries, the
state, in co-operation with international agencies, decides on drainage investments. Under
these conditions, there is a clear need to appraise the relevance of the investment after a few
years. This requires indicators that may differ from those needed by farmers and drainage
practitioners.

Performance Assessment and Potential Indicators for Drainage Systems 3

 1.2. Definitions
In order to be clear about the various assessments that can take place during a water
management system lifecycle the following definitions may serve as a guide:
1. Benchmarking (BM) can be defined as: "a systematic process for securing continual
improvement through comparison with relevant and achievable internal or external norms
and standards" (Malano 2001). Benchmarking is an externally focused activity that aims
at comparing organisations, finding comparable organisations and searching for best
management practices. These practices become references for other comparable
organisations. Benchmarking indicators are designed to enable comparison of
organisations and provide the best sensitivity in the vicinity of an identified target value.
Some performance assessment indicators, which are more likely to scan internal
processes, cannot be used for benchmarking.
2. Performance Assessment (PA) of irrigation and drainage systems. Definitions vary but
describing PA as “a methodology for investigating, using a limited number of indicators,
the level of service of installed systems” would seem to capture what most of the PA
researchers and practitioners seem to consider its most essential attributes.
3. Monitoring and Evaluation (M&E) applies to impact assessments of interventions rather
than to performance assessment. It is a more generic methodology than performance
assessment. The focus of M&E is often internal to the organisation. Although comparison
amongst schemes within the organisation is common, comparing it with external
organisations is not.
4. Quality Control (QC) is more related to engineering procedures; it aims at checking
construction requirements and specifications. Performance Assessment applies during
the whole lifetime of the structure, while Quality Control is generally executed only during
and just after the construction of the structure/system.
5. Research and development (R&D), provides new technologies or methods that differ from
providing information about functioning of structures. Research is not PA, although
applied research may use the data generated by BM, PA, M&E and QC.
Elaboration and explanations of terms used like indicators, target values, etc. follow below.

1.3. Scope of drainage performance assessment

Performance assessment is an essential part of management. Each organisation ought to
have objectives and means to achieve them (Figure 1) and it is essential for it to assess how
these objectives are fulfilled (effectiveness) and how its means and resources are utilised
(efficiency).

objectives
eff
ce

ec
an

1
veti
ev

ne
rel

ss

means, efficiency performance

resources
2

Figure 1: The management triangle

Performance Assessment and Potential Indicators for Drainage Systems 4

Performance can be defined and assessed from a variety of perspectives; it is (1) the degree
to which an organisation’s products and services responds to the needs of their customers or
(2) the degree with which the organisation uses the resources at its disposal (both strategic
and operational,(Bos 1996)), or (3) the degree to which the systems complies with the design
criteria (technical). The peculiarities of these are presented in Table 1.
Applied to drainage, operational performance refers to the level of compliance of a system
with its design criteria while strategic performance refers to its outputs, which are not
necessarily only the result of the drainage system performance. Operational assessment can
be regional or very site specific and in the latter case detailed technical design criteria are
used for indicator values. In practice, at field level, operational performance is related to a
proper removal of surface water, after rainstorms, or irrigation application and/or with a
proper control of the water table levels. At the system level, it is related to the proper removal
of the water discharged from the fields. Strategic performance is related to the different roles
of drainage in increasing and regulating crop production. It encompasses salinity control in
the root zone, contribution to rural development, and drainage effects on health and
environment. At a technical and operational level a simple direct indicator such as water table
behaviour in time, or water table depth considered spatially at certain critical periods, is
generally sufficient.

Table 1: Operational versus strategic performance assessment

Technical Operational Strategic

Level of functioning Level of service Cost-effectiveness;
environmental impact, crop
responses
Compliance with technical Compliance with (system) General (physical, economic)
design criteria design criteria objectives of drainage
Quality, effectiveness Efficiency Efficacy, viability
Short term Short term Medium and long term
Designers, consultants, drainage Drainage agency/company Drainage agency, higher
agency government departments i.e.
water board, donors

Performance of drainage systems is the result of several processes occurring at different

space and time scales (Figure 2). At small scales (plant, plot) biophysical processes are
predominant. At intermediate scales (fields, farm) technical aspects are predominant. At
larger scales socio-economic and political aspects are predominant.
In irrigated systems, it is questionable whether the strategic performance can be seen as the
performance of the drainage system alone: it is very difficult to separate the performance of
the drainage system from that of the irrigation system. Bearing this in mind, we will
nevertheless focus on drainage aspects.

1.4. Implementation of Performance Assessment Programmes

The standard PA procedure applied to the rehabilitation of an existing drainage system has
been schematically presented in Figure 3. The addition of the term “standard” is meant to
indicate that the proposed procedure is expected to have wide applicability but that other
procedures may be needed as well (Smedema 1996).
It involves three sequential steps. Each subsequent step is only undertaken when the
previous step has confirmed its necessity and, therefore, the PA process may end after each
step. It may not be necessary to complete all three steps or to do them in the indicated order.
The Preliminary Investigation may for example be stopped when the complaint assessment
indicates that the complaints do not require further action (and it would be resumed when the
complaints persist or when other new information becomes available). Each step requires a

Performance Assessment and Potential Indicators for Drainage Systems 5

set of indicators whereby the indicators will generally become more specific and detailed as
the procedure enters the next step.

Time

Region, country,
decade irrigated area

year farm

cropping
plot
season

rainfall,
irrigation plant
event

Space

Figure 2: Space and time scales to consider for performance assessment of drainage
systems

Figure 3: Proposed Standard Performance Assessment Procedure (Smedema et al. 2004).

Performance Assessment and Potential Indicators for Drainage Systems 6

 1.4.1. Preliminary Investigations (First Step)
This first step is proposed to include the following four activities:
1. Complaint management.
2. File/database search: this includes the age of the project together with the applied
technology (materials and construction methods); the applied quality control; the contract
documents; and other indications may be included and each indication may have value
singularly or in combination with others.
3. Agricultural data search: crop productivity and cropping pattern.
4. Rapid appraisal: a short field survey to assess the drainage conditions.
During this first step, the need for the second step is assessed. The latter step requires
considerable field work and expenditure and should only be undertaken when the preliminary
investigations have confirmed that there are sound indications that there are indeed
waterlogging and/or salinity problems in the area or in a considerable part of the area, and
that these problems are most probably due to a malfunctioning of the drainage systems.

1.4.2. Primary Investigation (Second Step)

This step is followed when there is a major waterlogging and/or salinity problem in the area
and these problems are due to a malfunctioning of the drainage systems. In this step, this
assumption is confirmed or rejected by collecting indicator data (e.g. water table depth and
soil salinity) and comparing these to the accepted standards of good performance. This step
may be broken down into two sub-steps:
1. Data collection and processing: monitoring the selected indicator parameters followed by
some form of processing to facilitate the use of the collected data.
2. Data evaluation: comparing of the collected indicator data with the accepted standards on
the basis of which judgments can be made on the performance of the drainage systems.
It is of course possible that these Primary Investigations reveal that there is no real
waterlogging and salinity in the area or that the prevailing conditions are not due to
malfunctioning of the drainage systems. In this case the observed problems are properly
reported and the performance assessment of the drainage system is ended.
The frequency of investigation for this step depends on the Preliminary Investigation and may
be different for each situation. Note that Standard M&E has been excluded from this
description: man-power and funds determine the frequency primarily in that case.
Observations of the Standard M&E collection in step 1 may be used for the PA step 2
purpose.

1.4.3. Cause Analysis (Third Step)

This phase is entered when the Primary Investigations have confirmed that the performance
of the installed drainage systems do not meet the expected standards. The remaining task is
then to identify the cause(s) of the under-performance of the system(s).
As will be clear from the foregoing, the process of PA of drainage (Figure 3) is shown from
primarily technical rationale. Moreover, the example suggests an ad-hoc approach to PA in
particular when cause analysis is added to the PA process. This is in practice often the case.
In light of the benchmarking of PA, and the service oriented definition of PA, it should be
evident that the process should be much more structural and systematically approached in a
water management organisation.

Performance Assessment and Potential Indicators for Drainage Systems 7

2. Parameters, Indicators, Criteria
The basic information for performance assessment is named hereafter “parameter”.
Parameters can be provided either by measurements or by field observations. Thus a
parameter can be either qualitative or quantitative.

Indicator/criteria

Parameter 1
target level and
Parameter 2
allowable range
Parameter 3 of deviation

indicator moderately
poor
criteria

poor
indicator
value

Figure 4: Indicators, criteria, target levels.

Performance has to be assessed by use of a set of indicators associated with criteria and
target levels (Figure 1). Combinations of parameters transform them into indicators. Criteria
are ranges of indicator values used to classify the indicator values in acceptable or non-
acceptable ranges (e.g. the soil salinity can be classified as non saline, slightly saline, saline
or severely saline). A nominal value and a range of deviation of acceptable values around the
nominal value define target levels.
The major challenge of Performance Assessment is not so much defining the indicators but,
rather, recommending the criteria with target levels and ranges. Target values are most likely
climate and site specific. They have been or need to be derived from case studies. A number
of case studies were available and summaries of the case studies are given in Chapter 3.
They will be used to derive the criteria and target levels.

2.1. Properties of Performance Indicators

Assessment indicators for drainage and irrigation systems performance have to be
characterised by the following (Bos, Murray-Rust et al. 1994a):
A number of parameters can make up an indicator
Indicator: calculated or measured variable which can change with time.
Parameter: a variable measured at a certain point in time.

Scientific basis
The indicator should be based on an empirically quantified, statistically tested causal model
of that part of the system process it describes. Discrepancies between the empirical and
theoretical basis of the indicator must be explicit. To facilitate international comparison of
performance assessment studies, indicators should be formatted identically or analogously
as much as possible (Bos & Nugteren 1990, ICID 1978, Wolters 1992).

Performance Assessment and Potential Indicators for Drainage Systems 8

The indicator must be quantifiable
The data needed to quantify the indicator must be available or obtainable (measurable) with
available technology. The underlying measurements must be reproducible.

Reference to a criterion
This is, of course, obvious from the definition of a performance indicator. It implies that
relevance and appropriateness of the criterion, its target values and tolerances, can be
established for the indicator. These target values (and their margin of deviation) should be
related to the level of technology.

Provide information without bias

Ideally, performance indicators should not be formulated from a narrow ethical perspective.
This is, in reality, extremely difficult as even technical measures contain value judgments.

Provide information on reversible and manageable processes

This requirement for a performance indicator is particularly sensible from the system
manager’s point of view.

Nature of the indicator

An important factor influencing the selection of an indicator has to do with its nature: the
indicator may describe one specific activity or may describe the aggregate or transformation
of a group of underlying activities. Indicators ideally provide information on an actual activity
relative to a certain target value. An indicator can be a direct and/or indirect indicator
(Smedema and Vlotman 1996). A direct indicator preferably is a function of one parameter
only, but should always fully describe the criterion selected for the particular rationale,
without depending on other indicators. For instance, the water table can be fully attributed as
being controlled by a drainage system, even though it is affected by rainfall and irrigation
losses. Soil and water salinity on the other hand can never be a function of the water table
alone, but depend heavily on the water management regime and water qualities. In general
direct indicators are operational indicators, while strategic indicators are generally indirect
indicators. For each rationale a prime indicator can be identified, which is either direct or
indirect.

Ease of use and cost-effectiveness

Particularly for routine management, performance indicators should be technically feasible,
and easily used by agency staff given their level of skill and motivation. Further, the cost of
using indicators in terms of finances, equipment, and commitment of human resources,
should be well within the agency’s resources.

2.2. Potential parameters

The foregoing not only gives guidance for the qualifications of a good indicator but is equally
applicable to the underlying parameters. From research, monitoring and evaluation a long list
of parameters that can help with the formulation of indicators is given in Annex A (Table 7). In
the next section, a list of potential indicators using one or more of the parameters for the
appropriate purpose of the PA is given.

2.3. Selected indicators as function of rationale

A long-list of potential indicators for drainage system assessment is provided in Annex B,
Table 9. This list is the result of a workshop held at the Drainage Research Institute (DRI) in

Performance Assessment and Potential Indicators for Drainage Systems 9

Egypt in 1996 (Smedema and Vlotman 1996). Parameters needed for each indicator are
shown as well.

Indicators from Irrigation Performance Assessment

From the work done by the ICID working group on performance assessment the following
indicators that relate to drainage were proposed (Bos et al. 1994b, Bos 1997):
Depth to ground water
Rate of change of depth to groundwater = new depth – old depth
old depth
Flooding
Impact of Flooding = Area subject to Flooding
Total irrigable area
System Drainage Ratio
System Drainage Ratio = Total drained volume water system
Actually delivered volume into system
Salinity
Relative Change of EC = New EC value – Old EC value
Old EC value
or
EC Increment Ratio = EC downstream of user – EC upstream of user
EC Upstream of user

To judge the performance of a drainage system or irrigation system one needs to have
access to indicators concerning the performance of both. In many cases indicators cannot
stand alone and will only in relationship with others, give a satisfactorily explanation of
observed trends or relative changes (ratios).
To select the most appropriate indicator, the purpose or rationale of the performance
assessment needs to be defined. Is the rationale operational or strategic in nature? Various
rationales are listed in Annex B (Table 9) together with a long-list of drainage system related
potential indicators. In addition the table contains information on whether the indicator is
considered direct or indirect and whether the rationale is considered operational or strategic.
The rationale of performance assessment starts with a question. Then one has to question
whether we indeed are dealing with performance assessment or research, as will be clear
when we consider the typical questions that are asked in relation to assessment of drainage
systems. For example:
3. Does the system function according to design criteria?
4. Is rehabilitation of a system needed?
5. What is the impact of the drainage system on crop yield/farmers income?
6. What is the best drain depth, drain spacing or design discharge?
7. Does this drain envelope work well?
8. Which is the best drain envelope material to select?

Performance Assessment and Potential Indicators for Drainage Systems 10

When we consider each of these from a PA point of view, the following results:
1. Functioning of the drainage system. This is clearly a question that calls for a PA. The
rationale is operational. When complaints about a system reach the appropriate
authorities, requests for PA will be issued. Questions asked will be: does the system
function according to design criteria? According to operational criteria? According to the
agreed service level? Potential prime and direct indicators are:
• Water table draw down rate (surface and subsurface drainage);
• Q-H plots (subsurface drainage);
• Extent and duration of flooding after rainfall (surface drainage primarily).
2. Need for rehabilitation. This also is clearly a call for PA, but it is strategic in nature. Major
indicators can be technical, statistical or financial. The responsible authority needs to
decide whether the system needs to be replaced or that continued maintenance is
acceptable. Prime indicators are:
• Water table depth as function of time (hydrograph, Sum of days Exceeding specified
Watertable depths (SEWxx, where xx is the specified depth)
• Water table depth as function of area;
• Change in number of appropriate complaints;
• Change in maintenance frequency and costs;
• Changes in Manning/Strickler coefficients.
3. Relation to crop-yield/farmers-income and the drainage system. Is the drainage system
responsible for changes in crop yield? If so, to what extent? This is to be considered a
strategic question, but does it require performance assessment? From the system
manager’s point of view: no. From a funding agencies point of view the question is highly
relevant, but it requires more research than performance assessment. It is a comparative
study.
4. Best drain depth, drain spacing or design discharge. The rationale for the PA activity is to
improve the design criteria. The level is operational. But it is not of interest to the system
operator or manager. It is research.
5. Functioning of the drain envelope. Does the envelope perform according design criteria
yes or no? Although a system manager may not be directly interested in this, it could be
the cause of the poor performance of the drainage system. We are caught between a
rock and a mountain. Yes, it is operational performance assessment, but is already at the
next level of PA; cause analysis (Figure 3). Prime and direct indicators are:
• Entrance resistance;
• Water table draw down rate.
6. Which is the best envelope material? Although related to the previous question it should
be clear that this is not PA but research.

Performance Assessment and Potential Indicators for Drainage Systems 11

Table 2: Potential indicators for drainage PA.

Technical / operational Strategic

1. Water table as function of time 1. Frequency of complaints
2. Water table draw down rate
3. Water table depth as function of the area
Direct

4. Q-h plots
5. Water level in canals
1. Relative ground water depth (ICID) 1. Trafficability
2. Ratio of unit discharge to design
Intermediate

discharge
3. Q/H ratio
4. Drainage system discharge as function of
time (q-t, Q-t )
5. Drainage intensity (a)
1. Calculated hydraulic conductivity 1. Sustainability of drained area (modified
compared with design ICID for irrigable area)
2. Head loss fraction (he/htot) over time 2. Frequency, level and costs of
3. Required spacing compared with design maintenance
spacing 3. Weed control intensity
4. Entrance resistance re=he/q 4. Sediment control intensity
5. Relative change of collector drainage 5. Age of structures and frequency of
water quality over time maintenance
6. Sediment in pipes, manholes 6. Relative change of soil salinity
7. Ratio of overpressure in collectors 7. Soil salinity over time
8. Age of the drainage system 8. Relative change of soil alkalinity (SAR or
9. Ratio of design discharge versus actual ESP)
discharge over time 9. Soil alkalinity (SAR or ESP) over time
10. Changes in head-discharge relationships 10. Yield over time
11. Changes in the Manning/Strickler 11. Relative yield change
coefficient 12. Relative cropped area change
12. Rootzone water movement ratio 13. Variability of crops
(leaching) 14. Farm net benefit
15. Net present worth (NPW)
16. Benefit cost ratio (B/C)
17. Internal rate of return (IRR)
Indirect

18. Benefits to the farmers

19. Employment
20. Workability

Performance Assessment and Potential Indicators for Drainage Systems 12

 2.4. Indicators
To achieve the earlier indicated replicability of the indicator, the description of the indicator
should contain the following information:
The name of the indicator and:
name of prime SEW80
1. Whether it is a technical, operational or indicator Operational
strategic indicator? strategic/operational …..
2. The rationale of the PA and the criterion to formula
be used, including target levels and
ranges. rationale Does the system
criterion function
3. An explanation of what is expected that
target level Sum of exceedance xx
the indicator will signify (justification of
use). range Less than ..% of time
..% - ..%
4. Whether it is a direct or indirect indicator?
5. If it is an indirect indicator, which justification of use Indicates potential of
supporting indicators are needed. waterlogging

6. The parameters needed to determine each

direct/indirect Direct at tertiary level
of the indicators.
7. The method of measurement of the supporting indicators none
parameters.
8. The frequency and density (if applicable) parameters needed Elevations and depth
of measurement of the parameters. prime indicator below surface

9. The method (formula) of calculating the measurement Automatic or manual

indicators. method(s);
The following indicators are most generally
used (Vincent 2002): frequency/density daily
parameters
Direct indicators
● Water table depth as function of time
(hydrograph)
● Water table draw down rate
● Water table depth as function of area
● Water level in canals

Indirect indicators
● Evolution of the Manning-Strickler coefficient
● Number and frequency of complaints (note: complaints may not be relevant for the
rationale)
● Q-h plots
● Crops responses (prime indicator)
● Evolution of soil and/or water salinity (prime indicator)
It may be noted that prime indicators can be both direct and indirect indicators. In most cases
they will be indirect indicators when analysis on the data and/or the calculation of the
indicator goes beyond simple trend analysis.

Performance Assessment and Potential Indicators for Drainage Systems 13

 2.4.1. Water table as function of time (hydrograph).
Because water table as function of time requires observation of water table depths in
observation wells or piezometers for a considerable amount of time (preferable 2-3 full cycles
of weather/irrigation/drainage conditions) the indicator is most useful at the Primary
Investigation step of the PA. The indicator is considered a prime and direct indicator as it is
relatively independent of other indicators and can lead by itself to a judgement of the
functioning or effectiveness (efficiency) of the subsurface drainage system. Typically, the
exceedance of a certain predetermined level, during a selected period of time, is the criterion.
Examples of this are given in the Egyptian case study where it was deemed desirable to have
the water table fluctuate predominantly between 0.3 and 0.8 m below the surface.

Parameters to be collected
date of measurement, a depth reading (with unit and reference level indicated), elevations of
measuring point, elevation of surface level at measuring point, and elevation of bottom of the
observation well or piezometer, elevations of invert of drains, number of days since last
irrigation in fields adjacent to the observation point. In addition it is advisable to determine its
location in an x-y co-ordinate system, which will be useful for area assessment of the extent
of waterlogging; however this is not needed at tertiary level.

Method(s) of measurement
Water levels can either be determined by automatic water level recorders which will give a
continuous, or close to continuous measurement (depending on interval of measurement set
when dataloggers are used), or can be measured manually. In the latter case discrete
measurements in time are obtained. Generally the most frequent observation is daily, but
intervals matching irrigation cycles (5, 7, 10, 14 days etc.) are common, while, depending on
man-power and funding, monthly or 6 monthly (or seasonal), observations are also used.
Oosterbaan (1994) observed that the water table mean values of daily and monthly periods
(or even longer intervals) coincide, but the standard deviation of the monthly (longer) value is
much smaller than that of the daily average. The longer interval observations tend to be less
skewed and hence the mean value represents the frequency distribution better. These
observations, however, could not be confirmed from long-term data from Egypt (Vlotman and
Eissa 2000, see also Part B, section 2) It should be considered whether these long-term
interval means correlate better with crop response than the short interval means.

Frequency and density of measurement

The frequency of measurement depends on the purpose of the indicator: if it is to be used to
obtain a general impression of the water table fluctuation in the area then fortnightly or
monthly observations may be sufficient. If the objective is to determine the response of the
drainage system (i.e. draw down curves see indicator 2.4.2), then continuous or hourly
observations may be necessary for a statistically adequate number of sets to represent the
area of interest. To determine Sum of days Exceeding specified Watertable depths (SEWxx),
where xx is the specified depth, daily measurements are desirable.
For area assessment of the water table (see also indicator in section 0), the density of
measurement will be determined in most cases by topographical features. A grid of 500 x
500 m works well in relatively flat areas. Where major drains and irrigation canals cross the
network needs to be intensified locally to assess the effect/influence of the canal on the water
table. For drainage system and drain line (lateral) assessment, it is common to install
observation wells midway between parallel drains and density will be a function of drain
spacing. FAO guidelines (Dieleman and Trafford 1976), which were primarily based on
experiences from temperate humid regions, suggest three sets of observation wells along a

Performance Assessment and Potential Indicators for Drainage Systems 14

drain line. This works well with uniform rainfall or irrigation over the area, but not with non-
uniform irrigation. In the latter case a randomly distributed grid of observation wells is better1.

Method of calculating and using the indicator

Most commonly the water table depth is converted to water table elevation and plotted
against time, setting the vertical axis (elevation) such that the top of the graph represents the
surface elevation. In that case both an impression of depth and elevation is obtained. If
certain dips or rises need to be explained the indicator needs to be combined with rainfall
and irrigation events, as well as the cropping pattern. If funds allow daily observation is
recommended. With daily observations also, monthly average, seasonal average, critical
growth period average, or exceedance of certain levels, can also be determined.

Justification
To judge the functioning of the drainage system, water table depth (elevation with respect to
MSL) will allow us to judge whether the design water table depth is achieved for most of the
time and during critical growing periods of the crop. If the frequency of observation is
sufficient it will allow us also to determine the number of days it takes to reach the design
depth after an irrigation (or rainfall) event (see indicator 2.4.2). Continuous or long term
observations will allow us to determine the periods of high and low water table depths. The
trend of the water table over time will allows us to judge whether there is a gradual decline
(or improvement) in the functioning of individual drains and/or the system if a number of
observation wells that represent the system is analysed.

Indicator range
To avoid waterlogging (aeration problems to the crop) water table depth should not be
shallower than 1.0m for more than 3 - 5 days after an irrigation or rainfall event. In Egypt the
root zone typically extends to 0.6 - 0.8 m below the land surface and a water table 1 m below
the surface for 3 days is no problem for crops with those rooting depths.

2.4.2. Water table draw down rate.

One of the more direct indicators on how a particular subsurface drainage line is performing
is to study the water table draw down rate.
name of prime indicator Draw down rate
strategic/operational Operational
formula …..
Rationale Does the system function according design
Criterion Below design depth 6 days after recharge event
target level In 6 days at design depth
range Plus-minus one day
justification of use Indicates adequate water table lowering
direct/indirect Direct
supporting indicators none
parameters needed prime indicator Elevations and depth below surface, date
measurement method(s); Automatic or manual
frequency/density parameters Daily or twice daily. Continuous when automatic.
Select a reading from the continuous
observation

1
This recommendation is for indicator 2.4.1 and 0 only, and does not apply when site
specific information is needed, such as Q-h relations or draw down curves.

Performance Assessment and Potential Indicators for Drainage Systems 15

Parameters to be collected
The parameters to be collected, their frequency, density of measurement are the same as for
indicator 2.4.1

Method of calculating and using the indicator

A draw down curve can be divided into three parts (Figure 5): the rising water table, the fast-
falling part and the slowly falling part. The rate for each part or several parts together can be
determined. If needed, this rate can be extrapolated when no records over a sufficient length
of time are available. A sufficient amount of time is determined by the criterion. For instance,
if the criterion specifies that X days after the recharge event the water table should be at
depth Y, then the duration of measurement should be X+ 2 days for instance.

Justification
To judge the functioning of the drainage system, or the efficacy of a certain drain envelope,
the draw down rate of a number of observation wells in the area, is good direct indicator.

Indicator range
See also the Egypt case study.

Tangent 1

Tangent 2
Falling Falling
part 1 part 2

Figure 5: Principles of Drawdown Ratio Determination

2.4.3. Water table as function of area.

This indicator has been used traditionally in Egypt, and is also promoted by the World Bank.
However, it suffers, at the moment, from clear definition of the period during which the
observations should be taken, and the area to be considered is not clearly defined. The
indicator is considered a direct indicator.
The parameters to be collected and the method(s) are the same as for indicator 2.4.1.

Frequency and density of measurement

The frequency is subject to similar considerations as described for indicator 2.4.1. The
density of measurements is, however, subject to further consideration for Egyptian

Performance Assessment and Potential Indicators for Drainage Systems 16

conditions. EPADP collects pre-drainage investigation data at a grid of 500 x 500 m. DRI
collects their data as function of the lateral spacing, usually for research purposes rather than
for PA. Further work to determine the best grid distances for certain conditions and
requirements are subject for further study by DRI. The M&E Project at EPADP proposes to
relate measuring locations to collector lay-out.

Method of calculating and usage of the indicator

All available data sets can be used, when the Global Positioning System co-ordinates are
known for all data points. Contour plots can be made manually or generated by computer
programs and frequency distributions of selected classes of water table depths can be
calculated. Time series will indicate shifts from one class to the other and this helps the
decision maker to decide on the effectiveness of certain measures, or the seriousness of the
problem. All data points should be related to the layout of the drainage system and plotted on
the 1:10000 design maps. The commercial computer program Surfer allows calculation of
areas between certain contour intervals in a somewhat round about way, but at an affordable
price. More sophisticated programs like ArcView and others allow calculation of areas based
on contour plotting too, but the learning curves of those programs are higher as is the cost.
Use of a planimeter is still quite acceptable and by far the cheapest from an acquisition point
of view.

Justification
This indicator indicates the magnitude of the problem observed, or the effectiveness of
certain measures. To judge the functioning of the drainage system this indicator is essential.

Indicator range
Some tentative values have been presented earlier such as (midway) water table depth less
than 0.8 - 1 meter below ground level in more than 10% of the survey area, or a water table
less than 100 cm depth below soil surface in at least 75% of the area (Small, 1992). It may
be noted that present EPADP design criteria for subsurface drainage systems merely
prescribe that the average water table depth should remain below 1 m.
Examples. See case studies of Pakistan, Egypt and India.

Number and frequency of complaints:

Experience with complaint assessment was gained on one of the projects in Egypt (see Part
B, section 3 for details). The process to determine the indicator is similar to that of the
others:
● parameters to be collected.

● Method(s) of measurement.
● Frequency and density of measurement.
● Method of calculating and using the indicator.
● Justification.
● Indicator range.

2.4.4. Q-h plots

This is a traditional indicator but in many cases the interpretation of the data will be difficult
as was shown in the first case study from Egypt (Section 2).

Performance Assessment and Potential Indicators for Drainage Systems 17

Parameters to be collected
Midway water table depth (one reading or average of several along the drain line), discharge
at the same time, general description of irrigation events along the drain line, or rainfall
information. Free flow or submerged condition at the location of discharge measurement.

Method(s) of measurement
The water table depth can be determined from piezometers or observation wells, either
manually or automatically. Levels are to be converted to MSL preferably, so determination of
elevations with respect to a nearby datum point is essential. Discharge can be measured,
with bucket and stopwatch, with weirs, or with a current meter. The latter is not
recommended for pipe flow measurements.
All measurements should be collected at the same time or the discharge measurement
should be as close as practical after the time of the head observation. In heavy textured soils
the time between the measurements may be several hours, but in light textured soils this
should be within the hour. In theory the travel time of the water in the drain is an indication of
the theoretically allowable delay between the head measurement along the line and the
discharge measurement at the end.
Maps indicating field sizes and cropping patterns along the drain should be made each
season. Fields irrigated should be indicated on this map for each Q-H measurement.
The Q-H assessment originated from rainfall zone where uniform rainfall along the length of
the drain line occurs. Under irrigated conditions this indicator is more complicated to use.
(Eissa, Hoogenboom et al. 1996) proposed methods to take into account the irregular
irrigation pattern along the drain line.

Frequency and density of measurement

There is no real frequency to be recommended and density of measurement is not applicable
in this case. To get a good graph from which conclusions can be drawn one should get a
good set of observations at low, medium and high discharges, for various seasons.

Method of calculating and using the indicator

Convert the water table depth readings to elevations and then calculate the pressure head
with respect to the centreline of the drain in the case of drain diameters up to 80 mm, and
with respect to the invert of the drain at the location opposite the water table measurement
for drains with larger diameters. Convert the discharge measurements to mm/d by dividing
the discharge with spacing and length on the drain. Plot all data in an X-Y plot and draw also
the expected relationship according the drainage formula used for the design (i.e.
Hooghoudt, Glover-Dumm, etc). For hydraulic conductivity use initially the design
parameters, but it is probably better to obtain the hydraulic conductivity of the soil between
the water table observation well and the drain line. Other parameters used in the design such
as equivalent depth, depth to impermeable layer, etc. should be used. Based on the
dimensions of the drain line and its slope one could also estimate the maximum expected
discharge using the Manning equation.

Justification
Under uniform rainfall or irrigation conditions the indicator will give an idea of whether the
system functions as expected. Under irrigated conditions, it is rare that along the complete
length of the drain irrigation application is uniform. Thus, a large scatter of the data will occur
from which little or nothing can be concluded. Unless uniform recharge conditions during
measurements exist, this indicator is not recommended for use.

Performance Assessment and Potential Indicators for Drainage Systems 18

Indicator range
The data should be clustered around the theoretical Q-H curve from the drainage formula
used in the design, and below the maximum expected or possible discharge through the
drain line when the pipe is just flowing full and no overpressure is taking place.

2.4.5. Water level in canals

The maintenance of drainage canals is a major concern of managers. Vegetation growth and
siltation affect the hydraulic characteristics of the structure. A consequence of this is a rising
of the water level in the canal resulting in submerged out-falls of laterals or worse, blockage
of laterals by sediments. Thus the observation of the water level in drainage canals is a key
indicator in the performance of drainage systems. The canal water level has to be related to
rainfall or irrigation events or to discharge measurements, to determine whether operation is
within the acceptable range or whether maintenance is required. Observation points may be
located in the vicinity of structures across the canal (bridges, culverts, etc.) or at any point in
a reach located away from any distortion due to structures etc. If all the observation points
are topographically referred to a reference plan, interpretation can be made in terms of water
level and the indicator becomes a screening indicator as well. Where the water level rises
over time for the same flow conditions it can be suspected that the canal requires
maintenance.

Parameters to be collected
Date of measurement, depth reading (with unit and reference level indicated), elevation of
measuring point, elevation of surface level at measuring point, elevation of bottom of canal at
the installation, elevation of bottom of the canal at the date of the observation, distance to the
origin along the canal, date of rainfall event, amount of water during this event, number of
days since last irrigation in fields adjacent to the observation point. In addition it is advisable
to determine its location in an x-y coordinate system, which will be useful for assessment of
command area affected by the canal performance and for maintenance planning.

Frequency and density of measurement

There is no real frequency to be recommended. Measurements have to be carried out during
rain events or at representative periods of the irrigation season.

Method of calculating and using the indicator

Compute the Manning Strickler coefficient, convert the canal water depth readings to
elevations and then compare to elevation obtained for similar irrigation or rainfall events;
compare to the design elevation. If discharge measurements are available, plot
discharge/water depth curves. Draw the water level for the different events; detect the
sections were the water levels are rising over time and any structures which induce high
head losses. Refer the water level to the position of the bottom of laterals and determine the
flow conditions at the lateral (free flow or not).

Justification
Siltation, vegetation, clogged or under designed structures across the canal or any other
obstacle raises the water level in the canal. Under uniform rainfall or irrigation conditions the
indicator, or combined with discharge measurements, will give an idea whether the system
functions as expected. Only when uniform discharge conditions during measurements exist,
is this indicator recommended for use unless the water level between representative
measurement points can be established.

Performance Assessment and Potential Indicators for Drainage Systems 19

Indicator range
The data should be clustered around the water level computed for the discharge rate. For
climates with rainy seasons, the discharge may exceed the design discharge if the canal
plays a role in transiting water caused by surface runoff.

2.4.6. Evolution of the Manning-Strickler Coefficient

As the vegetation grows in the canal, the roughness of the banks increase, and the Manning-
Strickler Coefficient decreases. Thus this Manning-Strickler coefficient should be an indicator
of the performance of the canal.

Parameters to be collected
The same parameters as for the indicator water level in the canals that have been measured.
In addition, discharge measurements have to be carried out for the associated water level
measurements.

Frequency and density of measurement

There is no real frequency to be recommended. Measurements have to be carried out during
rain events or at a representative time after irrigation events.

Method of calculating and using the indicator

Compute for different reaches and for a wide range of discharge and/or water levels the
Manning Strickler coefficient from the formula:

Q = K sS I R 2H 3
Q = discharge (l/s)
K = Manning Strickler coefficient
I = Slope of the bottom of the canal
S = Wet section (m²)
R = hydraulic radius (m)
A restricted number of measurements can be considered if the measurement point is
upstream far away from a hydraulic singularity or structure in order to avoid downstream
control. If a free flow model is available (i.e. calibrated and validated), Manning-Strickler
coefficient/water height curves can be plotted (Figure 6).

Performance Assessment and Potential Indicators for Drainage Systems 20

 1,8
Vegetalised ditch
Water height (m)
1,6
Water height
1,4

1,2

1
Low vegetalised d itch

0,8

0,6

0,4

0,2

0
0 5 10 15 20 25 30 35
Manning Strickler coefficient

Figure 6: Water height variation in a trapezoidal canal as function of the Manning Strickler
coefficient, uniform regime

The waterline considerably rises considerably with the diminution of the Manning Strickler
coefficient from 33 (new earth ditch) to 5 (highly vegetated canal).
● For a Manning-Strickler coefficient of 15, the water line rises 50 %,

● For a For a Manning-Strickler coefficient of 5, the water line rises 150 %.

● Manning-Strickler coefficient of 10, the water line rises 80 %,
● For a Manning-Strickler coefficient of 5, the water line rises 150 %.

Justification
If the measurement point is far enough upstream from a hydraulic obstruction, a uniform
regime can be considered for a given discharge. The Manning-Strickler formula applies and
the Manning Strickler Coefficient can be computed

Indicator range
The data should be included between very low values and 33 which is the value for a new
earth canal.

2.4.7. Crops responses

Crop response is clearly the best measure in terms of Performance Assessment for drainage
systems. The crop response is usually expressed in terms of yields and thus the
measurement has to refer to the state-of-the-art agronomic science. It is not our purpose
here to provide a review of yield measurement methods. Nevertheless, the following points
have to be considered as soon as crop responses surveys concern a PA programme:
● At the plot scale, yield measurement is meaningless without measurements of yield
components. Yield measurement is a very comprehensive measurement which integrates
all the factors affecting a cropping season. A yield value may be the result of the action of
various limiting factors at various phenological stages of the crop. Thus the observed yield
value is not necessarily linked to the limiting factor under concern and so yield component

Performance Assessment and Potential Indicators for Drainage Systems 21

 measurement is the only way to link a limiting factor to a phenological stage and finally to
the yield.
● It is not possible to improvise an agronomic protocol. Agronomists (and preferably local
ones) have to be involved in the PA program, including the monitoring of crop responses.
Examples of crop yields monitoring are given in case studies 5 and 6 in Part B.

2.4.8. Salinity control

In arid and semi-arid irrigated areas, drainage should provide control of waterlogging and
salinity in the root zone. Performance assessment with regard to waterlogging is reasonably
straightforward but not so with salinity.
The assessment with respect to salinity has to be characterized based on a system that
includes the irrigation system, the drainage system, and the crop. Since water is the salt
transport medium there has to be an adequate amount applied uniformly and there has to be
adequate capacity to remove the excess water applied to leach the salt from the soil profile.
Therefore, the aim of salinity control is to maintain and/or reduce the salinity in the crop root
zone to acceptable levels relative to crop tolerance, to do this as uniformly as possible, and
to minimize the mobilization of salt from below the root zone.
The other consideration is whether the area being assessed is undergoing reclamation or is
in full production. There would be different criteria developed for these conditions. The
following discussion assumes that the area is in full production.

Parameters to be collected
1. Bulk soil electrical conductivity is the main parameter to be collected. It is recommended
that this be measured using an electromagnetic induction (EM) method to facilitate rapid
collection and analysis. This can be done with an EM-38 or similar equipment. This
assessment should be aimed at the root zone, therefore shallower readings are more
relevant e.g. using the EM38 in the horizontal mode gives a better reflection of shallow
soil depths. A detailed description of use and calibration of EM equipment can be found in
Rhoades et al. (1999).
If bulk electrical conductivity measuring equipment is not available then soil sampling
alone must be used. In this case the intensity of sampling is much reduced and so the
variability/uniformity of soil salinity is difficult to assess.
2. The EM measurements need to be correlated to electrical conductivity data measured on
saturation extracts (USSL, 1954) from soil samples taken to the maximum depth of the
main root zone of crops in the rotation. The soil sampling locations should be determined
based on statistical analysis of the data collected during the EM survey. This can be done
with the ESAP software (Lesch, et al., 2000). The bulk soil electrical conductivity and
sample locations may be geo-referenced using a global positioning system (GPS), or
logged manually where this is not available.
3. The crop rotation must be quantified.
4. The electrical conductivity of the irrigation water should be measured.

Frequency and density of measurement

1. The bulk soil salinity should be measured annually prior to the major cropping period,
generally the spring. Ideally the soil water content at this time should be relatively high
and uniform (Lesch, et al., 1998). The density of measurements taken will be determined
by the equipment available. The measurements should be in the form of a mapping
survey, usually in a grid pattern, in order to capture variability in soil salinity within a field
or drainage system.

Performance Assessment and Potential Indicators for Drainage Systems 22

 Mobile EM systems with GPS units are capable of taking data on a small grid in a short
period of time while manual collection is done a larger grid size to save time. The actual
grid size needs to be determined based on the field size and any prior knowledge of the
salinity distribution. Secondary indicators such as crop health, drain position, slope, soil
changes and the presence of salt on the soil surface may be used in determining the grid
spacing.
Particular care should be taken when mapping horizontal pipe drainage systems to
ensure readings are taken perpendicular to the drains at intervals of half or quarter the
drain spacing in order to reflect the variation in salinity with distance from the drains,
Rhoades et al. (1999) gives examples of this.
2. The soil sampling density (number of holes) for calibration of the EM is based on a
statistical analysis of the EM data. A 1.0 m sampling depth is usually adequate to
characterize the root zone of most crops.
3. If there is reason to believe that the irrigation water quality changes over the irrigation
season, then the water quality data should be taken at each irrigation. Otherwise, only
periodic sampling is needed.

Method of calculating and using the indicator

The main indicator will be the bulk soil salinity as a saturated paste value. This value will be
determined by calibration of EM readings to saturated extracts measured on soil samples.
This can be done with the ESAP software (Lesch, et al., 2000).
Soil salinity, as an indicator can then be plotted as contour maps to determine the spatial
distribution, see Rhoades, et al. (1999). This mapping can be done using software such as
SURFER (1986) or manually. The maps should have the tile lateral and main locations and
other important features such as irrigation channels marked on them to identify areas that are
receiving inadequate leaching water or have reduced drainage capacity as evidenced by
higher salinity values."
The highest salinity values would be expected at the mid-point between drain laterals and at
the tail end of surface irrigated fields. Irrigation uniformity may be characterized by variations
in the salinity particularly between the head and tail end of the field.
This analysis of the variability of salinity is particularly useful in that often there may only be
localised problems within drainage systems. These problems may be caused by
maintenance problems e.g. a blocked drain/failed pump, or they may relate to inadequacies
in the design of the system, e.g. one part of a field may have a much less permeable soil that
was not detected during the investigations for design. Furthermore the analysis of variability
can assess other localised problems such as seepage from irrigation channels or non-
uniform distribution of irrigation water, both in field and between fields.
For an overall assessment a median value for salinity should be calculated for the field or
area. The mapped salinity may also be used to develop classes of salinity that may be
compared over the years. These values can then be used in subsequent years to determine
if salinity is being controlled across the field or area with time.
A secondary indicator is changes in cropping pattern to more salt sensitive crops, as this is
an indication of reductions in salinity in the root zone.

Justification
Bulk salinity that characterizes the root zone salinity over the long term is the only measure
that is indicative of the operation of the system. The spatial and temporal distribution of EC
will give an indication of either the lack of water for leaching or the reduced function of the
drainage system. The drainage water salinity (sometimes regarded as the leaching fraction)
is not an adequate parameter since salt may be mined from the region below the root zone

Performance Assessment and Potential Indicators for Drainage Systems 23

giving high drainage water salinity. This high drainage water salinity may be interpreted as
leaching of salts from the root zone and hence maintenance of a salt balance, however
subsurface drainage systems move salt from well below the root zone depth and so the salts
in the drainage water cannot automatically be assumed to have been removed from the root
zone.

Indicator range
The range of soil salinity will be a function of the salt tolerance of the crop being grown on the
field. These values may be determined from the literature, e.g. Maas and Hoffman (1977).
The maximum value to sustain production will be the threshold value at which yield
reductions begin.
Irrigation water quality will influence the soil salinity and must be considered when evaluating
the long term response of salinity in the profile. The bulk salinity should be compared to the
salt tolerance values for crops in the rotation and assess whether the overall system
composed of the irrigation and drainage components is maintaining the root zone salinity at
levels that don’t limit production. Theoretical calculations of the soil salinity with adequate
leaching can also be made (Hoffman, 1990) or calculated using programs such as WATSUIT
(Oster and Rhoades, 1990) and compared with the measured values, (Rhoades et al., 1999).

2.5. Reliability and usability of the Performance assessment results.

To achieve certain levels of accuracy or reliability of the Performance Assessment it is
necessary that a certain number of indicators be used for certain levels of adequacy in
explaining the changes. To put it differently: to explain the change in a criterion (i.e. yield)
adequately, say 75% accuracy, or completely (100%) a certain number of indicators, or all
indicators are necessary. Most often, when the question of why a certain criterion is changing
is raised, not all data will be available (immediately) and it is then necessary to indicate
qualitatively, the accuracy or reliability of the Performance Assessment in explaining the
change.
The prime indicator here is defined as the one indicator that signals that something is
changing noticeably such that decision makers request explanations. A prime indicator
should not be confused with a primary or direct indicator as defined before. A prime indicator
for a certain rationale can be an indirect indicator (or secondary indicator). For instance
changes in yield or crop response may be considered a prime indicator but is by no means a
direct indicator. Indicators qualifying the water management and crop management (fertiliser,
pesticide applications, planting dates, variety etc.) are needed as well.
The number of indicators will vary, but to be classified as a set that can give the best results
to meet the objective of the PA the set needs to be able to explain the objective or the
change in the prime indicator for 100%. The very minimum is the use of one indicator (Figure
7). So far, only for the functioning of the subsurface drainage system a one parameter “set”
explains the functioning 75 - 100%. In this case one may argue about whether the adequacy
of the Performance Assessment is 75% or 100% based on the one indicator assessment.

Performance Assessment and Potential Indicators for Drainage Systems 24

 Levels of Accuracy/Reliability of Performance Assessment (PA)

• 1 (minimal) • 2 • 3 • 4 (best)
• only one (prime) • prime indicator + • 90% of the • all indicators listed
indicator 50% of indicators indicators listed for for the rationale are
i.e. listed in Annex B for particular rationale available
selected rationale are available.
Yield
Salinity • can explain 100% of
• can explain 50% of • reasonable accurate the change in the
Water Table the change in the water balance can prime indicator
prime indicator not be made.
• can explain 75% of
the change of the
prime indicator

Minister Drainage Departments Research organizations

Country Government
and Operational &
Province/state Local government
System Management
Watershed/project
Drainage/irrigation system Research & executing
Field Research Organisations

Figure 7: Level of accuracy/reliability of Performance Assessment.

The best level of PA is achieved if all indicators listed for the rationale are available. This
however, will seldom be the case. Possibly a fully controlled experiment may achieve this
level. Generally one or more of the indicators will be lacking. Two additional levels have been
arbitrarily determined (Figure 7): Level 2 is achieved when the prime indicator and 50% of the
indicators listed in Annex B are available. It is assumed that these indicators can explain 50%
of the objective of the PA. This is a rather subjective statement, but it is important that we
assign some adequacy rating to the PA. Level 3, the one below best, is achieved when 75%
of the objective can be met with the available indicators. The most common indicators lacking
at Level 3 are those that relate to producing a closing water balance (one or more
parameters for calculating indicators related to drainage, irrigation, pumping, evapo-
transpiration, etc. are not available). When the water balance is not accurate (this can
possibly be expressed using standard error assessments) the level can be maximum Level 3.
When there is a relatively high degree of accuracy achieved in the water balance
calculations, the level of accuracy of the PA is greater than Level 3 but perhaps not yet Level
4. In those cases assigning values of 3.25, 3.5 and 3.75 could be considered.
Level 1 of PA adequacy is the one most commonly available initially and typically at high
government levels when questions are first raised; i.e. yields are dropping but the why cannot
be adequately explained. The government will respond with requesting more details, which
will require further investigation and PA which may lead to Level no 2 type of PA. This level
may be achieved after a preliminary investigation has been performed such as indicated in
Figure 3. This, for instance, could be the type of PA that a drainage department would
require in order to make its planning decisions, or it could be PA needed for Operational and
System Management. Level 3 and 4 are typically the type of PA performed by research
organisations, but preferably to be used by local, provincial and national governments as
well. Level 3 may be achieved after the Primary Investigation has been completed. A Level 4
PA is likely to be achieved only after extensive research. Note that although the area
encompassing the PA (country, province/state, watershed, project, drainage system, or field)
is only indicated with level 3 and 4 of the PA (Figure 7) it may apply to all levels. Depending
on the number and level of accuracy as well as reliability of the quality of the data collected
several classes of data may be distinguished as indicated in Figure 7.

Performance Assessment and Potential Indicators for Drainage Systems 25

3. Benchmarking Drainage Systems in IWRM
Benchmarking is a relatively new phenomenon in IWRM and has not yet been introduced to
the drainage scientists peer group in this format. Yet it is not new, as PA and other types of
assessment have been around for many years. What is new is to formulate a well-defined
process of benchmarking of irrigation and drainage systems, using the latest Information and
Communication Technologies (ICT) available. The benchmarking activity will hopefully lead
to the establishment of indicator and target values for well-specified conditions. The lack of
comparative data has been the main drawback of Performance Assessment to date (very
similar to problems experienced with benchmarking in the business world and at government
levels). The emergence of ICT (incl. remote sensing and GIS applications) has given a new
vigour to the PA activities (incl. private business comparison and service oriented utility
companies comparisons; e.g. in the Dutch drinking water industry, (VEWIN 2001), which may
be extended to include drainage systems. The TF4 has made a start with on-line
benchmarking and uses the guidelines of Malano and Burton (Malano 2001) as its basis. In
order for drainage to become part of benchmarking the key (water management system)
descriptors shown in Table 3 need to be included.

Table 3: Descriptors for benchmarking water management systems

Descriptor Possible options Explanatory notes
Drained area Area with gravity drainage system. Defines the magnitude of drainage
Area with controlled (incl. pumped) systems that are part of the water
drainage management
Allows qualification of size of drainage
systems
Type of drainage Gravity; controlled; pumped. Affects the potential level of
Surface; subsurface (hor.); vertical performance.
Type of Government agency; private Affects the potential level of
management of company; joint government performance. Note that because
drainage system agency/farmer; farmer-managed, or drainage and irrigation are often still
any other appropriate descriptor. separate entities this information is to be
stored for both drainage and irrigation
separately also when in one
organisation

To quantify the performance of drainage systems in the framework of benchmarking the use
of the following indicators are proposed (Table 4).

Performance Assessment and Potential Indicators for Drainage Systems 26

Table 4: Proposed indicators for benchmarking the performance of drainage systems
Domain/rationale Performance indicator Explanatory notes
Service delivery Total annual volume of drainage
performance water removal (m3/year, m3/ha)*
Total annual volume of drainage NB Water Boards in the Netherlands
water treatment for reuse (m3/year, include sewage water treatment in
m3/ha)* their package of water management
Drainage ratio* Indicates efficiency and degree of
control.
Financial Drainage system investment, Standard financial indicators include
operational and maintenance costs. investment and operational cost. No
additional indicator needed but
terminology presently used should
include drainage, not just irrigation:
use water management.
Productive efficiency No additional indicators; costs of
drainage to be added to irrigation
costs, incl. water treatment costs.
Environmental Water quality** (irrigation, Separate quality reporting per type
performance drainage*): Salinity (mmhos/cm) of conveyance system.
Water quality** (irrigation, Separate quality reporting per type
drainage*): Biological (mg/litre) of conveyance system.
Water quality** (irrigation, Separate quality reporting per type
drainage*): Chemical (mg/litre) of conveyance system.
(Average*) depth to groundwater SEWxx indices, where xx denotes
(m) the target depth in cm (see Annex 2)
Change in water table over time (m) Trend analysis
Spatial distribution of water table Degree of waterlogging
depth*.
Salt balance (tonnes) Indirect indicator due to dependence
on irrigation and salt concentration
in different waters.
* drainage related additions to the list produced by Malano and Burton (2001) and not yet
included in current ICT implementation of benchmarking at the IWMI website.
** water quality contaminant groups need to be reformulated using standard water quality
terminology used by WHO, EPA’s , to accommodate use with IWRM; see also text below.
SEW = Sum of Exceedance of Water table (elevation)

In the foregoing the current layout of the benchmarking indicator list has been followed as
much as possible. The asterisk indicates where a change or addition is proposed. The
proposed indicators are primarily operational and strategic. It should be noted that the
environmental performance indicators, in particular that of water quality, need further
elaboration, perhaps following the typical division of groups of contaminants of water quality
standards of the drinking water industry, Environmental Protection Agencies (EPA) and the
World Health Organisation (WHO). This will enhance synergy between different disciplines
dealing with water quality standards in IWRM and assist with providing guidelines for
drainage and sewage water reuse. Depending on the use of the water for ecology,
agriculture, livestock, swimming, drinking, industrial, the indicator target values will be set
differently (depending on the rationale). Indicator values suggested are given by Kielen and
Martinez (Kielen 2002) CV 2000 and may be found on the US EPA and WHO websites.
Benchmarking these data will assist greatly in eventually setting (more) realistic target values
compared to the rather conservative values recommended presently. The groups of
contaminants are:

Performance Assessment and Potential Indicators for Drainage Systems 27

● Physical parameters, such as colour, taste odour, turbidity;
● Micro organism (bacteriological), covering most of the human and animal fecal waste.
● Disinfectants and disinfectant by-products (i.e. chlorine and others) used for drinking water
disinfection and additives used to control microbes;
● Inorganic contaminants, ranging from the heavy metals to salts, nitrate, phosphorous,
selenium, etc. expressed as indicator ratios e.g. SAR, SARadj, BOD, COD, pH, etc.;
● Organic contaminants, which include most of the pesticides.
Benchmarking of irrigation and drainage is in its infancy. Yet fast experiences on the
performance of drainage systems exist but because of lack of a unified reporting system, the
data may be hard to compare, while the format in which the data are presented may also
inhibit use. This is further compounded with issues of reliability of the data. Regarding the
achievement of drainage projects it might be considered that operational drainage PA
programmes are scarce. Experiences are known in Australia, Egypt, France, The
Netherlands, India, Morocco, Pakistan, and Tunisia.
In France PA programmes were linked to drain spacing assessment and to economic
aspects (Zimmer 1991; Zimmer 1996). The work used to be carried out at the scale of control
plots. Regional approaches were scarce and reporting is limited to grey literature. Other
aspects that can be pointed out from the experience gained in France during the 80’s are the
following: (1) farmers usually have a good knowledge of the efficiency of their drainage
networks and relevant PA programs could start with farmer enquiries; (2) PA programs
benefit much from R&D programs and also from farm monitoring programs. Research results
may be important for drainage performance assessment since they provide a general
framework for interpretation of the field observations. Having a field experiment included in
the program is very helpful since it is possible to investigate more precisely the effects of
drainage; (3) the sampling of subsurface drainage systems and farms to be investigated in
the PA program is very critical. Although still under development, remote sensing techniques
might be very useful for that purpose (Ferrante 1995; Vincent, Vidal et al. 1996).
In the Netherlands, observation wells networks exist but are not organised in a drainage PA
program. Several networks such as The Waterboards report their results of water
management and flood protection in Annual Reports that more and more follow the
guidelines of accounting standards and from these standard ratios such as suggested in
Malano and Burton (Malano 2001) for financial benchmarking may be obtained. The Dutch
Agricultural Economics Institute recently issued a report (Diederen 2002) which highlights the
costs of water management in the Netherlands including flood protection. There is lots of
benchmarking data available in the Netherlands but it is not in a format readily useable with
the current I&D benchmarking of ICID TF4.
In Tunisia, all the irrigated areas are equipped with observation wells (grid 500*500 m), but
interpretation is difficult because of the variable number of piezometers due to degradation
(Vincent 2001) and thus the only option is to analyse for long term trends (Hachicha 2001). In
Morocco, very good results were achieved, but this has remained at the research level.
In Pakistan extensive research and work has been done on drainage projects constructed
over the last two decades. In particular, the Fourth Drainage Project has been reported on
extensively (Vlotman 1994; Vlotman 2002) but as is the case with all these works the
reporting format does not lends itself easily for inclusion of the data in the benchmarking
database. Nevertheless the guidelines prepared by the Working Group on Drainage used the
results to formulate part of the recommendation reported herein.
Australia is one of the case study areas of the TF4 project and a programme has been
underway in Australia initiated by the Australian National Committee of ICID (ANCID) since
1998 (ANCID 1999; ANCID 2002). The first benchmarking report for 1997/98 reported on 33
irrigation systems and used 15 performance indicators. The 1998/99 benchmarking report
reported on 46 systems and used 47 performance indicators. The 2000/01 report covers 40

Performance Assessment and Potential Indicators for Drainage Systems 28

irrigation water providers with 65 indicators (12 on system operation, 14 on environmental
issues, 25 on business processes and 14 financial indicators). Actual data are provided on a
CD-ROM including the data of the previous 2 periods. To describe the water management
systems 67 descriptors are used, of which approx. 15 apply to the drainage system, Table 5.

Table 5: Drainage System Descriptors used by ANCID.

Descriptors Surface Drainage Descriptors Subsurface Drainage (SSD)

Area serviced in ha Area serviced in ha
Portion of irrigated land serviced by surface Portion of irrigated land serviced by SSD in %
drains in %
Area not drained but requiring drainage (ha) Area not drained but requiring drainage (ha)
Number of drainage customers Number of drainage customers
Type of surface drains Type of collection system from SSD
Length of open drain and waterway (km) Number of groundwater level bores monitored
Drainage Pipeline length (km) Average depth to water table (m)
Where surface drains outfall Where sub-surface drains outfall
Number of groundwater pumps
Outfalls monitored for flow, % of total Outfalls monitored for flow, % of total
Outflows monitored for quality, % of total (salinity, Outflows monitored for quality, % of total (salinity,
nutrients) nutrients)

Measures in place to offset downstream impacts Measures in place to offset downstream impacts
of drainage (pay, dilute, works, other) of drainage (yes/no)

Indicators that pertain to drainage and water quality in the 2000/2001 report are shown in
Table 6. It may be noted that a rather large number of the indicators listed are primarily
descriptive in nature, with few giving actual values. Hence a more consistent separation of
qualitative and quantitative indicators is recommended. Table 5 and Table 6 were derived
from output tables rather than input data sheets, where this separation would have been
more evident perhaps.

Performance Assessment and Potential Indicators for Drainage Systems 29

Table 6: Indicators pertaining to drainage as used by ANCID

Domaine/Rationale Indicator
Operational None listed, although 28% of outfalls were monitored for flow of the
surface drainage system and none of the SSD. The indicators below
were given as part of the environmental issues, but are elsewhere
commonly reported under operational performance assessment
indicators.
Environmental Environmental issues 1, 2 and 3 (e.g. salinity, water table within 2 meter
of surface, soil acidity, vegetation decline, etc.)
Environmental Management Plan, EMP (y/n)
EMP part of business plan
EMP linked to regional strategy
EMP linked to Catchment Action Plans
Progress towards ISO 9000 certification
Progress towards ISO 14001 Certification
Groundwater data density (irrigated hectares per bore)
Average depth to water table
Portion of irrigation area with water table less than 2 m depth
Median groundwater salinity (EC)
Salt load in (tonnes/annum)
Salt load out (tonnes/annum)
Salt load in (tonnes/annum)
Nature Conservation activities 1,2, and 3
Cultural Heritage Conservation Initiatives, Issue 1,2, and 3
Tail water and Drainage Portion of farms with whole farm plans (%)
water Reuse Portion of farms with recycling systems (%)
Volume of water reused on farms (ML/a)
Regionally; is reuse from drains encouraged (Y/N)
Regionally Volume reused from drains (ML/a)
Business Standard business (financial) management ratios, none separate for
drainage
Financial Total expenditure for Environmental issues 1, 2 and 3
Total expenditure for Nature Conservation activities 1,2, and 3
Total expenditure for Cultural Heritage Conservation Initiatives, Issue
1,2, and 3

As may be clear from the selected indicators in Table 6 few indicators will indicate drainage
system performance, and hence additional indicators, such as the direct indicators in Table 7
will have to be included.
In India extensive reporting of drainage developments took place on the Rajad Project (Rajad
Project Staff 1995; Rajad Project Staff 2001). Several Pilot areas were established and
traditional pilot area research was executed. Although some information appropriate for PA
can be derived the results reported are typically research oriented.

Performance Assessment and Potential Indicators for Drainage Systems 30

4. Conclusions and Recommendations
The global experience with performance assessment of drainage systems is scarce and what
exists is very often more research oriented than truly PA. Even if the tools (i.e. indicators)
used are common to both PA and Research, the actual practice of benchmarking of drainage
systems is still to be developed. Even in Australia, where benchmarking of irrigation and
drainage systems has taken off spectacularly, the drainage components included are
primarily descriptive in nature.
Perusing the literature available on performance assessment of drainage systems, it seems
apparent that there is no simple diagnosis method or tool available for the determination of
the factors that make up the performance of a drainage system. The determination of
appropriate factors to date relies mostly upon local expertise and/or research. Poor drainage
performance may result from several deficiencies related to design, materials, construction,
maintenance etc., which are not easy to diagnose several years after drainage construction.
Thus thorough examination of the system components (at least of selected ones) is needed.
The diagnosis method should rely upon a very good theoretical expertise of drainage
function.
Nevertheless, benchmarking the performance of drainage systems can be done. Using
simple (direct) indicators that are easy to measure is a golden rule for successful
benchmarking. Many variables may describe the functioning of drainage systems; they range
widely in ease of collection and only the most direct are suitable for benchmarking as they
are also the easiest to measure. The variables describing the water table behaviour and its
spatial distribution are the easiest to measure and provide a good assessment of the
functioning of the drainage system. For this reason the indicators to be used with PA and
Benchmarking should be based on water table depth measurement. Although watertable
depth is easiest factor to measure this does not mean it is cheap to measure. Those who
have to maintain an observation well network know that such very basic data require
substantial investments in data collection and processing. These costs are considerable
especially if they are to be born by financially weak water user associations, or, for that
matter, water management organisations in general. Hence both in technical and financial
terms a considerable amount of work is to be done before drainage system performance
assessment becomes sufficiently standardised to be able to contribute to benchmarking. The
cases undertaken by the Task Force 4 (TF4) of ICID are therefore a major step in the right
direction, but Performance Assessment needs to be embedded structurally and financially in
existing water management organisations, whether governmental or private. In addition,
appropriate drainage system indicators need to be built into the database established as part
of the TF4 benchmarking project.

Performance Assessment and Potential Indicators for Drainage Systems 31

ANNEX A POTENTIAL PARAMETERS FOR OPERATIONAL AND STRATEGIC
DRAINAGE INDICATORS

Table 7: Potential Parameters for Drainage Indicators

No. Parameters
SUBSURFACE DRAINAGE SYSTEM
SYSTEM DESIGN FEATURES
1 Monthly rainfall isoheytals of project/area
2 Monthly evapo(trans)piration isoheytals of the project/area
3 Topo maps showing geo-hydrological features with respect to water quality sample locations
4 Soil profile data: texture, infiltration, hydr. cond., salinity, sodicity.
5 Pipe diameters
6 Pipe elevations
7 Lateral and collector slopes
8 Lateral and collector lengths
9 Area drained per lateral/collector
10 Outlet elevations of pipes in manholes
11 Condition at outfall into sump or open drain
12 Midway design water depth
13 Drain depths
14 Drain spacings
15 Design drainage coefficient
16 Design hydraulic conductivity
17 Period on which design was based (watertables and rainfall in years before).
18 Determination of drainage coefficient (drainable surplus over which period, i.e. annual)
19 Drain spacing equation used
20 Pipe design equations used (Visser, Manning, Uniform or non-uniform flow)
21 Safety factors used
22 Depth of impervious layer
23 Disposal of drainage water through .....(open drains, pumping into ...)
24 Net cultivated area per area unit
25 Crop yield per net area
26 Available types of crops to be cultivated
CONSTRUCTION FEATURES
27 Method of construction (by hand, machines, etc.)
28 Problems during construction
29 Type of envelope material if any
30 Pipe material
31 Manhole features (material, diameter, plugged, silttrap, etc.)
32 Constructed under dry or wet conditions
33 Lateral alignment after construction
34 Collector alignment after construction
35 Actual spacing
36 Project cost estimated
37 Actual project cost
38 Year and duration of construction

PERFORMANCE FEATURES

Performance Assessment and Potential Indicators for Drainage Systems 32

No. Parameters
Map with all measuring locations, field boundaries, and irrigation units, marked (see also 177),
39 GPS
40 Time of monitoring after construction
41 Time of monitoring after last irrigation (and/ or no 55)
42 Midway water table elevation (between laterals)
43 Natural surface elevation at point of water table observation
44 Water table elevation at envelope/drain interface
45 Actual & calculated hydraulic conductivity
46 Water level (elevation) in drain pipe
47 Discharge of lateral (dimensions at location, water depths u/s, d/s, sill elev., zero point elev.)
48 Discharge of collector (dimensions at location, water depths u/s, d/s, sill elev., zero point elev.)
49 Condition at flow measuring point submerged or free flow
50 Water level just outside pipe trench (25 - 40 cm from the outside of the pipe)
51 Soil profile data: texture, infiltration, hydr. cond., salinity, sodicity.
52 Water elevation in manhole
53 Net cultivated area per drainage unit
54 Crop yield per net area
55 Location of irrigated field at time of water table depth observation
OPERATION AND MAINTENANCE FEATURES
56 Sediments in pipes (0.25 diam., 0.5 diam. etc)
57 Sediments in manholes
58 Sediments in pumping sumps (if applicable)
59 Quantity of pipes blocked (laterals & collectors)
60 Frequency of cleaning pipes
61 Quantity of pipe length in need of cleaning
62 Funding available for maintenance over the years
63 Hours of pumping per day/ hours of pump break-downs
64 Pump maintenance needs
65 Operation by farmers, government?
66 Operation by organisation who designed and constructed?
67 Assessment of maintenance status: excellent, good, medium, poor
WATER QUALITY
68 EC
69 Method of determination EC
70 SAR
71 Method of determination of SAR
72 RSC
73 Method of determination of RSC
74 Chemical composition of EC, SAR, RSC
75 Items of expanded list in Table 8 (items 4 - 36)
76 Method of determination of items 4 - 36 of Table 8
Pesticides, quantities and methods of determination (e.g. insecticides, fungicides, herbicides,
77 acaricides, rodenticides)
78 Selenium, quantities and methods of determination

SURFACE DRAINAGE SYSTEM

79 DESIGN FEATURES
80 Items 1,2,3, and 4 as above

Performance Assessment and Potential Indicators for Drainage Systems 33

No. Parameters
81 Lined or unlined
82 Typical side slopes
83 Design discharge
84 Method of runoff determination
85 Period on which design was based (watertables, drain discharges and rainfall in years before).
86 Determination of drainage coefficient (drainable surplus over which period, i.e. annual)
87 Method of hydraulic design (manning, steady, non-steady, computer model, etc)
88 Disposal of drainage water

CONSTRUCTION FEATURES
89 Method, year and duration of construction
90 Problems during construction
91 Cross-sections at regular intervals (elevations, width, free board, etc.)
92 Details of all structures in the drain
93 Longitudinal profile (elevations, bedslope, side slope, etc)
94 Project cost estimated
95 Actual project cost

PERFORMANCE FEATURES
96 Map with all measuring locations marked
97 Discharge
98 Water elevation upstream of flow measuring point
99 Water elevation downstream from flow measuring point
100 Method of flow measurement
101 Condition at flow measuring point submerged or free flow

OPERATION AND MAINTENANCE FEATURES

102 Side slope stability
103 Maintenance records
104 Time and periods of desilting
105 Reconstruction
106 Amount of sediments
107 Aquatic weed growth assessment and cleaning
108 Funding available for maintenance over the years
109 Operation by farmers, government?
110 Operation by organisation who designed and constructed?
111 Assessment of maintenance status: excellent, good, medium, poor

WATER QUALITY
- Items 68 - 78 as above

Hydrologic and other parameters of area of drainage system

RAINFALL
112 Daily
113 Weekly
114 Monthly
115 Seasonal

Performance Assessment and Potential Indicators for Drainage Systems 34

No. Parameters
116 Yearly
117 Method of data collection
118 Conditions at data collection point (standard FAO/WMO or non-standard; explain)

EVAPO(TRANSPI)RATION
119 Daily
120 Weekly
121 Monthly
122 Seasonal
123 Yearly
124 Method of data collection/calculation (Class A pan, Penman, other)
125 Conditions at data collection point (standard FAO/WMO or non-standard; explain)

IRRIGATION SUPPLIES AND EFFICIENCIES

Primary canal level
126 Daily flows
127 Weekly flows
128 Monthly flows
129 Seasonal flows
130 Yearly flows
131 Canal loss assessments, and year of, method used
Secondary canal level
132 Daily flows
133 Weekly flows
134 Monthly flows
135 Seasonal flows
136 Yearly flows
137 Canal loss assessments, and year of, method used
Tertiary canal (field) level
Surface irrigation supply:
138 Daily flows
139 irrigation interval flows
140 Monthly flows
141 Seasonal flows
142 Yearly flows
143 Tubewell irrigation supply (discharge volumes in time)
144 Private tubewell irrigation supply (discharge volumes in time)
145 Pumped (lift) irrigation supply other than government and private tubewell (volumes in time)
146
147 Average volumetric field applications + method of determination
148 Average volumetric deep percolation losses + method of determination
149 Type of irrigation: rotation, demand, etc.
150 Method of irrigation: border, furrow, sprinkler, trickle, subsurface, etc.

Irrigation water quality

151 EC
152 Method of determination EC

Performance Assessment and Potential Indicators for Drainage Systems 35

No. Parameters
153 SAR
154 Method of determination of SAR
155 RSC
156 Method of determination of RSC
157 Chemical composition of EC, SAR, RSC
158 Items of expanded list in Table 8 (items 4 - 36)
159 Method of determination of items 4 - 36 of Table 8
Pesticides, quantities and methods of determination (e.g. insecticides, fungicides, herbicides,
160 acaricides, rodenticides)
161 Selenium, quantities and methods of determination

GROUND WATER TABLE FLUCTUATIONS

162 Daily (Stevens recorder, datalogger, manual collection)
163 Weekly
164 Bi-weekly
165 Monthly
166 Seasonal
167 Yearly
168 Density of observations
169 Method(s) of observations

WATER BALANCE PARAMETERS to explain rise and fall of water table

170 Recharge from rainfall (inl. surface drains)
171 Runoff of surface drains
172 Recharge of irrigation canal system
173 Recharge of field irrigation losses
174 Recharge/withdrawal through public and private tubewells
175 Withdrawal through evapotranspiration from vegetation (incl. crops)
176 Withdrawal by evaporation from bare/fallow soil surface

LAND USE AND AGRICULTURAL PRACTICES

177 Seasonal land use maps (in space and time)
178 Number of irrigations typical for certain crops and quantities per irrigation
179 Fertilizer applications
180 Pesticides used

Performance Assessment and Potential Indicators for Drainage Systems 36

Table 8: Possible contaminants in drainage water that may be measured.

WHO Pakistan EPA

Normal
standards standards for
Sn drain
Contaminant/Parameter for potable municipal
No quality
water waste water
standards
quality emissions
Standard water quality measurements
1 Electric Conductivity ECw (mmhos/cm or dS/m)
2 Residual Sodium Carbonate, RSC (meq/l)
3 Sodium Adsorption Rate, SAR (-)
Expanded list of environmental factors to assess
water quality
4 Temperature °C
5 pH 7.0 - 8.5 9.5 6.9
6 Biochemical Oxygen Demand (BOD5 at 20 °c (mg/l) 200 80
7 Chemical Ogygen Demand, COD (mg/l) 400 105
8 Total Suspended Solids, TSS (mg/l) 400 200
9 Total Disolved Solids, TDS (mg/l) 500 5000 3500
10 Disolved Oxygen, DO (mg/l)
11 Color Unit (pH)
12 Turbidity ( Nephelometric Turbidity Units, N.T.U.)
13 Taste (objectionable or not)
14 Odour (Treshold Odour Numbers, T.O.N.)
15 Colliforms per ml
16 Faecal Colliforms per ml
17 Faecal Streptococci per ml
18 Grease and Oil (gm/l) 20 10
19 Phenali Compact (µg/l) 1.5 0.3
20 Detergents (mg/l) 30 20
21 Ammonia, NH3 (mg/l) 75 40
22 Chronium, Cr (µg/l) 2 1
23 Copper, Cu (µg/l) 4 1
24 Nikkel, Ni (µg/l) 2 1
25 Chloride, Cl^-1 (mg/l) 1000 1000
26 Chloor, Cl2 (mg/l) 1 1
27 Calcium, Ca (mg/l) 75
28 Floride, Fl (µg/l) 10
29 Iron, Fe total (µg/l)
30 Mercury, Hg (µg/l) 0.1 0.01
31 Magnesium, Mg (mg/l) 50
32 Hardness (mg/l) 100
33 Sulphates, SO4^-2 (mg/l) 200 1000 1000
34 Aluminium, Al
Pesticides (e.g. insecticides, fungicides, herbicides,
35 acaricides, rodenticides) 0.75 0.15
36 Selenium

Performance Assessment and Potential Indicators for Drainage Systems 37

ANNEX B RATIONALE AND LONG-LIST OF INDICATORS FOR DRAINAGE SYSTEM
PERFORMANCE ASSESSMENT.
Indicators have been identified as direct ([nearly] independent indicators, sometimes also
referred to as primary indicator) and indirect indicators (dependent on other indicators as well
and also referred to as secondary indicator). Direct indicators are left justified in the column,
while indirect indicators are right justified. Sometimes no agreement on whether an indicator
was direct or indirect could be reached and hence the mark is in the centre of the cell. The
idea is that direct indicators can satisfy the rationale without have to consider other
indicators. A typical example is the indicator(s) to determine the functioning of a drainage
system. Water table alone can serve to assess the functioning. However, many engineers
prefer to have information on soil salinity as well. Yet, high salinity can mean that either the
drainage system does not function or not enough irrigation water is applied. In that case
salinity is only an indirect indicator which will by itself cannot lead to conclude whether the
drainage system is working or not.

Performance Assessment and Potential Indicators for Drainage Systems 39

Table 9: Potential indicators for Irrigation and Drainage System Performance Assessment

Purpose of Performance Assessment (Rationale)

Parameters needed Economic Drainage design criteria
Functioning
(from Table 7 unless Need for Crop Farmers drain drain design drain
No. Indicators drainage
otherwise indicated) rehabilitation yield income depth spacing discharge envelope
system
Dir. Indir. Dir. Indir. Dir. Indir.D. I. D. I. D. I. D. I. D. I.
Subsurface Drainage System
1 Water table as function of time 39,40,41,42,43,177 X X X X X
2 Watertable draw down rate X X
3 Relative ground water depth (ICID) 9,39,40,41,42,43,177 X X X X X X
4 Water table depth as function of the area 9,39,40,41,42,43,177 X X X
Ratio of unit discharge to design
5 discharge 9,15,40,(47 or 48) X X X X
6 Q-h plots 6-16,19-23,39,42-50,55 X X
7 Q/H ratio 6-16,19-23,39,42-50,55 X X
8 Trafficability Cemagref X X
Drainage system discharge as function
9 of time (q-t, Q-t ) 9,40,47- or 9, 41,47 X X X X X
10 Drainage intensity (a) 6,41,42,46-or-9,15,42,47 X
Calculated hydraulic conductivity
11 compared with design 16,6,9,41,42,46,47,22,35 X X X
12 Head loss fraction (he/htot) over time 6,42,46,50 X X
Required spacing compared with design 6,41,42,46-or-
13 spacing 9,15,42,47,35 X X X X
14 Entrance resistance re=he/q 8,46,50,47 X X
Relative change of collector drainage
15 water quality over time 3,68,40 &70,40
Sustainability of drained area (modified
16 ICID for irrigable area) 24,40,53 X

Performance Assessment and Potential Indicators for Drainage Systems 40

 Purpose of Performance Assessment (Rationale)
Parameters needed Economic Drainage design criteria
Functioning
(from Table 7 unless Need for Crop Farmers drain drain design drain
No. Indicators drainage
otherwise indicated) rehabilitation yield income depth spacing discharge envelope
system
Dir. Indir. Dir. Indir. Dir. Indir.D. I. D. I. D. I. D. I. D. I.
Maintenance
17 Sediment in pipes, manholes 56, 57 X X
18 Ratio of overpressure in collectors 10,52
19 Age of the drainage system 40
20 Frequency of complaints X
Frequency, level and costs of
21 maintenance 60-67, 102-111 X

Surface Drainage System

Ratio of design discharge versus actual
22 discharge over time X
23 Changes in head-discharge relationships X
Maintenance
24 Weed control intensity X
25 Sediment control intensity X
Age of structures and frequency of
26 maintenance X

Water balance
Rootzone water movement ratio 112-118, 119-125, 138 -
27 (leaching) 145 X X X

Soil
Change in hydraulic conductivity over
28 time 16,45 NOT

Performance Assessment and Potential Indicators for Drainage Systems 41

 Purpose of Performance Assessment (Rationale)
Parameters needed Economic Drainage design criteria
Functioning
(from Table 7 unless Need for Crop Farmers drain drain design drain
No. Indicators drainage
otherwise indicated) rehabilitation yield income depth spacing discharge envelope
system
Dir. Indir. Dir. Indir. Dir. Indir.D. I. D. I. D. I. D. I. D. I.
29 Relative change of soil salinity 3,40,51 X
30 Soil salinity over time 3,40,51 X X X
Relative change of soil alkalinity (SAR or
31 ESP) 3,40,51 X
32 Soil alkalinity (SAR or ESP) over time 3,40,51 X X
33 Infiltration rate over time 3,40,51 NOT

Agronomic factors
34 Yield over time X X X X X
35 Relative yield change X X
36 Relative cropped area change X X X
37 Variability of crops X X
38 Changes in cropping pattern

Economic
39 Farm net benefit X X
40 Net present worth (NPW) X X
41 Benefit cost ratio (B/C) X X X
42 Internal rate of return (IRR) X X

Social
43 Benefits to the farmers X X
44 Employment X

Performance Assessment and Potential Indicators for Drainage Systems 42

 Purpose of Performance Assessment (Rationale)
Parameters needed Economic Drainage design criteria
Functioning
(from Table 7 unless Need for Crop Farmers drain drain design drain
No. Indicators drainage
otherwise indicated) rehabilitation yield income depth spacing discharge envelope
system
Dir. Indir. Dir. Indir. Dir. Indir.D. I. D. I. D. I. D. I. D. I.
45 Workability X X

Performance Assessment and Potential Indicators for Drainage Systems 43

PART B
This section of the report details practical case studies from Pakistan, Egypt, Morocco,
France and Australia.

Part B Contents
1. Pakistan - Water Table Analysis ................................................................................46
1.1. Response explanation water table fluctuations..............................................46
1.2. Trend analysis....................................................................................................47

2. Egypt - Water Table Analysis .....................................................................................48

2.1. Water table trends..............................................................................................48
2.2. Water table drawdown curves. .........................................................................50
2.3. SEWxx ..................................................................................................................51
2.4. Q-H plots.............................................................................................................54

3. Egypt - Complaint Assessment .................................................................................57

3.1. Rehabilitation of subsurface drainage systems and complaint
assessment ........................................................................................................57
3.2. Indicators for Rehabilitation .............................................................................57
3.3. Complaints collection........................................................................................58
3.4. Analysis of Complaints in Santa (stage 1 of the study) .................................60
3.5. Analysis of Water Table Depth in Santa ..........................................................61
3.6. Salinity ................................................................................................................63
3.6.1. Spatial Distribution of Age and Complaints in different Drainage
Directorates (stage 2 of the study) ...........................................................63
3.6.2. Complaints related to the year of installation ...........................................63
3.7. Conclusions and Recommendations...............................................................64

4. Morocco - Hydraulic System Indicators....................................................................66

4.1. Buried drainage systems: collectors and drains............................................66
4.1.1. Collectors. ................................................................................................66
4.1.2. Drains .......................................................................................................66
4.2. Performance of secondary drainage canals ...................................................67
4.2.1. Introduction to the canal under study. ......................................................67
4.2.2. Drainage design discharge and initial characteristics of the canal...........68
4.2.3. Drainage design discharge and present condition of the canal................69
4.2.4. Influence of the vegetation .......................................................................69
4.2.5. Sediment influence...................................................................................70
4.3. Conclusions .......................................................................................................70
5. Morocco - Crop yields in the Gharb Plain.................................................................71

Performance Assessment and Potential Indicators for Drainage Systems 44

6. France - Drainage Efficiency and Agronomic Outcomes ........................................74
6.1. Types and objectives of land drainage performance assessment
programs in France ...........................................................................................74
6.1.1. French PA programs launched by designers were of two main different
types:........................................................................................................74
6.2. Research and Development and Performance Assessment Programs .......75
6.2.1. Assessment of drainage efficiency and benefits ......................................75
6.2.2. Determination of the factors of drainage performances ...........................76
6.2.3. Methods and indicators ............................................................................76
6.3. Drainage efficiency: case study in eastern France ........................................76
6.4. Drainage benefits...............................................................................................80
6.4.1. Yield differences between trench and mid-spacing locations...................81
6.4.2. Yield components.....................................................................................82
6.4.3. Factors for drainage performance ............................................................83
6.5. Conclusions .......................................................................................................83
6.6. Acknowledgements ...........................................................................................84

7. Australia - Salinity Control in Vineyards...................................................................85

7.1. Site Details..........................................................................................................85
7.2. Salinity assessment using Electromagnetic survey technique.....................86
7.3. Conclusion .........................................................................................................90
7.4. Recommendations.............................................................................................90

REFERENCES .......................................................................................................................92

Performance Assessment and Potential Indicators for Drainage Systems 45

1. Pakistan - Water Table Analysis

1.1. Response explanation water table fluctuations

A water table hydrograph (Zuberi and Vlotman 1991), with key parameters that affect the
water table elevation is shown in Figure 8.

1/ 1/ 90 1/ 4/ 90 30/ 6/ 90 28/ 9/ 90 27/ 12/ 90 27/ 3/ 91 25/ 6/ 91

0.00 12.00
Rain f all
31.6 26.7
55.3 10.00

Eva pora tion in m m /da y

Depth in meter below NSL

102 mm

Pan ev ap o rat io n 8.00

1.50
6.00

6 " D r ai n 4.00
El evati on
W at ert ab le
3.00 2.00

0.00
0.00 70.0

60.0
Canal closure

Canal closure
Depth in meter below NSL

Discha rge in m ^ 3/s

50.0
A v . w eekly irr. can al f lo w
1.50 40.0

30.0
W at ert ab le
20.0

3.00 10.0

0.0
0.00 1.4

31.6 26.7 Rain f all

55.3 1.2
Depth in meter below NSL

Discha rge in m m /da y

102 mm 1.0
Su mp d isch arg e
1.50 0.8

W at ert ab le 0.6

0.4

3.00 0.2

0.0
1/ 1/ 90 1/ 4/ 90 30/ 6/ 90 28/ 9/ 90 27/ 12/ 90 27/ 3/ 91 25/ 6/ 91

Figure 8: Depth of water table over time in relation with rainfall, irrigation and growth
stages.

Performance Assessment and Potential Indicators for Drainage Systems 46

The drain elevation is indicated. Although water table as a function of time is considered a
direct indicator for the functioning of the drainage system, it is clear from Figure 8 that over a
short period (less than 2 years) other parameters are necessary to explain the various
fluctuations. There is a clear response to increased crop consumptive use and major rainfall
events. After canal closure in January the water table shows a rapid decline. However, it also
shows that the location of drain depth apparently has little effect on the fluctuation, and that
pumping of the sump of the subsurface drainage system follows the fluctuation of the water
table, rather than being a constant factor that lowers the water table. The reason for this is
that the sump unit of which the results are shown, is but one of many in the area, which is
also drained by traditional tubewell drainage schemes and thru the open surface drains
through which the sump units discharge. All drainage schemes are operational in the same
unconfined aquifer. The permeability of the aquifer is in the order of 50 m/d.

1.2. Trend analysis

A long-term record from the same area is shown in Figure 9. The individual annual ups and
downs are also visible, but because the record shows seven years of data, an overall
downward trend may be observed. However, if only data of the period Jan 88 – Jul 90 had
been used an upward trend would have been concluded. Therefore even with long-term
records care should be taken to consider whether the period might be part of a wet or dry
cycle in the weather pattern.
Surface elevation is 623.71 ft above M SL.
623.00
CHA K151, Fourth D rainage Proje c t, Pakistan
Elev ation w ith res pec t to MSL in f t.

622.00

621.00

620.00

619.00

618.00

617.00

616.00

615.00
5

2
85

86

87

88

89

90

91

92
l- 8

l- 8

l- 8

l- 8

l- 8

l- 9

l- 9

l- 9
n-

n-

n-

n-

n-

n-

n-

n-
Ju

Ju

Ju

Ju

Ju

Ju

Ju

Ju
Ja

Ja

Ja

Ja

Ja

Ja

Ja

Ja

Figure 9: Water table hydrograph Pakistan.

Performance Assessment and Potential Indicators for Drainage Systems 47

2. Egypt - Water Table Analysis
From Egypt similar data are available but conditions are completely different from Pakistan.
In Egypt, conditions are such that localised lowering of the water table is possible, and hence
the water table hydrograph is more of an indication of the functioning of the drainage system
in the area. Almost all subsurface drainage systems are gravity drains. The open drains have
large pumping stations.

2.1. Water table trends

An example of trend line analysis is shown in Figure 10 and Figure 11. Long-term
observation well data was analysed (Eissa 2000). Daily observations were available which
were analysed with daily, monthly, seasonal and 6 day after irrigation frequencies. All 6
observation wells with data from 1988–1995 showed similar characteristics in that the means
of each interval tended to be the same and those trends became more pronounced with the
larger interval.

Performance Assessment and Potential Indicators for Drainage Systems 48

 Drain Spac ing = 15 m Mas htul Pilot A rea (OW 1. Unit 1)
Drain Depth = 1.2 m Daily Watertable Depth Below Soil Surf ac e

Se 2

Se 3

Se 5
1

Ja 2

Ja 3

Ja 4

95
89

90

91
8

89

90

91

M 2

93

94

M 5
-9

-9

-9

-9
9

9
-9

-9
-8

-8

-9

9
p-

p-

p-
b-

b-

b-
n-

n-

n-

n-

n-

n-

n-
ay

ay

ay

ay
ep

ep
ct

ct

ct
Fe

Fe

Fe
Ju

Ju

Ju

Ja

M
O

S
Ground Level
0.000
Des ign W.T Lev el -0.9 m
-0.200
-0.400
WTD Below Soil Surf ac e (m)

-0.600
-0.800
-1.000
W
-1.200
-1.400
-1.600
Drain Depth -1.2 m
-1.800
y = 9E-06x - 1.2542
-2.000
-2.200

Series 1 Linear trendline

Mas htul Pilot A rea (OW 1. Unit 1)

Monthly A verage Watertable Depth Below Soil Surf ace
89

90

91

92

93

94

95
8

89

90

91

92

93

94

95
-8

-8

-9

-9

-9

-9

-9
b-

b-

b-

b-

b-

b-

b-
n-

n-

n-

n-

n-

n-

n-
ct

ct

ct

ct

ct

ct

ct
Fe

Fe

Fe

Fe

Fe

Fe

Fe
Ju

Ju

Ju

Ju

Ju

Ju

Ju
O

O
0.2
Ground Level
0
-0.2 Drain Spac ing = 15 m
WTD Below Soil Surf ac e (m)

Drain Depth = 1.2 m y = 0.0009x - 1.0161

-0.4
-0.6
-0.8 Des ign W.T Level
-1
-1.2
-1.4 Drain Depth
-1.6
-1.8
-2
-2.2

A v erage W.T.D Linear (A v erage W.T.D)

Mashtul Pilot A rea (OW 1. Unit 1)

Seas onal A v erage Watertable Depth Below Soil Surf ac e
89

90

91

92

93

94

95
8

4
t8

t8

t9

t9

t9

t9

t9
m

m
in

in

in

in

in

in

in
su

su

su

su

su

su

su
w

-0.2
y = 0.0059x - 1.0178
-0.4

-0.6

-0.8

-1

-1.2

-1.4

OW1 Linear (OW1)

Figure 10: Trend line analysis with daily, monthly and seasonal average midway water table
depth values.

What can be derived from Figure 10 and Figure 11 is that there is an upward trend and
hence a problem may be developing. However, this is not exactly a direct indicator of
drainage system functioning as it may have been caused by changes in irrigation

Performance Assessment and Potential Indicators for Drainage Systems 49

applications or increased reuse. Extensive reuse of irrigation water is common in the Nile
Delta.

Mashtul Pilot Area (OW 1. Unit 1)

Seasonal Average Watertable Depth & The Depth at 6 day from Irrigation

Drain Spacing = 15 m Drain Depth = 1.2 m

87

88

89

90

91

92

93

94

95
6

4
t8

t8

t8

t8

t9

t9

t9

t9

t9
m

m
in

in

in

in

in

in

in

in

in
su

su

su

su

su

su

su

su

su
w

w
0.2

Ground Level seasonal average

0
WTD Below Soil Surface (m)

-0.2 Depth after 6 days

-0.4
Linear (seasonal
average)
-0.6
Linear (Depth
-0.8 after 6 days)
Design W.T Level
-1

-1.2
Drain Depth
-1.4

-1.6

-1.8

Figure 11: Trend in seasonal water table hydrograph and six days after irrigation seasonal
water table.

2.2. Water table drawdown curves.

From a study of draw-down curves (Eissa 2000) it was observed that the water table six days
after irrigation was generally at the desired design depth (Figure 12). Consequently six days
after irrigation water tables were also analysed (Figure 11) and it was found the water tables
were generally 0.2 m higher than the average values. Periods of high water tables (close to
the surface) were generally caused by rice cultivation. Farmers then closed the subsurface
drainage system. Therefore these periods should not be considered when the water table is
used as indicator for drainage system functioning.

Performance Assessment and Potential Indicators for Drainage Systems 50

 W atertable Draw dow n After Irrigation
M ASHT UL PILO T AREA Mas h. U pper B. Mas h. Low er B.
drain spacing = 30m hydraulic head = 0.3 Drain depth = 1.2m

1.3 0.1

Ground level
1.1 -0.1

0.9 Initial WT level -0.3

H ydraulic head (m)

Upper boundary

Watertable depth (m)

0.57 m
0.7 -0.5
Average

0.5 -0.7

Lower boundary
0.3 -0.9

0.1 -1.1

-0.1 -1.3
1 2 3 4 5 6 7 8
D ays after irrigation

Figure 12: Example of water table draw down analysis results.

2.3. SEWxx
Another example of possible use of the water table hydrograph is to link with understanding
of crop responses. Crop responses vary with the pattern of the water table regime (the
duration and timing of the high water tables are especially important), these patterns may to
some extent be captured by the Wx and the SEWx indices. Calculation of SEWx indices is
shown in Figure 13

Figure 13: Calculating SEW values (Smedema et al. 2004)

Performance Assessment and Potential Indicators for Drainage Systems 51

Good correlations with a range of crop yields were obtained in the UK for the W40 Index
(Watertable shallower than 40cm) and for yields of sugar cane in Australia for the W50
(Watertable shallower than 50cm) index. Figure 14 shows two cases of similarly good
correlations for the SEW30 (Watertable shallower than 30cm) index. The nature of the
response shown by these indices depends on the period to which they apply (winter/off-
season indices mostly showing the adverse impact of high water tables on soil structure
while for the summer/growing season indices this is mostly the impact of oxygen deficiency).
In some cases weighted indices were used which took into account the occurrence of the
waterlogging in relation to the development cycle of the crops, but these have not always
proved to be superior to the non-weighted indices. For references and more detailed
discussion, see Smedema 1988.

Figure 14: Relationships between crop yield and SEW values (Smedema et al. 2004)

An example of this possible use of the water table hydrograph is where it was decided
arbitrarily that the water table would be allowed to fluctuate between 0.2 and 0.8 m below the
surface (Mahmoud et al. 2000). A visual inspection of the graph was used (Figure 15) to
judge whether the drainage system operated satisfactorily or not.

Performance Assessment and Potential Indicators for Drainage Systems 52

 Mashtul Pilot Area- Unit (1) Well (1)
Lat(10) 15m Spacing - 1.2m Depth (89-1995)
01-Nov-88 01-Nov-89 01-Nov-90 01-Nov-91 31-Oct-92 31-Oct-93 31-Oct-94 31-Oct-95
0.0

-0.3
Depth of water table (m) -0.5

-0.8
-1.0

-1.5

-2.0

-2.5

Date
W.T.Depth Wet stress Dry stress Linear (W.T.Depth )

Figure 4 Relationship between water table depth and time (Relations entre le table d’eau et le temp)

Mashtul Pilot Area- Unit (9) Well (5)

Lat (80) 30m Spacing - 1.2m Depth (89-1995)
01-Nov-88 01-Nov-89 01-Nov-90 01-Nov-91 31-Oct-92 31-Oct-93 31-Oct-94 31-Oct-95
0.0

-0.3
Depth of water table (m)

-0.5

-0.8
-1.0

-1.5

-2.0

-2.5

Date
W.T.Depth Wet stress Dry stress Linear (W.T.Depth)

Figure 5 Relationship between water table depth and time (Relations entre le table d’eau et le temp)

Mashtul Pilot Area- Unit (13) Well (7)

Lat (94) 15m Spacing - 1.7m Depth (89-1995)
01-Nov-88 01-Nov-89 01-Nov-90 01-Nov-91 31-Oct-92 31-Oct-93 31-Oct-94 31-Oct-95
0.0

-0.3
Depth of water table (m)

-0.5

-0.8
-1.0

-1.5

-2.0

-2.5

Date
W.T.Depth Wet stress Dry stress Linear (W.T.Depth)

Figure 6 Relationship between water table depth and time (Relations entre le table d’eau et le temp)

Figure 15: Water table fluctuation between 0.2 and 0.8 m below the surface.

Performance Assessment and Potential Indicators for Drainage Systems 53

 2.4. Q-H plots
It is common to display Q-H plots, where Q is the discharge from the subsurface drain of
interest and H the midway water table depth or midway water table head. Examples of these
plots may be found in description of measurement results of the Rajad Project (Rajad 1995)
and recently a number of these type of plots were produced at the Drainage Research
Institute (DRI) in Egypt. Without exception all plots show a cloud of data without any clear
relationship. However, when data are collected with attention to the irrigation schedules it is
expected that this indictor could be used to see whether the system operates according the
design criteria used.
To illustrate the difficulties with using the Q-H plots two examples are given. Both are from
the same pilot area in the Egyptian Nile delta (Bakr, Abdel Fattah et al. 1998). In Figure 17
the data collected each year from 1989 – 1995 are shown, as well as, a figure that combines
all. The maximum discharge with Manning (n = 0.014, S = 0.001) of the corrugated PVC
lateral at 1.2 m depth was determined to be 0.000295 m3/s or 2.75 mm/d using the full length
of the lateral of 310 m and a spacing of 30 m. This assumes uniform recharge of 2. 75 mm.
In some years the discharge was higher than the theoretical model, although not by much,
and this may be caused by the assumption of slope and the Manning coefficient.
Hydraulic conductivity values at the point of measurement were not available but pre-
drainage data (1980, Kavg80) and measurements at various locations in the area in 1995,
have been used in the Hooghoudt equation to obtain indicator lines. As can be seen the K-
value has a significant effect on the position of the indicator lines. No additional data was
available besides the water table depth data, so it could not be ascertained whether lines
were submerged or not. Also no data on the uniformity of irrigation along the line are
available. The geometric mean of the hydraulic conductivity of the 1995 did not differ
substantially from the average value as may be seen in Figure 16. Figure 16 shows the data
in one plot for all years, but for a drain spacing of 15 metres. The smaller spacing is the main
cause for the q Manning to be off the chart (13.4 mm/d). In both graphs, low discharge with
high head, could indicate submerged conditions, or a mismatch between time of water table
depth observation and discharge measurement. Although, taken on the same day, it could
not be ascertained whether they were within one hour of each other. Most probably not as,
the observers tended to take all the depths first and then the discharges.

Performance Assessment and Potential Indicators for Drainage Systems 54

 6.0 6.0
MPA-Unit (9)-We ll (5)-Lat (80) MPA-Unit (9)-We ll (5)-Lat (80)
2 2
30m Sp acing - 1.2m D ep th - A sur .=9300m 30m Sp acing - 1.2m D ep th - A sur .=9300m
5.0 5.0
qmanning p ip e flow ing full = 2.75 mm/d qmanning p ip e flow ing full = 2.75 mm/d
qmax according H ooghoudt at h = 1.2 m qmax according H ooghoudt at h = 1.2 m
q(manning) q(manning)
4.0 4.0
q10(mm/d)

q10(mm/d)
K max95 K max95
K min95 K min95
3.0 3.0
K avg95 K avg95
K avg80 K avg80
2.0 2.0
1989 1990

1.0 1.0

0.0 0.0
0.0 0.5 1.0 0.0 0.5 1.0
h(m) h (m )

6.0 6.0
MPA-Unit (9)-We ll (5)-Lat (80) MPA-Unit (9)-We ll (5)-Lat (80)
2 2
30m Sp acing - 1.2m D ep th - A sur .=9300m 30m Sp acing - 1.2m D ep th - A sur .=9300m
5.0 5.0
qmanning p ip e flow ing full = 2.75 mm/d qmanning p ip e flow ing full = 2.75 mm/d
qmax according H ooghoudt at h = 1.2 m qmax according H ooghoudt at h = 1.2 m
q(manning) q(manning)
4.0 4.0
q10(mm/d)

q10(mm/d)
K max95 K max95
K min95 K min95
3.0 3.0
K avg95 K avg95
K avg80 K avg80
2.0 2.0
1991 1992

1.0 1.0

0.0 0.0
0.0 0.5 1.0 0.0 0.5 1.0
h(m) h (m )

6.0 6.0
MPA-Unit (9)-We ll (5)-Lat (80) MPA-Unit (9)-We ll (5)-Lat (80)
2 2
30m Sp acing - 1.2m D ep th - A sur .=9300m 30m Sp acing - 1.2m D ep th - A sur .=9300m
5.0 5.0
qmanning p ip e flow ing full = 2.75 mm/d qmanning p ip e flow ing full = 2.75 mm/d
qmax according H ooghoudt at h = 1.2 m qmax according H ooghoudt at h = 1.2 m
4.0 q(manning) 4.0 q(manning)
q10(mm/d)

q10(mm/d)

K max95 K max95
K min95 K min95
3.0 3.0
K avg95 K avg95
K avg80 K avg80
2.0 2.0
1993 1994

1.0 1.0

0.0 0.0
0.0 0.5 1.0 0.0 0.5 1.0
h(m) h (m )

6.0 6.0
MPA-Unit (9)-We ll (5)-Lat (80) MPA-Unit (9)-We ll (5)-Lat (80)
q(manning)
2
30m Sp acing - 1.2m D ep th - A sur .=9300m 30m Sp acing - 1.2m D ep th - A sur .=9300m 2
5.0 5.0 K max95
qmanning p ip e flow ing full = 2.75 mm/d qmanning p ip e flow ing full = 2.75 mm/d K min95
qmax according H ooghoudt at h = 1.2 m qmax according H ooghoudt at h = 1.2 m
4.0 q(manning) 4.0 K avg95
q10(mm/d)

q10(mm/d)

K max95 K avg80
K min95 1989
3.0 3.0
K avg95 1990
K avg80 1991
2.0 2.0
1995 1992
1993
1.0 1.0
1994
1995
0.0 0.0
0.0 0.5 1.0 0.0 0.5 1.0
h(m) h (m )

Figure 16: Example of Q-H plots, Egypt.

Performance Assessment and Potential Indicators for Drainage Systems 55

 6.0
MPA-Unit (1)-We ll (1)-Lat (10)
15m Sp acing - 1.2m D ep th - A sur.=4050m 2 K max95
5.0 K min95
qmanning p ip e flow ing full = 13.37 mm/d
qmax according H ooghoudt at h = 1.2 m K avg95

4.0 K avg80
K geom95
q10(mm/d)

1989
3.0
1990
1991
2.0 1992
1993
1994
1.0
1995

0.0
0.0 0.5 1.0
h(m)

Figure 17: Example of Q-H plots, Egypt.

Performance Assessment and Potential Indicators for Drainage Systems 56

3. Egypt - Complaint Assessment

3.1. Rehabilitation of subsurface drainage systems and complaint

assessment
Subsurface drainage systems were installed in Egypt to lower the water table, and remove
salts from the plant root zone. Pipe drainage installation began in Egypt in the 1940s. In the
1960’s, lateral drains were installed at a fixed spacing of 60 m. Lateral drain spacing currently
varies from 20 to 80 m and depends on specified drainage design criteria and soil hydraulic
properties. It is estimated that 1.85 million ha have been subsurface drained, with 840 000 ha
remaining to be drained. However, there are indications that parts of the originally installed
drainage systems are not functioning properly.
Features of poor system performance include: reduced hydraulic capacity of the system;
overpressure in the pipes; higher Manning’s roughness coefficients in the pipes as a result of
sedimentation; higher water tables; desalinisation; reduction in crop yield, and a longer
reclamation period (DRI, 1993).
Many of the areas, which have been provided with subsurface drainage systems since the
1960’s have problems which are related to sedimentation; age; water logging and salinity.
These areas have passed their economical lifetime, which is usually 26 years.
The problems of those areas resulted from the fact that maintenance is not helping any more.
Considering, the lifetime of those areas, the materials and tools used for implementation at
that time and the concepts and criteria of both planning and design, then, rehabilitation may
be a must ((Rady 1993) and (Salman 1995).
There is an urgent need to precisely identify the areas currently requiring rehabilitation. Four
strategies were recommended for this process (DRI, 1993) as follows:
1. Complaints by farmers should be investigated with field observation.
2. Areas with any failure in the network should be identified.
3. Pipes, which demonstrate permanent over-pressure or are continually blocked with
sediment.
4. Long term monitoring program of water tables, drain discharges, crop yields, and soil and
water salinity be developed.
Contractors always implement rehabilitation works. The definition of rehabilitation says that
the rehabilitation is the new construction done by contractors to bring the existing system
back to its former proper functional state.
Smedema et al. 2004 mentioned that in order to assess the need for rehabilitation only two
parameters need to be considered: the water table depth in time and over an area and the
soil salinity. That is in addition to complaints and maintenance costs.
In the National Drainage Program (NDP), a 12-year program is included for the rehabilitation
of an area of 197 400 ha of which 54 600 ha has to be implemented in the first 6-year slice
(1993-1998) (DRP, 1995).

3.2. Indicators for Rehabilitation

The NDP identified general indicators and criteria for the selection of new areas in need of
subsurface drainage and gave some additional indicators and criteria for areas in need for
rehabilitation as:
1. A water table less than 100 cm depth below the soil surface in at least 75% of the area.

Performance Assessment and Potential Indicators for Drainage Systems 57

2. Soil salinity (ECe) exceeding 4 dS/m at 25 and 50 cm depth below soil surface.
3. Areas where decline in agricultural production (about 20-30%) is reported due to high
water table and/or saline conditions.
4. Effect of maintenance: cases where intensive maintenance and flushing do not result in
an improvement of the existing situation.
5. Farmers’ views and willingness to pay with respect to rehabilitation (complaints).
6. Age of the system.
7. Design and installation history.
A joint EPADP/DRI workshop was held in 1996, on performance assessment and
rehabilitation. It became clear during the workshop that in order to assess the need for
rehabilitation, the following indicators should be used:
1. The age of the systems.
2. The number of farmers’ complaints.
3. Level of water table.
4. Soil salinity.
5. Maintenance cost.
DRI started to study these indicators of rehabilitation in 1996 with the exception of the last
one. The maintenance cost was studied by EPADP’s M&E project. The aim of this study was
to improve the decision-making process on which EPADP can base their yearly programmes
for rehabilitation. The DRI research programme for rehabilitation was implemented in two
stages, each stage included some indicators.
In the first stage of rehabilitation study (DRP1), three indicators were tested in the Santa area
(central part of the Middle Delta) representing one of the old areas (the installation of
subsurface drainage systems started 20 years ago). These indicators were farmer’s
complaints, water table depth and salinity. The farmer’s complaints were collected in this
area (40,000 fedan) for a period of five years before the study. At the same time the
measurements of water table depth and salinity were undertaken for two periods of irrigation
midway between two laterals and in the middle of each collector on both right and left sides.
The research programme for rehabilitation in the second stage was implemented on a large
scale. The DRI focused during this stage on studying two indicators: age of the drainage
system and farmer’s complaints.
This chapter describes the results of the study that started in 1995 and completed in 1998.

3.3. Complaints collection

The current farmer’s complaints procedure can be explained as follows:
1. The farmers who have complaints go to the drainage sub centres at the drainage centre
(Figure 18) and file their complaints.
2. They submit the written complaints to the drainage engineer. The complaint contains the
farmer’s name, the location of the problem (the village’s name, the number of the
collector), and the nature of the complaint.
3. The drainage engineer collects the complaints and sends them to the drainage centre as
soon as possible to be studied and solved.
4. The drainage centres put the complaints in files and send a standard complaints
summary form to the Drainage Directorate General.

Performance Assessment and Potential Indicators for Drainage Systems 58

Figure 18: Number of complaints per Drainage Center.

In the second stage of the rehabilitation study (DRP) special arrangements have been
followed to be able to collect complaint data by DRI staff from different drainage centres. The
number of complaints for collectors, laterals and open drains was collected on a monthly
basis from drainage centres (Figure 19) in the West, Middle, and East Delta, and Middle
Egypt for the period 1993 - 1996. These data were tabulated and summarised.

Total no. of complaints

No. of collectors with complaints
Blocked/garbage
Broken/collapsed
Collectors Sediment
Trees/roots
Blocked manhole
Open joints
Other
Total no. of complaints
No. of laterals with complaints
Laterals Sediment
Cut
Other
Total no. of complaints
Open drains Weed/sediment
Other
Figure 19: Overview of type of complaints

Performance Assessment and Potential Indicators for Drainage Systems 59

 3.4. Analysis of Complaints in Santa (stage 1 of the study)
It was observed from the analysis of complaints that 40% of collectors were working without
any problems. The main problems observed for the rest of the collectors were caused by
high water levels and problems of weed growth in the open drains. The number of complaints
was classified into four classes as: 1-10, 11-20, 21-30 and 31-40 complaints.
Table 10 shows the frequency of complaints for the different Sub-Centers under study. It was
noticed that the majority of the collectors with complaints did not exceed the first class (1-10
complaints).This means that there were no effective problems that led to rehabilitation of the
network. The rest of the collectors represented by the rest of the classes mentioned in Table
10 were not in need for rehabilitation, even though all these areas were planned for
rehabilitation by EPADP based on the age as indicator only. Figure 20 shows the collectors
with/without complaints. It was observed from the figure that Tukh Mazyd, Belkeem and
Shubrakass sub centres were in most need of rehabilitation according to the complaints by
farmers. But at the same time, if the low number of these complaints per collector was
considered, one should have recognised that this number was less than 10 for most of the
collectors during the last five years. In other words, the number of complaints for these areas
was not significant and these collectors were not in need of rehabilitation.

Table 10: Frequency analysis of complaint numbers for the drainage sub centres of the Santa
Drainage Centre
Class of Number of collectors for different Sub Centers
Complaints Mehallet Shenrak Shubrakas Ekhnaway Tukh Belkeem Mit
Number Rouh Mazyed Haway
1-10 22 28 46 35 42 50 31
11-20 0 0 4 2 8 2 0
21-30 0 0 2 1 2 0 0
31-40 0 0 1 0 0 0 0

90
80
70
No. of collectors

60
50
40
30
20
10
0
Ekhnaway Shubrakas Mehalleyt Tukh Shenrak Belkeem Mit Haway
Rouh Mazyed

Coll. with complaints coll. without complaints

Figure 20: The relationship between the number of collector’s with/without complaints in
different sub centers of the Santa area

Performance Assessment and Potential Indicators for Drainage Systems 60

Generally, the results from the Santa area revealed that the number of complaints collected
over five years were relatively few for the large area served area. Most of these complaints
were solved by the sub centres. The total number of complaints does not give a real picture
about the need for rehabilitation in an area. In addition one needs to know the total collectors
served by the sub centre and the number of complaints per collector.

3.5. Analysis of Water Table Depth in Santa

In any irrigated field provided with a subsurface drainage system the rate of water table draw
down depends mainly on soil hydraulic conductivity, drain depth and drain spacing. The
desirable water table depth is one that creates an air water balance and does not create dry
or wet stress on the plant roots.
The drainage criteria in the present design of all drainage projects in Egypt are based on the
steady state Hooghoudt equation. A design discharge for a lateral drain is 1 mm/day. At the
same time the elevation of the water table above drain level is 0.4 m. Then the average drain
depth equals 1. 4 m below soil surface. The water table should be at about 1 m below soil
surface at the design discharge. In areas with upward seepage the design discharge is
increased by 0.5 or 1 mm/day (only in the Nile Delta). Water table recession in different sub
centres of the Santa area is illustrated in Figure 21.
Generally, it could be observed from this figure that: for Belkeem; Tukh Mazyed, Eknaway
and Shubraks sub centres, the average water table depth through the whole irrigation cycle
was 84, 63, 68, and 92 cm below the soil surface respectively. At the same time it was
observed that the average water table depth 6 days after irrigation was 87.5, 47, 60.5 and 59
cm below soil surface for the same sub centres, respectively. This means that these
collectors were not functioning as required, according to the criteria defined for Egyptian
soils.
For the Mehallet Rouh sub centre, it was observed that the average water table depth for the
whole irrigation cycle was 103 cm below soil surface. At the same time the average water
table depth was 103 cm below soil surface “6 days after irrigation”. This means that the water
table depth coincides with the criteria defined for Egyptian soils. This indicates that the
collectors under that sub centre were functioning well and were not in need of rehabilitation.
From the above-mentioned results, it can be said that, the water table draw down curve after
irrigation is a very important tool for defining to what extent recession takes place. If the
water table was still high 6 days after irrigation this means that the system was in need of
rehabilitation (main open drainage system is performing well). At the same time the open
drainage system had to be maintained because if there was any backward flow to the
subsurface drainage system it would have created many problems.

Performance Assessment and Potential Indicators for Drainage Systems 61

 0
BELKEEM
Soil Surface Hole 1
-20 Hole 2
Average

Water table depth (cm)

-40 Average cycle
-60

-80 87.5 84
7
-100

-120

-140
1 3 5 7 9 11 14 16 18
Days after irrigation

0
TUKH MAZYED
Soil Surface Hole 1
-20 Hole 2
Average
Average cycle
Water table depth (cm)

-40 47
-60 63

-80

-100

-120

-140
1 3 5 7 9 11 13 15 17
Days after irrigation

0
EKHNAWAY
Soil Surface Hole 1
-20 Hole 2
Average
Water table depth (cm)

-40 Average cycle

60.5
-60 68
-80

-100

-120

-140
1 3 5 7 9 11 13 15 17
Days after irrigation

0
SHUBRAKAS
Soil Surface Hole 1
-20 Hole 2
Average
Water table depth (cm)

-40 Average cycle

59
-60

-80
92
-100

-120

-140
1 3 5 7 9 11 13 15 17
Days after irrigation

0
MEHALLET ROUH
Soil Surface Hole 1
-20 Hole 2
Average
Water table depth (cm)

-40 Average cycle

-60

-80
103 103
-100

-120

-140
1 3 5 7 9 11 13 15 17
Days after irrigation

Figure 21: Water table recession in different sub centers of Santa area during irrigation cycle

Performance Assessment and Potential Indicators for Drainage Systems 62

 3.6. Salinity
The electrical conductivity (EC) in dS/m was measured simultaneously with the water table
depth in two auger holes for each sub centre of the Santa area. The location of the
measurement point was chosen at the midway between two laterals. The average values of
salinity ranged 1.59-1.69, 1.75-1.87, 1.72-2.64, 1.8-2.57 and 1.59-2.86 for Belkeem, Tukh
Mazyed, Ekhnaway, Shubrakas and Mehallet Rouh sub centres respectively. At the same
time, it was observed that salinity increased with time after irrigation. This comes from the
fact that the receding of the water table with time makes the concentration of salts in the
water table increase as a result of evaporation. The average salinity of ground water during
the study was about 2.0 dS/m in all sub centres. This means that the salinity of the water
table does not indicate any need for rehabilitation.

3.6.1. Spatial Distribution of Age and Complaints in different Drainage Directorates

(stage 2 of the study)
The distribution of complaints in the period 1993-1996, per 10,000 fed (Figure 22) shows that
for 75% of the observations the yearly number of complaints of the Drainage Centres is 35 or
less. The 90% cumulative frequency occurs at approx. 45 complaints per year, See
Technical Report 103 (DRI, 2000a).

70 100%

90%
60
80%
50 70%

60%
Frequency

40
50%
30
40%

20 30%

20%
10
10%

0 0%
5 10 15 20 25 30 35 40 45 50 55 60 65 More
Complaints per year per 10000 fed

Frequency Cumulative %

Figure 22: Frequency distribution of complaints for the period 1993-1996

3.6.2. Complaints related to the year of installation

The average year of installation of a Drainage Centre depends on the year of installation of
the individual Design Areas (contract areas) within each Drainage Centre. The average year
of installation was included in the complaints survey and was checked with the installation
data from the M&E database. From this database the installation year of the Design Areas
was taken and a weighted average was calculated (based on the size of the Design Areas)
for the Drainage Centre.

Performance Assessment and Potential Indicators for Drainage Systems 63

 40

35
Average complaints/yr (93-96)
per 10,000 fed 30

25

20

15

10

0
1960-1977 1978-1982 1983-1987 1988-1992 1993-1997
Construction Period

Figure 23: Complaints related to system age

The Drainage Centres were arranged according to their average installation year in five
groups: 1960-1977 (areas more than 20 years old), 1978-1982, 1983-1987, 1988-1992, and
1993-1997. An age of 20 years has been chosen because that is the period over which
farmers have to pay the instalments for their subsurface drainage system and it is considered
the lifetime of such a system in Egypt. For each group the average number of complaints in
the period 1993-1996 was calculated. From Figure 23, it is clear that the Drainage Centres
more recently constructed have fewer complaints than the older ones. The oldest areas,
constructed between 1960 and 1977, seem to have a similar number of complaints as the
systems of 10 to 15 years old. The reason for this could be that those areas are located in
the south of the Nile Delta, the highest part, with stable clay soils, where less problems with
water logging and salinity are to be expected and where no drainage envelopes are needed.

3.7. Conclusions and Recommendations

The main conclusion of the Workshop on Performance Assessment and Rehabilitation
(Smedema, Abdel Dayem et al. 1996) was that Rehabilitation can be considered equal to first
time installation and that to assess the need for rehabilitation water table depth and salinity
indicators can be used, with the standard design criteria as boundary values. The only
difference between rehabilitation and the installation of new systems is that the assessment
of complaints by farmers needs to precede the assessment of the other indicators.
The water table draw down curve after irrigation is a very important tool for defining to what
extent recession takes place. If the water table was still high 6 days after irrigation this means
that the system was in need of rehabilitation (main open drainage system is performing well).
At the same time the open drainage system must be maintained because if there were any
backward flow to the subsurface drainage system it would create many problems.
To select (part of) sub centre areas which are in need of rehabilitation, one could simply list
the Drainage Centres with the highest number of complaints per 10,000 fed and start the pre-
drainage survey in those areas. The result of the pre-drainage survey would actually
determine if rehabilitation is needed, based on the measured water tables and the soil
salinity.

Performance Assessment and Potential Indicators for Drainage Systems 64

Not all the Drainage Centres with a high number of complaints do necessarily have the oldest
Design Areas. The farmers in Egypt are paying for the installation of the subsurface drainage
systems. During 20 years after the construction, instalments are paid by the farmers to the
government through taxes levied by the agricultural co-operatives. The maintenance
departments of EPADP remain responsible for the maintenance of systems during their
lifetime. It would be difficult to rehabilitate (i.e. construct a new drainage system) in areas that
haven’t been repaid completely by the farmers, unless EPADP recognises that the original
drainage system was not designed, installed, or maintained well enough and waives the
farmers the remaining instalments.
The complaint information from EPADP used for this study was available only per Drainage
Centre. Each Drainage Centre contains older and newer Design Areas and it is not clear
which Design Areas generated the complaints listed for the Drainage Centre. A more
accurate complaint assessment can only be made when for each recorded complaint the
associated Design Area is also well known. In that case pre-drainage investigation could start
at specific Design Areas instead of in the whole Drainage Centre.
The following recommendations for drainage complaint assessment can be made:
1. Start a rapid appraisal in the Drainage Centres, which have an average age of more than
20 years. It is clear that that 75% of the yearly complaints observations of the Drainage
Centres have an average number of complaints of 35 or less. Therefore, based on the
data presented in Tec. Report No. 103 by DRI, as a first guideline, the following areas
should be selected for their investigation to rehabilitation: South Ibrahemia, Zagazeg,
Meet Ghamr, South Aga, Sheben el Kom, and South Kafr el Zayyat. This rapid appraisal
should include the visual inspection of the open drains (water level, weed, blockage),
manholes (water level, sediment, damage), fields (waterlogged, crop condition, salt) and
interviews with the farmers who logged most of the complaints;
2. Based on the rapid appraisal select the areas where the problems can not be solved with
(improved) maintenance;
3. Start a standard pre-drainage investigation in the selected areas in order to determine the
need for a subsurface drainage system. The decision should be based on the same
criteria, used in new areas (without a drainage system). If needed, install new drainage
systems as needed;
4. Implement the M&E improved complaint assessment system in all the areas provided
with subsurface drainage systems in Egypt. The complaint assessment system
developed by the M&E project includes detailed information on the location of the
complaint source and the reason for the complaint. The results are stored in the M&E
database.
5. Repeat the exercise with the improved complaints data from the M&E database. Show
the spatial distribution of the complaints from the Design Areas instead of the Drainage
Centres.

Performance Assessment and Potential Indicators for Drainage Systems 65

4. Morocco - Hydraulic System Indicators
The performance of surface and subsurface drainage systems is a major concern for the
managers of the irrigated zone of the Gharb plain (Morocco, approximately 100 km north of
Rabat). The irrigated (and drained) zone covers an area of 100 000 ha and includes 14500
km of drainage canals, subsurface collectors and PVC drains. The challenge of the
performance assessment is appreciated when the total length of the canals is compared to
the total length of irrigation canals (2300 km), or tracks (1500 km).
The Performance Assessment program has focussed on buried systems (drains and
collectors) and on open canals. In this paper we summarise the results on buried systems
and present our work on open canals

4.1. Buried drainage systems: collectors and drains

The irrigated zone is equipped with 8500 km of PVC drain pipes and with 1500km of
concrete culverts for collectors. Pipes are laid with a trench machine and culverts with a
mechanical shovel. Drains and collectors are connected through asbestos-cement manholes.

4.1.1. Collectors.
Our on farm enquiries revealed failures which are hereafter classified by the increasing order
of intensity of induced hydraulic disturbance:
● Broken manholes (frequent).

● Poor junctions between concrete elements of collectors, or broken elements (rare).

● Disrupted or clogged outfalls.
● Inverse slope on drains.
Manholes or concrete elements failures induce soil erosion above the pipe. The resulting
hole has a peculiar shape, which is well known to farmers. Such holes are used by farmers
as inlet wells in order to try to solve surface drainage problems. A consequence of this is an
excessive hydraulic head that prevents the downstream drains from operating at standard
flow conditions. Although this practice is ineffective in addressing surface drainage concerns,
it does provide a kind of flushing self-maintenance of the collector. Pits dug across the
collectors have shown low silting levels and they might have a level of performance close to
the original level (Frejefond 1993).
The question of invisible faults still remains. If they affect the performance, they might be
detected by use of enquiries, complaint management and, under certain conditions, by
remote sensing (Vincent, Vidal et al. 1996). If information is organised in a GIS, basic queries
may help detection of possible failures (for instance: root clogging due to fences installed
above the collector, roads or tracks constructed after the installation, etc.).

4.1.2. Drains
Several authors (Lahlou and Hamdi 1989; Frejefond 1993) Jellouli 1980 have reported
inverse slope. The relationships between efficiency and inverse slope are not clear. Bouarfa
et al., 1999, have shown discharge limitations on some plots of the Gharb drainage
experiment which seems to be due to local inverse drain slopes. These limitations, which
occur for specific discharges higher than the design discharge rate, are not shown to have a
significant influence either on the duration of the water table at drain mid distance, nor on the
yields of sugar cane.
Mineral clogging is not likely to occur. In this area the clogging is unlikely according to the
criteria as outlined by Vlotman (1999); This has been confirmed on the terrain by our

Performance Assessment and Potential Indicators for Drainage Systems 66

personal observations and those of the local manager, which are more or less described in
(Lahlou and Hamdi 1989; Frejefond 1993) Jellouli 1980.
Corrugated PVC drains, even if they were imperfectly installed, are still in good condition.
Restoring correct downstream flow conditions in the collectors and in the canals would
provide proper efficiency, close to the original. Thus, the performance and the maintenance
of collectors and drainage canals have to be examined first. The question of the collectors
has been summarised in the previous section; the following section shows what has been
done to assess the performance of drainage canals.

4.2. Performance of secondary drainage canals

Performance assessment of drainage canals is the concern of the local manager of the
Gharb zone, as he is charged with the maintenance of 4800 km of secondary and tertiary
drainage canals. Drainage canals are designed to maintain the water below the level of
buried collectors at the design discharge drainage rate.
A secondary canal has been chosen and the flow of this canal has been studied in order (1)
to modernise the hydraulic functioning of it, and (2) to simulate this flow under different
conditions of maintenance or of discharge. For this purpose, a free surface flow model has
been used (TALWEG/FLUVIA software; (Baume and Poirson 1984)

4.2.1. Introduction to the drainage canal under study.

The canal is located in the Oued Beth district, in the southern part of the Gharb zone, the
measurement points are easily accessible and downstream flow conditions at the junction to
the primary are well controlled (never submerged). The vegetation conditions are
homogeneous which allows us to consider a unique Manning-Strickler along the entire canal.
The cross section is a trapeze (width at the bottom 2m; depth 4 m, slope of the bank 1/1.5).
The total length is 3000 m. It shows 4 culverts for roads and tracks. This canal is
representative of canals present in the Gharb.
The following measurements have been carried out: discharges (current meter), water and
sediments depths, rainfall, size and characteristics of the culverts. We have drawn existing
cross and longitudinal sections (Figure 21) and have compared our findings with the initial
design study, which has provided the design discharge rate and the original longitudinal
section. The present bottom longitudinal profile has also been drawn (i.e. with sediments).

Assumptions for the model, calibration and validation are described in Vincent, Jamet et al.
(1999). The differences between simulated and measured waterlines for a given discharge
are low and it can be considered that the model provides a good image of the reality. We
have simulated various flow conditions for the design discharge rate - the canal when new,
and the canal under various conditions of sedimentation.
The design discharge rate is 2.2 m3/s upstream and 3.3 m3/s downstream.

Performance Assessment and Potential Indicators for Drainage Systems 67

 Culvert 1 :1800 L=20m
26,5

Actual bed height

25,5 Original bed height

24,5
Culvert 2 :1800 L=20m
Height (m)

23,5
Culvert 3 : 2*900 L=20m
Culvert 4 : 1800 L=20m

22,5

Culvert 5 : 1800 L=19m

21,5
0 500 1000 1500 2000 2500 3000
Cumulative distance (m)

Figure 24: Longitudinal section of the drainage canal.

4.2.2. Drainage design discharge and initial characteristics of the drainage canal
The hydraulic functioning of the canal has been tested using the initial size and
characteristics of the canal. Two simulations were computed - with and without the culverts
(Figure 25). The simulation without the culverts shows a stable and acceptable water level
below the bottom of the collectors. The simulation that included the culverts resulted in an
irregular and rising water level such that upstream of the culverts the collectors are
submerged. Thus, since its construction, the canal has been unable to convey the design
discharge.

Figure 25: Waterlines at initial conditions with and without culverted passages

The first step in checking the Performance Assessment of a canal is to compare its
compliance with design parameters. This step would also help to determine where it would
be sensible to undertake particular monitoring.

Performance Assessment and Potential Indicators for Drainage Systems 68

 4.2.3. Drainage design discharge and present condition of the canal
This simulation has been computed with a Manning Strickler coefficient of 5 and with the
present bottom longitudinal profile. This value comes from the calibration of the model and is
logical considering the poor condition of the canal. There is however overtopping of the canal
banks for that drainage discharge rate (Figure 26). Although calibration of the model is no
longer valid it still describes the canal functioning with frequent overtopping of the canal
banks at this point as has been reported by farmers.

Figure 26: Waterline at present conditions of sedimentation and vegetation

The overtopping of the canal is not due to the height of sediments in the canal but to a
reduction of the section of culverts owing to siltation both inside and around them. A
Performance Assessment program should first involve culvert maintenance operations. The
findings would then assist in decision making in the case of insufficient funds allocated to
canal maintenance.

4.2.4. Influence of the vegetation

Several simulations have been computed in order to reproduce the consequences of
vegetative growth at various levels of intensity. For this purpose it has been considered that
changes in the Manning Strickler coefficient could properly represent variations in vegetation.
Simulations have been computed with Manning-Strickler of 5, 15, 25 and 33.
It has been shown that in the upstream vicinity of the culverts the influence of the changes in
the Manning Strickler coefficient is limited: this is due to the influence of the culvert, which
raises the waterline and reduces the water velocity upstream of the structure. The roughness
induced head losses being a function of the flow velocity; incidences of the variation
coefficient are limited Figure 27.
The incidence of the Manning Strickler coefficient is obvious far away upstream of the
culvert, see for instance, Figure 27 - the reach between culvert 4 and 5 where the water lines
rise roughly 0.7m above their initial levels.

Performance Assessment and Potential Indicators for Drainage Systems 69

 27,5

Initial bed height

26,5 Water level with a Strickler of 5

Water level with a Strickler of 15

25,5 Water level with a Strickler of 25

Water level with a Strickler of 33

Height (m)
24,5

23,5

22,5

21,5
0 500 1000 1500 2000 2500 3000
Cumulative distance (m)

Figure 27: Waterline variation as function of the variation of the Manning Strickler coefficient
(trapezoidal reach)

A visual assessment of Figure 27 seems to represent a functioning threshold for a Manning-

Strickler value ranging between 5 and 15. These values correspond to extremely degraded
ditches. These ditches would easily recover acceptable performance if a substantial amount
of vegetation was removed.

4.2.5. Sediment influence

Variations in sediment height have been tested as well. Results are Vincent, Jamet et al.
(1999) and are not detailed here, as they do not show huge influences on the waterlines in
reaches between the present conditions and the originals. In the case under study, the strong
head losses through the partly clogged culverts mask the possible influence of the sediment
height.

4.3. Conclusions
The degree of vegetation seems to strongly influence the waterlines of highly contaminated
ditches. The role of sedimentation in reaches has not appeared as the parameter most
limiting flow. Thus the maintenance techniques for vegetated ditches should treat vegetation
rather than sedimentation.
The role of clogged culverts has been highlighted. Because of these restrictions, the ditches
are unable to transit drain design discharge rates and are now overflowing because of their
poor condition. Thus, all the culverts would benefit from priority treatment. In the case of a
limited budget we would recommend cleaning the culverts first, prior to the removal of
vegetation from the canal reaches and, lastly, the removal of all sediments.
In terms of design, we would recommend the use of parallel piped culverts whose height
would be the same as the diameter of the cylindrical culverts. Such parallel piped culverts are
known to be less prone to siltation.

Performance Assessment and Potential Indicators for Drainage Systems 70

5. Morocco - Crop yields in the Gharb Plain
In the Gharb plain of Morocco the performance of the drainage systems was a matter of
concern for the irrigation agency (Office Régional de Mise en Valeur Agricole du Gharb,
Zimmer, 1999 #25). Yields of the different crops produced in the area are routinely recorded
at different levels. The approach utilised [El Amraoui, 1998 #26] was comparable to that
presented above since in the Gharb Plain, drainage is mainly installed to reduce winter
waterlogging that occurs during rainy periods.
In the Moroccan context, the climate variability is much higher than in temperate climates.
Thus, yields can be reduced during both dry and wet years. The correlation between yields
and rainfall depths during winter months was determined for different crops. The best
correlation was obtained for the months November to January which have the highest rainfall
depths. Average rainfall for this period is 218mm with a standard deviation of 113mm Figure
28. Data from several years were compared with the average distribution which provided
return periods for the different rainfall depths. A similar index as that utilised in the above
example was computed by dividing each winter rainfall depth (for the three selected months)
by its average value. In the Gharb, an index of 1.5 has a return period of 6 to 7 seven years;
an index of 2 has a return period of 10 to 15 years. This index was then utilized to evaluate
(1) the effect of waterlogging on yields and (2) to compare the performances of different
sectors of the irrigated area.
1.20

1.00

0.80
frequency

0.60
observed
fitted to Normal Law
0.40

0.20
average value

0.00
0 100 200 300 400 500 600

Rainfall depth November-January (mm)

Figure 28: Cumulative distribution of rainfall depths of months November to January in the
Gharb region

Two typical examples of the results are presented in Figure 29 and Figure 30. In the case of
winter cereals, the yields are affected in dry as well as in wet years because these crops are
not irrigated. For other crops such as sugar beet or sugar cane which are irrigated, dry winter
seasons have no effect on the yields. This shows an important difference between drainage
and irrigation: whereas drought stress can be fully suppressed with irrigation, it would not be
cost-effective to totally remove waterlogging stress with a drainage system. Therefore, the
performance of drainage systems should be assessed using a sensitivity threshold that is
shifted more toward rainy seasons Figure 28. The corresponding increase in return periods
could then be compared to the design parameters of the project. Due to the limited number of
years available, this approach could not be carried out here.

Performance Assessment and Potential Indicators for Drainage Systems 71

 35.000

30.000 Winter wheat

fitted curve
25.000
yield (T/ha)

20.000

15.000

10.000

5.000
average value

0.000
0.0 0.5 1.0 1.5 2.0 2.5
rainfall index

Figure 29: Winter wheat yields versus rainfall index in the Gharb plain (d’après [El Amraoui,
1998 #26])

60.000

Sugar beet
50.000

40.000
Yield (T/ha)

30.000 fitted curve

20.000

10.000
average value

0.000
0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

Rainfall index

Figure 30: Sugar beet yields versus rainfall index in the Gharb plain (d’après [El Amraoui,
1998 #26])

Besides the management of extreme wet years it is also important that the maximum yields
obtained in “average years” be increased after project implementation. In most of the cases,
implementing a drainage system should induce modifications in the farming practices of the
farmers (better fertilisation or tillage practices for instance). A general increase in the yields
should therefore be an objective of the projects. This result has not clearly been achieved so
far in the Gharb Plain as can be determined by comparison of the maximum yields of drained
areas with those of nearby waterlogged areas [El Amraoui, 1998 #26], Bentiss

Performance Assessment and Potential Indicators for Drainage Systems 72

 before project after project
implementation implementation
(irrigated crop)

Yield
increase
yield

yield
dry years wet years dry years

sensitivity Rainfall sensitivity Rainfall

threshold index threshold index

Figure 31: Theoretical relationships between yield and rainfall index before and after
implementation.

Although somewhat difficult to use to determine absolute values, the yield versus index
relationships is very useful for comparing the relative performances of different sectors in a
given drained area. Several comparisons were carried out in the Gharb Plain (see example in
Table 11) which could help managers in two ways: (1) in helping to better focus the
maintenance or rehabilitation programs and (2) in selecting the crops best suited to the
prevailing conditions in the different sectors.

Table 11: Relative drainage performances of different sectors of the Gharb plain. Low figures
refer to classes of high sensitivity (from [El Amraoui, 1998 #26])

Area Mograne PTI STI Beht

Sensitivity to drought conditions
Winter cereals 4 3 2 1
Citrus - 2 1 3
Sunflower cropped area 2 3 1 3
Sensitivity to waterlogging condition
Winter cereals 1 2 2 2
Sugar beet 1 4 2 3
Sugar cane 1 3 2 -
Citrus - 2 1 0
Sunflower cropped area 1 2 2 0

Performance Assessment and Potential Indicators for Drainage Systems 73

6. France - Drainage Efficiency and Agronomic Outcomes
Although of interest for many drainage practitioners and donor agencies, land drainage
performance assessment (PA) is a rather new topic. Very few methodology papers dealing
with PA for the drainage sector have so far been published. Common views of the drainage
community regarding the definition and the fields of interest of PA are still to emerge and the
relevance of indicators is still to be discussed.
On the other hand the large body of experience existing within the drainage community on,
for instance, the hazards of dysfunctional drainage networks, drainage benefits for crops,
trafficability control, salinity etc. is very useful for the formulation of methods for land drainage
PA. For practical purposes, it should be kept in mind that PA should be applicable to large
areas and that it should rely on observable or measurable indicators which can be made
available easily (Bos 1996; Smedema and Vlotman 1996).
Different types of performance assessment programs were launched in France during the
1980’s. Indeed, at the beginning of that decade, subsurface drainage construction increased
dramatically to a rate of 110 to 130 thousand hectares per year which was then sustained for
almost ten years. This increase, combined with a lack of information regarding drainage
criteria, raised a lot of concerns among practitioners regarding the relevance, the efficiency
and the durability of the drainage techniques commonly utilized.
This paper firstly reviews the French PA programs and then discusses their complementarily
with research and development (R&D) programs. It then proposes and discusses two
complementary aspects of PA programs which are (1) the assessment of drainage efficiency
and benefits and (2) the determination of the factors explaining the performances.

6.1. Types and objectives of land drainage performance assessment

programs in France
PA programs were necessary in France because little R&D had been conducted before the
80’s on drainage efficiency and on drainage design criteria especially regarding drain spacing
calculation. The relevance of usual approaches referring to classical drainage formulae was
doubtful in the absence of field validation and owing to the high variability of the French
waterlogged soils. Moreover construction techniques were still evolving with the introduction
of better quality fittings and of more accurate grading control systems (Lesaffre 1990).
Most PA programs were therefore dictated by the needs of designers for precise references
on the efficiency of drainage design and construction techniques they were promoting. They
had however, to be conducted on field subsurface drainage systems that had often been built
in another era with different approaches and technologies. The need for rehabilitation of
existing drainage schemes which is one of the major objectives of PA programs (Smedema,
Abdel Dayem et al. 1996) was not a concern.

6.1.1. French PA programs launched by designers were of two main different types:
Local surveys of existing networks were implemented to diagnose the main reasons for poor
drainage function (mineral or iron clogging, construction deficiencies, poor maintenance etc.).
These surveys were part of the methodology developed in the framework of the so-called
“Reference Area Program” (Favrot 1984; Favrot 1987). The typical size of a Reference Area
was about 10000 ha; that is the size of a small agricultural region in France. They were
based on farmer enquiries as well as field observations by simple pit digging and system
examination. These surveys proved to be very relevant in providing feed-back from the field
as well as discussions between practitioners and farmers or between farmers themselves.
Regional programs aiming at improving the design and construction of subsurface drainage
were established after specific problems were diagnosed by drainage practitioners either in
terms of drainage malfunction or in terms of poor performance; they involved surveys and

Performance Assessment and Potential Indicators for Drainage Systems 74

field measurements without heavy measurement equipment; GIS techniques were used in
certain of these programs to help identify the key factors explaining the lack of performances
(Vincent, Frejefond et al. 1993).
Some PA programs were also initiated by farmer boards or associations. They were regional
programs aiming at the assessment of drainage benefits, both in terms of crop yields or in
terms of influences of drainage on farmer incomes. It is important to recall here that in
Western Europe, the farmer is the one who decides to invest in drainage improvements even
if his investment is in general subsidized. The history of drainage development shows that
the farmers need to be convinced of the interest of drainage improvements much more than
for irrigation. Irrigation is in general an obvious need for crops while drainage needs are more
indirectly perceived unless severe trafficability problems are faced. In this context PA
Programs were needed to precisely assess the benefits of drainage, increase the awareness
of farmers of the importance of waterlogging and possibly demonstrate to farmers the cost
efficiency of drainage improvements.

6.2. Research and Development and Performance Assessment Programs

PA programs need to be distinguished from R&D programs (Smedema and Vlotman 1996).
The distinction between PA and R&D programs is related (1) to the lack of possibility to
address all problems met in actual field conditions in field experiments and (2) to the need for
PA results to be quickly operationalised. This is, for instance, the case for construction or
maintenance aspects. PA programs are nevertheless complementary to R&D programs and
it would be often beneficial to start them at the same time, PA programs providing questions
from the field where many parameters are uncontrolled, R&D programs providing guidance
on specific problems which can be addressed in experiments. It should also be pointed out
that R&D programs on PA would be very relevant to test and compare indicators and
methods.
Another aspect relating PA to R&D programs is the fact that PA programs should not be
confined to the measurement of performance indicators but should also provide information
on the factors explaining the performances (Smedema, Abdel Dayem et al. 1996). The
peculiarities of these two components of PA programs are described below.

6.2.1. Assessment of drainage efficiency and benefits

Drainage efficiency can be defined as the level of performance measured against the
design objectives of the drainage improvements (i.e. is the water table satisfactorily
controlled with regard to the design criterion? or does the time needed to achieve trafficability
comply with the needs of the farmer?). Drainage efficiency assessment is carried out at the
system level.

Drainage benefits refer to crop yields, farmer income increases, cost reduction (i.e.
machinery costs or costs/benefits resulting from trade-offs between different components of a
farming system) or to environmental or socio-economic aspects. They are in general the
main objectives of drainage, but they are neither explicitly taken into account in the design
criteria nor measurable in a short period in situ. The time and space scales of the
investigations needed for the assessment of drainage benefits are large: they go far beyond
the system level and encompass the human and natural environment of that system; they
would often need specific monitoring programs.
The distinction between drainage efficiency and drainage benefits is more or less parallel to
that between operational and strategic performances as defined by Bos et al (Bos 1994c). As
defined by this author, operational performance is concerned with the level of service of the
equipment (drainage efficiency) whereas strategic performance is concerned with the
development (agricultural, human, financial etc.) authorised by the level of service achieved
(drainage benefits).

Performance Assessment and Potential Indicators for Drainage Systems 75

 6.2.2. Determination of the factors of drainage performances
Knowing the efficiency or benefits of the drainage improvements is not enough. PA programs
are conducted either when new drainage projects are planned or when it becomes
questionable as to whether rehabilitation is needed. They often aim at improving design,
construction or maintenance practices. Therefore it is important that parallel to the
assessment of the drainage efficiency, investigations be carried out to determine which are
the critical factors of efficiency.
The determination of the factors explaining drainage performances requires specific
investigations in general at field level.

6.2.3. Methods and indicators

Relevant methods and indicators depend very much on the types and objectives of PA
programs. A few examples of methods used and results obtained are presented below for (1)
drainage efficiency, (2) drainage benefits and (3) factors of drainage performances.

6.3. Drainage efficiency: case study in eastern France

The eastern part of France (Lorraine region) is characterized by widespread and highly
variable heavy clay soils. Drainage improvements started quite early in that region, i.e. in the
mid 70’s. As waterlogging is very critical for many farms of that region, designers and farmers
were much concerned about the subsurface drainage systems efficiency. A 3 year PA
program was initiated in 1987 as some doubt arose regarding this efficiency in certain types
of soils. The questions of the designers were twofold: (1) are the drainage performances
related to the type of soil, and if so, (2) should the drainage design be changed, or would it
be possible to prioritize drainage according to the type of soil?
The program had two major components as shown in Figure 32; (Zimmer 1991). In a first
stage, drainage networks were selected and their efficiencies determined; the method
comprised enquiries with the farmers as well as system diagnosis. In a second stage, the
systems investigated were classified according to drainage efficiency and type of soil and
water table regimes were measured by use of tensiometers while the farmer indicated his
tillage operations and his appreciation of the field trafficability.
For the first stage of the program, a key factor to success was the sampling of the fields to
investigate. The sample had to be representative of the different types of soils, the age and
the design of the drainage systems had to be well known and the farming practices (crop
rotations, tillage practices) had also to represent the actual practices. The field trafficability
assessment by the farmers proved to be very accurate in most cases but there was evidence
that their appreciation was to some extent influenced by the intensity of the labour shortages
faced during spring periods. The accuracy was ascertained by comparisons with naturally
well drained or poorly drained soils which helped when comparing the judgements between
the farmers. Tensiometer monitoring associated with trafficability judgement in the second
stage also demonstrated the accuracy of that assessment.

Performance Assessment and Potential Indicators for Drainage Systems 76

 Selection of drainage Farmer enquiries
systems to investigate - drainage needs
- trafficability
- performance
evolution

System diagnosis:
soil, crop, system,
components (pipes
outlets, couplings...)

Water table and

Selection of fields to
farmer's operation
be monitored
monitoring

Detailed soil
investigations

Figure 32: Example of performance assessment methodology used in a 3 year program in

France during the 80’s

One of the soils monitored during the second stage of the program was perceived as very
difficult to improve by subsurface drainage. The program included an experiment comparing
two drainage techniques, namely (1) conventional 10 m spacing laterals installed by trencher
and (2) mole drains discharging into 30 m spacing laterals. To facilitate the performance
assessment, the two plots were compared to a third experimental plot installed in a clayey
soil with good natural drainage. This naturally well drained soil was located in the vicinity of
the two others so that crops and farming practices could be identical in all three plots.
Water pressures were monitored at different depths between 20 cm and 1m for a few months
in spring 1988 (from March 18th until June 16th) within the three plots, Figure 33. The total
amount of precipitations was 285 mm during that period. The differences between the time
series of measurements of the three plots was not obvious; the tendency was that the
naturally well drained plot had lower water pressures during dry periods. During wet periods
the differences were not significant.

Performance Assessment and Potential Indicators for Drainage Systems 77

 20

0
Water pressure head (cm)

-20

-40

-60

-80

-100

-120
18/03/1988

28/03/1988

07/04/1988

17/04/1988

27/04/1988

07/05/1988

17/05/1988

27/05/1988

06/06/1988

16/06/1988
good natural drainage mole drainage trencher

Figure 33: Pore water pressure sequence at the bottom of the plough layer in three plots of a
pilot drained area in eastern France in spring 1988

The farmer’s tillage operations were recorded during the same period as well as his
comments on the field trafficability, Table 12. The trafficability conditions were quite different
in the three plots. From the data recorded, it appears that a threshold value of -20 to -25 cm
of pressure head at the bottom of the plough layer is necessary to achieve satisfactory
trafficability. This value compares well with the common criteria used in different countries for
trafficability. For example, in France, a depth of 0.45m of the water table is commonly used.
In the Netherlands, a water pressure head of -40cm at a depth of 0.05m is a common
criterion. These values are identical to a pressure head of -25cm at a depth of -0.2m if a
hydrostatic pressure head profile is assumed above a water table located at a depth of
0.45m.

Table 12: Farmer operation, water pressure heads and trafficability conditions in three plots
of clayey soils of the eastern part of France (spring 1988)

Operation Date Water pressure head, Farmer appreciation of

25 cm depth trafficability
Fongicide application 03.26 well drained: -17 cm Bad conditions in three plots
trencher: -12 cm
mole drainage: -12 cm
Fertiliser 04.01 well drained: -32 cm Good conditions
application trencher: -15 cm Bad conditions
mole drainage: -15 cm Bad conditions
Seedbed preparation 04.02 well drained: -32 cm Good conditions
trencher: -12 cm Bad conditions
mole drainage: -12 cm Bad conditions
Seedbed preparation 04.07 well drained: -57 cm Good conditions in three plots
trencher: -15 cm
mole drainage: -25 cm
Fertiliser application 04.15 well drained: -22 cm Difficult in three plots
trencher: -12 cm
mole drainage: -15 cm

Performance Assessment and Potential Indicators for Drainage Systems 78

The time series of water pressures was then re-examined by use of water pressure
exceedance duration curves Figure 34, the exceedance duration is expressed as a number
of measurements; 42 values were recorded every 2-3 days in a three months period). This
new representation of the data clearly evidences the differences between the three plots and
especially between the poorly drained plots and the well drained plot. Indeed the differences
between the naturally well drained and the poorly drained situations are very important
around the threshold value. Assuming the recorded values are representative of the three
months period, the trafficability of the naturally well drained soils was good for 60 to 70 % of
the period. That of the poorly drained soil was good only for 10 to 20% of the period.

20

-20
water pressure head (cm)

-40

-60 good natural drainage

mole drainage
-80
trencher

-100

-120
0 5 10 15 20 25 30 35 40 45
exceedance duration (number of measurements)

Figure 34: Exceedance durations curves of pore water pressures at the bottom of the plough
layer in three experimental plots in eastern France in spring 1988

The method could then be applied to all the monitored fields and provide relevant
classification of the performances. In general the water pressure analysis confirmed the
assessment made by the farmers. It also helped build a typology of the water pressure
profiles in heavy clay soils with practical applications for drainage design (Zimmer and
Lesaffre 1989). It was clearly shown in this program that dipwell measurements did not
provide accurate data in heavy clay soils; water table elevations calculated from tensiometer
data and dipwell measurements could in some places be totally different.
The conclusions of the program can be summarized as follows:
Provided farmers, soils and drainage system sampling is done carefully, farmers can provide
very relevant indicators of the performances of their networks; subsurface drained soils could
be ranked according to the delay needed to achieve trafficability after an average rainfall
period in early spring (2 year return period); farmer’s appreciations indicated that severe
constraints occur when this delay is more than 4 days;
Regarding PA indicators, it was shown that water table (or pore water pressure) regime may
be relevant only if adequately processed; exceedance duration curves provide relevant
indicators if threshold values for water table elevations or pore water pressures can be
determined;
Finally, a good insight of the relationships between drainage performances and soil types
was provided to drainage practitioners.

Performance Assessment and Potential Indicators for Drainage Systems 79

 6.4. Drainage benefits
The assessment of drainage benefits was carried out in France by farmer boards or farmer
associations. Indicators chosen were mostly crop yields and farmer income increases. In the
second case, a common method was to compare the average income of farmer groups
having or not having invested in drainage. This indicator is easy to collect but it suffers from
the following drawback: a higher income is not necessarily the effect of drainage
improvements; it can simply indicate that farmers who invest in drainage have a better
economic situation than those who do not invest. This drawback is a very general
characteristic of the difficulty encountered when assessing drainage benefits; the level of the
indicator depends on many other factors than drainage. A PA program based on drainage
yields measurements at the field scale will illustrate here these difficulties and ways to tackle
them.
The program was launched in the late 70’s in the French department of Eure et Loir located
100km west from Paris. Yield values were monitored by the technical services of the local
farmer board in several subsurface drained fields. At that time the benefits of drainage
improvements in the loamy alfisoils of the region had not been clearly demonstrated.
The field experiment of Arrou, which had been extensively monitored by Cemagref for 15
years, was included in the sample of fields monitored. Data presented below are from this
field experiment which is located in a farm representative of the farms of the region. Three
situations are compared: (1) without drainage improvement, (2) with 10 m spacing laterals
without secondary treatments (the common technique in use justified by the shallow depth of
the impervious barrier); (3) with 20 m spacing laterals without secondary treatment.
Crop rotations of the region include winter wheat or barley, maize and forage crops
(ryegrass, peas). Results for winter wheat, which is the most frequent crop in the rotation, are
presented in Figure 35. It should be noticed that a period of about 10 years was necessary to
obtain these data. During that decade, the humid seasons have been classified according to
a climatic indicator defined as the ratio between the rainfall depth of the season (December
to May) versus 30 year average rainfall depth for the same period.
When the rainfall depth is less than the average value, the yields are not affected by excess
water; in that case, other factors explain the variability of the yields. As soon as the rainfall
depth exceeds this average value, the yields are dramatically reduced in the undrained plots.
No significant differences are noticeable between the spacing of 10 and 20 meters.

Performance Assessment and Potential Indicators for Drainage Systems 80

 8
7
6
5
Yield
4
(T/ha)
3
No drainage
2 Subsurface 10m
1 Subsurface 20m
0
0 50 100 150 200 250
Rainfall index %

Figure 35: Yield values of winter wheat for improved and non improved drainage conditions
versus rainfall index in Arrou’s field experiment (France). Index is the ratio
between the rainfall depth of the rainy season (December to May) and average
rainfall depth for the same period.

The practical use of these results would require statistical analysis of the climatic data to
determine the frequency of occurrence of dry and wet winter seasons. Besides, in the
absence of data for undrained situations, it would be necessary to investigate whether or not
the results are related to drainage improvements. For this purpose, two types of indicators
have been tested in that program: (1) yield differences between trench and drain mid-spacing
locations and (2) yield components.

6.4.1. Yield differences between trench and mid-spacing locations

Yield differences between trench and drain mid-spacing location proved to be very relevant
indicators of the effect of drainage on yields. If excess water is totally removed by drainage,
then the differences in yield between the two locations should be very little. Differences in
yields between two fields can be attributed to drainage if they are comparable to the
differences between trench and drain mid-spacing locations in the field with the lowest yield
value.
In the field experiment of Arrou, the differences of yields between trench and mid-spacing
locations of two plots with lateral spacings of 10 and 20m have been compared for the same
series of years as in figure 4. They are represented in Figure 36 versus the same climatic
index as that of figure 4. For the 10m spacing laterals, no differences in yields between
trench and drain mid-spacing locations are observed except for rainfall indexes as high as
200 (twice the average rainfall depth during winter and early spring months). For the 20m
spacing laterals, differences are measured even at low rainfall indexes, thus showing that
some excess water occurs with that spacing even for rather dry winter seasons.
These results have two practical consequences: (1) they demonstrate that the differences in
yields between the two spacings are related to drainage, and (2) they allow a precise
evaluation of the theoretical benefits of drainage as related to the rainfall depth of the winter
and to the drain spacing.

Performance Assessment and Potential Indicators for Drainage Systems 81

 1.2
Subsurface 10m
1
Subsurface 20m
0.8
Yield
differences 0.6
(T/ha)
0.4

0.2

0
0 50 100 150 200 250
Rainfall index %

Figure 36: Yield differences of winter wheat between drainage trench location and drain mid-
spacing versus rainfall index in Arrou’s field experiment (France). Each value is
the average of 10-20 samples of 1 m²

6.4.2. Yield components

Yield components refer to the different organs of a plant contributing to its yield. For example,
yield components of winter wheat crops are (1) the plant density, measured at the end of the
winter season, (2) the average number of spikes per plant, (3) the average number of grains
per spike and (4) the average weight of the grains. From plant physiology considerations,
one can assume that the different yield components are good indicators of the conditions that
prevailed during the corresponding stages of the development of the crop. Several regional
programs have been launched in France to evaluate the potential yields of crops in different
types of soils by use of yield component measurements. These indicators were also
measured in the program launched in the Eure et Loir department. An example of the results
obtained in Arrou is presented in Table 13.
Under French soil and climate conditions, the first three yield components of winter wheat are
mostly affected by winter and spring excess water. The fourth one is, on the contrary, mostly
affected by summer water shortages. From the example of Table 13, there is evidence that
the crop of the undrained plot suffered from poor drainage conditions throughout the winter
and early spring seasons. When regional references are available, it is quite easy to attribute
any yield increase to drainage conditions if these three yield components have not been
affected.
Yield components or yield differences between trench and mid-spacing locations are relevant
to identify whether yield increase can be attributed to drainage improvements. These
indicators might be out of reach on a routine basis in PA programs. They might however, be
available through other programs carried out by agronomists. In case they are not, they
ought to be measured at least in a sample of representative fields included in the program.

Performance Assessment and Potential Indicators for Drainage Systems 82

Table 13: Yield components of winter wheat measured in drained and undrained plots of the
field experiment of Arrou, France in 1984-1985 (Guerin 1986).

Period of (a) subsurface (b) no Ratio drainage

measurement Drainage (10m Subsurface vs no drainage
spacing drainage (a)/(b)
laterals)
Plants/m² End of winter 249 226 1.10
Spikes/plant End of winter 1.81 1.48 1.21
Grains/spike Early spring 40 30 1.33
1000 grains weight (g) Early summer 39 37 1.05
Yield T/ha Summer 6.5 3.0 2.16

6.4.3. Factors for drainage performance

Although very crucial for drainage PA programs, there is no simple diagnosis method or tool
available for the determination of the factors of the performances. This determination relies
upon expertise or research. Poor performances may result from several deficiencies related
to design, materials, construction and maintenance which are not easy to diagnose several
years after drainage construction. Thus thorough examination of the system components (at
least of selected ones) is needed. If this examination does not provide satisfactory
explanation, which will quite often be the case, then drainage function analysis is required.
The diagnosis method should rely upon a very good theoretical expertise of drainage
functioning. Relationships between drainflow rates and water table elevations in different
locations of the field could provide the basis of this diagnosis.
It should also be stressed that quality assurance has to be improved in drainage projects.
Quality assurance starts with proper choice of good quality drainage materials and thorough
follow-up of the construction. It also relies upon good records of all characteristics of the
drainage networks including construction plans.

6.5. Conclusions
Drainage performance covers at least three different aspects: drainage efficiency, drainage
benefits and factors of drainage performance.
(1) Drainage efficiency is related to the hydraulic functioning of the drainage system and to
the satisfactory achievement of drainage criteria. Drainage efficiency refers to
operational performance. Required PA indicators should reflect as closely as possible
the characteristics of this functioning.
(2) Drainage benefits are related to the improvements of the farming and socio-economic
system allowed by the drainage program. Drainage benefits refer to strategic
performance. PA indicators are not directly linked to drainage functioning and it should
be ascertained that the measured benefits actually result from the drainage
improvements.
(3) Determination of the factors of drainage performances is needed whenever the level
of service does not match the expectations. Simple methods for this determination still
need to be developed.
Other aspects that can be pointed out from the experience gained in France during the 80’s
are as follows: (1) farmers usually have a good knowledge of the efficiency of their drainage
networks and relevant PA programs could start with farmer enquiries. (2) PA programs derive
much benefit from R&D programs and also from farm monitoring programs. Research results
may be important for drainage performance assessment since they provide a general
framework for interpretation of the field observations. Having a field experiment included in
the program is very helpful since it is possible to investigate more precisely the effects of

Performance Assessment and Potential Indicators for Drainage Systems 83

drainage; (3) the sampling of subsurface drainage systems and farms to be investigated in
the PA program is very critical. Although still under development, remote sensing techniques
might be very useful for that purpose (Ferrante 1995; Vincent, Vidal et al. 1996).

6.6. Acknowledgements
The author acknowledges the contribution of the “Chambre d’Agriculture d’Eure et Loir” who
provided some of the data utilized in this paper as well as that of M. Renat, private consultant
who was very helpful in the PA program on heavy clayey soils in Lorraine. The paper also
benefited from the comments received from various drainage colleagues.

Performance Assessment and Potential Indicators for Drainage Systems 84

7. Australia - Salinity Control in Vineyards
This case study is from the Murrumbidgee Irrigation Area in New South Wales, Australia and
tracks the establishment of a vineyard on a very saline soil. The vineyard had subsurface
drainage installed at 20m spacing at approximately 2m depth. For a detailed description of
the drainage system and layout refer to Hornbuckle et al. (2002).

7.1. Site Details

The site was located in the Murrumbidgee Irrigation Area (MIA) in New South Wales,
Australia on a previously un-irrigated area. The MIA is located in a semi-arid environment,
where summers are characterized by hot, dry conditions and winters are mild. Mean annual
rainfall for the area is 418mm and ranges between 140 –700mm. Potential evapo-
transpiration is around 1800mm. The MIA is supplied with irrigation water diverted from the
Murrumbidgee River, supplied by large catchment dams located in the Snowy Mountains.
Irrigation water is of low salinity and typically ranges from 0.05-0.15 dS/m.
The surface soil is a shallow clay loam passing quickly through to light clay. The grey subsoil
appears at a 0.75m depth and continues to more than 2m with increasing heaviness. Both
soft and hard carbonates are present.
The 1.2 ha site was very saline having been an un-irrigated parcel of land surrounded by
irrigated areas. The average soil salinity from 22 cores to 1.8m deep is shown in Figure 37.

Soil Salinity ECe (dS/m)

6 7 8 9 10 11 12 13 14
0

-0.2

-0.4

-0.6
Depth (m)

-0.8

-1

-1.2

-1.4

-1.6

-1.8

Figure 37: Soil salinity, average of 22 cores

It can be seen that there is a distinct increase in soil salinity with depth, which is typical of
saline soil profiles in the area, Hornbuckle and Christen (1999).
The site was installed with subsurface drains 1.8m deep and 20m apart. These were 100 mm
diameter slotted corrugated pipe with a gravel envelope. The site was first irrigated in
September 2001 with four irrigations being applied before planting with Semillon grapevines
in July 2002. Irrigations were applied through a furrow system with furrows located each side
of where the future vine row was to be situated.

Performance Assessment and Potential Indicators for Drainage Systems 85

 7.2. Salinity assessment using Electromagnetic survey technique
Soil salinity distribution over the vineyard was measured using the GeonicsTM EM 38 device
following the standard operating procedures outlined by McNeil (1986). The site was
rectangular being 60m wide and 200m long, and the survey was undertaken in a grid format
down every second row of vines to give a 6m by 6m grid, as shown in Figure 38, resulting in
270 data points. The EM38 instrument was used in the vertical (EMv) and horizontal modes
(EMh). The EMv mode is more sensitive to salinity in depths below 0.35m whereas the EMh
is more sensitive to salinity at depths above 0.35m (McNeil, 1986).

Figure 38: EM38 Survey grid

Three surveys were taken, one before the drains were installed and then at yearly intervals after
the installation. The data from the EM38 surveys were then plotted using SURFER 8.0TM to
provide a spatial map of the soil salinity in the vineyard, Figure 39 and Figure 40.
The surveys show that before the drainage there was a wide distribution of soil salinities across
the vineyard. The highest salinities being near the bottom end as shown by the red ‘hot spots’.
The surveys in the two modes show that the soil has a much higher salinity with depth as shown
by the EMv results in Figure 39. The surveys show that over time the salinity in the soil is
reduced, until after two years the soil salinity in the upper layers is uniformly low, Figure 39. The
salinity in the deeper layers is not reduced as much but most of the ‘hot spots’ have declined
markedly.

Performance Assessment and Potential Indicators for Drainage Systems 86

160 160 160
Uncalibrated EMh

540
520
140 140 500
140
480
460
440
420
400
120 120 120 380
360
340
320
300
280
100 100 100 260
240
220
200
180
160
80 80 80 140
120
100
80
60
60 60 60

0 20 40 0 20 40 0 20 40

16/06/2000 16/05/2001 03/06/2002

Figure 39: EM38 vertical reading for three dates (ECa in ms/cm).

160 160 160

Uncalibrated EMv

480
460
140 140 140
440
420
400
380
120 120 360
120
340
320
300
280
100 100 100 260
240
220
200
180
80 80 80 160
140
120
100
80
60 60 60

0 20 40 0 20 40 0 20 40

16/06/2000 16/05/2001 03/06/2002

Figure 40: EM38 horizontal reading for 3 dates (ECa in ms/cm).

These results give a qualitative assessment of the reduction in soil salinity, spatially, but do not
provide a quantitative estimate of the reduction in the soil salinity and hence the effectiveness of

Performance Assessment and Potential Indicators for Drainage Systems 87

the drainage system. The EM38 survey data need to be calibrated before the effectiveness of
the system can be determined. The calibration was done by processing the data from the EM38
survey using the ESAP (ECe Sampling, Assessment and Prediction) software from the U.S.
Salinity Laboratory, California (Lesch, et al. 2000), (website
http://www.ussl.ars.usda.gov/models/esap-95.htm).
Using the ‘RSSD’ component of ESAP an optimal soil sampling design was created based on
the conductivity survey data. This design resulted in the recommended 12 points distributed as
shown in Figure 41. Soil cores were taken to a depth of 1.8m at 0.3m intervals at these sites.
These samples were then analysed for electrical conductivity in a saturation paste (ECe).
The soil sample salinity data were then used to calibrate the EM38 data using the ‘ESAP-
Calibrate’ component of the software. The multiple linear regression function was used to
convert the soil salinity data for each soil layer, the results of the predicted versus measured ECe
can be seen in Figure 42.

Figure 41: Optimal soil sampling design provided by ESAP-RSSD

Performance Assessment and Potential Indicators for Drainage Systems 88

Figure 42: Predicted versus measured ECe using ESAP-Calibrate (different colours refer to
sample depth)

Using this calibration the original EM38 survey data was calibrated to provide a spatial salinity
map of ECe. The effectiveness of the drainage system in promoting plant growth was assessed
based on the yield potential estimated from the soil salinity data. To do this, the calibrated EM38
salinity data for the rootzone (0.8m) were averaged and used to predict the relative yield
potential using a Maas and Hoffman (1977) crop salt tolerance function. The yield potential was
then mapped for each survey, Figure 43.
This mapping shows that initially the area was very saline and would have had none or very poor
growth. After the first year the reclamation was proceeding well and the farmer planted the
grapevines. By the third year most of the area was reclaimed apart from a section at the bottom
corner. These results were borne out by the growth of the grapevines. In the red areas the vines
struggled to survive after planting. However, after the third year the vines in the whole area were
growing well.

Performance Assessment and Potential Indicators for Drainage Systems 89

160 160 160
Relative Yield %

100
95
140 140 140 90
85
80
75
70
120 120 120 65
60
55
50
45
100 100 100 40
35
30
25
20
80 80 80 15
10
5
0
-5
60 60 60

0 20 40 0 20 40 0 20 40

16/06/2000 16/05/2001 03/06/2002

Figure 43: Grapevine relative yield potential using Maas and Hofman (1977) crop salinity
tolerance function

7.3. Conclusion
Using this methodology it is possible to have spatial estimates of salinity that are useful in
assessing drainage performance. From this data we can see that the drainage system
performance in this case was excellent in reducing salinity both quickly and uniformly across
the area. The use of the crop salt tolerance function provides an indication of the effect of the
salinity on yield and hence the economic value associated with drainage performance.

7.4. Recommendations
From this and other work the authors highly recommend the use of electromagnetic surveys to
assess the spatial variability in soil salinity. This knowledge is useful before the installation of
drainage in the design phase as well as in assessing drainage performance after installation as it
can guide where drainage is most required.
In this example the drainage reclamation was uniform, in other situations it is not and the spatial
survey can provide indications as to where further drainage work is required. This can be seen in
an example from an irrigated pasture farm. Figure 44 shows an EM38 survey three years after
the subsurface drainage was installed. The blue area in the left of the survey is fully reclaimed
and drainage can be reduced to a management level. In the rest of the area where the drains
were installed the soil salinity is low, but there are some ‘hot spots’ remaining. These hot spots
can be addressed by a range of actions such as installing more drainage, deep ripping or the
addition of gypsum. These results can also be used to indicate the type of management that
should be applied to different sections of drainage depending upon their salinity and water table
status, Christen et al (2002).

Performance Assessment and Potential Indicators for Drainage Systems 90

 300
290
280
270
260
250
240
230
220
210
200
190
180
170
160
150
140
130
120
110
100
90
80
70
60
50

Figure 44: EM38 survey of irrigated pasture three years after subsurface drainage
installation (ECa ms/cm) Note: Yellow lines are subsurface drainage pipes

Performance Assessment and Potential Indicators for Drainage Systems 91

REFERENCES
ANCID (1999). Australian Irrigation Water Provider - Benchmarking Report for1998/99. Victoria, Australia:
69 pp.
ANCID (2002). Australian Irrigation Water Provider - Benchmarking for 2000/2001. Victoria, Australia: 4 pp
+ CD-ROM with all data.
Bakr, M., M. Abdel Fattah, et al. (1998). Design Criteria Mashtul Pilot Area (1985 - 1995). Cairo, Drainage
Research Institute (DRI): 50.
Baume and Poirson (1984). THALWEG-FLUVIA, Cemagref.
Bos, M. G. (1994). Suggested Indicators for Assessing Irrigation and Drainage Performance in Egypt.
Report to the Drainage Panel. Wageningen, The Netherlands, ILRI: 28.
Bos, M. G. (1996). Performance indicators for irrigation and drainage. Workshop on the evaluation of the
performance of subsurface drainage systems, 16th ICID Congress, Cairo, Egypt, ICID-Cemagref.
Bos, M. G., S. Abdel -Dayem, et al. (1994). Assessing Performance of Irrigation and Drainage
Performance: Examples from Egypt. 8th IWARA World Congress on Water Resources, Cairo, Egypt.
Bos, M. G., D. H. Murray-Rust, et al. (1994). “Methodologies for assessing performance of irrigation and
drainage management.” Irrigation and Drainage Systems 7(4): 231-261.
Christen, E.W, Hornbuckle, J.W. and Herath, S. (2002). Management options to reduce salt loads from tile
drainage in the Campaspe West District. CSIRO Land and Water, Technical Report, CSIRO Land
and Water, Griffith, NSW, Australia.
Diederen, P. J. M., Bunte, F., van Staalduinen, L., Huygen, A. And Uytewaal E. (2002). De prijs van water;
een financiele en institutionele analyse van het waterbeheer in Nederland. Rapport 3.02.01 (in Dutch,
The price of water; a financial and institutional analysis of water management in the Netherlands).
Den Haag, Nederland, 01 Landbouw Economisch Insitute (LEI),: 71 pp.
Dieleman, P. J. and B. D. Trafford (1976). Drain Testing. FAO Irrigation and Drainage Paper 28. Rome,
Italy, Food and Agriculture Organisation of the United Nations: 10-20.
Eissa, M. (2000). Watertable Indicators for Pilot Area Research. Drainage Research Project. Cairo, Egypt,
Drainage Research Institute.
Eissa, M., P. J. Hoogenboom, et al. (1996). Determination of Q-H Relations of Field Drains under Egyptian
Conditions. Workshop on the Evaluation of Performance of Subsurface Drainage Systems, 16th ICID
Congress, Cairo, Egypt, ICID-Cemagref.
Favrot, J. C. (1984). Acquisition des données nécessaires au draiange par la méthode des secteurs de
références. 12th ICID Congress, Colorado USA, ICID.
Favrot, J. C., et Lesaffre, B., (1987). Défautes de fonctionement et réhabilitation des réseaux de drainage
souterrains. 13 th ICID Congress,, Csablanca, Morocco, ICID.
Ferrante, J. E. T. (1995). “Situation of remote sensing in Brazil and its application to irrigation, drainage
and salinity problems.” Water report 4: 165-172.
Frejefond, E. (1993). Evaluation des performances des réseaux de drainage. Mise au point d'une méthode
combinant télédétection et diagnostic de terrain dans la plaine du Gharb-Maroc. Strasbourg,
ENGEES: 139.
Guerin, T. (1986). Diagnostic du risque de lessivage d'azote en fin de campagne culturale. Comparaison
d'une parcelle drainée et d'une parcelle non drainée., Cemagref.
Hachicha, M., Bahri, A., Zimmer, D. (2001). Effets à long terme de deux périmètres irrigués de la Basse
Vallée de la Medjerda. Economie de l'eau en irrigation, 14-16 novembre 2000, Hammamet.
Hoffman, G.J. 1990. Leaching fraction and root zone salinity control. In K.K. Tanji (ed). Agricultural Salinity
Assessment and Management. ASCE Manuals and Reports on Engineering No. 71 pg 237-261. New
York, N.Y.Lahlou, O. and Hamdi (1989). Problématique du drainage dans le Gharb. Kenitra,
ORMVAG: 1-13.

Performance Assessment and Potential Indicators for Drainage Systems 92

Hornbuckle, J.W. and Christen, E.W., (1999), Physical properties of soils in the Murrumbidgee and
Coleambally Irrigation Areas, CSIRO Land and Water, Technical Report 17/99, CSIRO Land and
Water, Griffith, NSW, Australia
Hornbuckle, J.W., Christen, E.W and Faulkner, R. (2002). Preliminary Investigations of a Multi-level
Subsurface Drainage System for Improving Drainage Water Quality. Irrigation 2002, Irrigation
Association of Australia Conference, 21-23 May 2002, Sydney.
Kielen, B. M., J. (2002). "New FAO Irrigation and Drainage Paper on Water Quality."
Lesaffre, B. (1990). Fench programm for technology research in drainage. Some illustrations. 7th Annual
World Bank Seminar, Baltimore, USA, World bank.
Lesch, S.M., J. Herrero, and J.D. Rhoades. 1998. Monitoring for temporal changes in soil salinity using
electromagnetic induction techniques. Soil Science Society America Journal. 62(1): 232-242.
Lesch, S.M., J.D. Rhoades, and D.L. Corwin. 2000. The ESAP-95 version 2.0 R. user manual and tutorial
guide. USSL Research Report # 146.
Maas, E.V. and G.J. Hoffman. 1977. Crop salt tolerance – current assessment. Irrigation and Drainage
Divison, ASCE. 103, pg 115-134.
Malano, H. (2003). Report of Task Force (TF4) on Benchmarking of Irrigation and Drainage Projects. New
Dehli, India, ICID: 19 pp.
Malano, H., Burton, M. (2001). Guidelines for benchmarking performance in the irrigation and drainage
sector. Rome, FAO-IPTRID.
McNeil, J.D., (1986), Geonics EM 38 ground conductivity meter. Operating instructions and survey
interpretation techniques. Technical Note TN-21, Geonics Ltd., Canada
Oosterbaan, R. J. (1994). Agricultural Drainage Criteria. Drainage Principles and Applications, ILRI
Publication 16. H. P. Ritzema. Wageningen, The Netherlands, ILRI: 635-690.
Oster, J.D. and J.D. Rhoades. 1990. Steady state root zone salt balance. In K.K. Tanji (ed) Agricultural
Salinity Assessment and Management. ASCE Manuals and Reports on Engineering No. 71 pg. 469 -
481. New York, N.Y.
Rady, M. A. (1993). Main policies and guideline for rehabilitation of tile drainage networks. Cairo, Egypt,
Egyptian Authority for Drainage Projects: 24.
Rajad Project Staff (1995). "Analysis of Subsurface Drainage Design Criteria, 1995 Edition. Rajasthan
Agricultural Drainage Research Project (Rajad), Chambal Command Area Development, Rajasthan,
India. Sep 95."
Rhoades, J.D., Chanduvi, F. and Lesch, S (1999) Soil salinity assessment Methods and interpretation of
electrical conductivity measurements. FAO Irrigation and Drainage Paper no. 57. Rome.
Salman, A. F. (1995). Rehabilitation policies and priorities of field drainage. Cairo, Egypt, Egyptian
Authority for Drainage Projects: 18.
Smedema, L.K., Vlotman, W.F and Rycroft, D.W. 2004. Modern Land Drainage. Planning, Design and
Management of Agricultural Drainage Systems. A.A. Balkema Publishers, Leiden, The Netherlands,
446 pp.
Smedema, L. K., S. M. Abdel Dayem, et al. (1996). Performance assessment of land drainage systems.
Workshop on the Evaluation of Performance of Subsurface Drainage Systems, 16th ICID Congress,
Cairo, Egypt, ICID-Cemagref.
Smedema, L. K. and W. F. Vlotman, Eds. (1996). Workshop on Performance Assessment of Drainage
Systems. Cairo, Egypt, Drainage Research Institute (DRI) and Egyptian Public Authority for Drainage
Projects (EPADP).
Smedema, L.K. 1988: Watertable control indices for drainage of agricultural land in humid climates.
Agricultural Water Management 14, p 69-77.
Smedema, L. K., Abdel Dayem, S.M., Vlotman, W.F., Abdel Aziz, A., Van Leeuwen, H. (1996). Key note
address for Performance assessment of land drainage systems. Workshop on the Evaluation of
Performance of Subsurface Drainage Systems,, Cairo, Egypt, CIID-Cemagref.

Performance Assessment and Potential Indicators for Drainage Systems 93

SURFER, 1986. Reference Manual. Golden Software Inc. Golden, CO.
U.S. Salinity Laboratory Staff. 1954. Diagnosis and Improvement of Saline and Alkali Soils. USDA
Handbook No. 60. 160 pgs.
VEWIN (2001). Reflections on Performance 2000; Benchmarking in the Dutch Drinking Water Industry.
Rijswijk, The Netherlands., VEWIN, the Netherlands Waterworks Association: 77 pp.
Vincent, B., E. Frejefond, et al. (1993). drainage performance assessment by use of remote sensing
techniques in irrigation and drainage., Montpellier, Cemagref-FAO.
Vincent, B., A. Vidal, et al. (1996). Use of satellite remote sensing as an indication for the assessment of
waterlogging or salinity as an indication of the performance of drained systems. Workshop on the
Evaluation of Performance of Subsurface Drainage Systems, 16th ICID Congress, Cairo, Egypt, ICID-
Cemagref.
Vincent, B., Jamet, L. Baqri, A. Chaumont, C. (1999). Performances hydrauliques des systèmes de
drainage et d'assainissement. Atelier International sur le drainage de la plaine du Gharb, Rabat,
Actes éditions.
Vincent, B., Kchouk, A., Hachicha, M. (2001). Evaluation de la performance des systèmes de drainage
dans le cas du périmètre irrigué de Kalaât Landelous. Economie de l'eau en irrigation, 14-16
novembre 2000, Hammamet.
Vincent, B., A. Vidal, et al. (1996). Use of satellite remote sensing as an indication for the assessment of
waterlogging or salinity as an indication of the performance of drained systems. Workshop on the
Evaluation of Performance of Subsurface Drainage Systems, 16th ICID Congress, Cairo, Egypt, ICID-
Cemagref.
Vincent, B., Vlotman W.F., Zimmer D. (2002). Performance assessment and potential indicators for
drainage systems; Draft Publication ICID Working Group on Drainage.
http://www.icid.org/w_body.html#22, then go to DRAINLINE.
Vlotman, W. F. (1999). Drain envelopes. Wageningen, ILRI: 10-29.
Vlotman, W. F., Bhutta, M.N., Syed Rehmat Ali Shah, and Bhatti, A.K. (1994). Design, Monitoring and
Research (Supporting Report) Fourth Drainage Project, Faisalabad, 1976 - 1994. Lahore, Pakistan,
IWASRI Publication No. 159, International Waterlogging and Salinity Research Institute, Lahore,
Pakistan,: 260+ pp.
Vlotman, W. F., Jansen H. J., Snellen, W. B. (2002). Controlled Drainage; an additional paradigm for
integrated water management (in press). 7th Netherlands ICID meeting, April 2002, Nijmegen, The
Netherlands.
Vlotman, W. F. and Bhutta, M. N (2002). Case Study Fourth Drainage Project, Pakistan (Draft). In: Vincent
et al. 2002. Performance assessment and potential indicators for drainage systems; Draft Publication
ICID Working Group on Drainage. http://www.icid.org/w_body.html#22, then go to DRAINLINE.
Zimmer, D. (1991). “Evaluation de la qualité de fonctionnement des réseaux de drainage, application aus
sols argileux gonflants de Lorraine.” Etudes du Ceùagref, hydraulique agricole 10: 17_46.
Zimmer, D. and B. Lesaffre (1989). Subsurface drainage flow patterns and soil types. ASAE summer
meeting, Quebec, ASAE.
Zimmer, D. (1996). Lessons gained from the land drainage performance assessment programs in France
during the 80's. Workshop on the Evaluation of Performance of Subsurface Drainage Systems, 16th
ICID Congress, Cairo, Egypt, ICID-Cemagref.

Performance Assessment and Potential Indicators for Drainage Systems 94