eBook
Available on Compatible NOOK devices, the free NOOK App and in My Digital Library.
Related collections and offers
Overview
COMPREHENSIVE coverage of all aspects of public water supply and treatment for a worldwide market * INCLUDES more coverage of US, European and WHO standards and practice * COVERS important current topics such as EIA and cryptosporidium outbreaks
Product Details
| ISBN-13: | 9780080520803 |
|---|---|
| Publisher: | Elsevier Science |
| Publication date: | 06/06/2000 |
| Sold by: | Barnes & Noble |
| Format: | eBook |
| Pages: | 712 |
| File size: | 10 MB |
About the Author
Civil engineer specialising in water supply planning and distribution system management; responsibilities include research in the UK and overseas as well as project execution in Europe, Asia, Middle East and North and South America.
Read an Excerpt
WATER SUPPLY
By Alan C. Twort Don D. Ratnayaka Malcolm J. Brandt
Butterworth-Heinemann
Copyright © 2000 A. C. Twort, D. D. Ratnayaka and M. J. BrandtAll right reserved.
ISBN: 978-0-08-052080-3
Chapter One
Public water supply requirement and its measurement
1.1 Categories of consumption
It is useful to divide public water consumption into the following categories.
(1) Domestic
In-house use – for drinking, cooking, ablution, sanitation, house cleaning, laundry, patio and car washing.
Out-of-house use – for garden watering, lawn sprinkling and bathing pools.
Standpipe use – from standpipes and public fountains.
(2) Trade and industrial
Industrial – for factories, industries, power stations, docks, etc.
Commercial – for shops, offices, restaurants, hotels, railway stations, airports, small trades and workshops, etc.
Institutional – for hospitals, schools, universities, government offices, military establishments, etc.
(3) Agricultural
Agricultural use is for crops, livestock, horticulture, greenhouses, dairies, farmsteads.
(4) Public
Public use is for public parks, green areas, street watering, water mains and sewer flushing, fire-fighting.
(5) Losses
Distribution losses – leakage from mains and service pipes upstream of consumers' meters or property boundary; leaks from valves, hydrants and washouts, leakage and overflows from service reservoirs.
Consumer wastage – leakage and wastage on consumers' premises and from their supply pipes, misuse or unnecessary use of water by consumers.
Metering and other losses – source meter errors, supply meter errors, unauthorised or unrecorded consumption.
Many domestic supplies are not metered. In the UK about 14% of domestic supplies in England and Wales were metered in 1999, but none in Scotland and Northern Ireland. A DoE report mentions that in Denmark only town houses are normally metered; in the Netherlands about 24% of houses are unmetered; and in Germany only 30–40% of households are individually metered, the rest being block-metered. In the USA, whilst metering of domestic supplies is widespread, it is not universal. New York City did not start a programme to meter all domestic supplies until 1987. This was in response to the Delaware River Basin Commission's requirement that the major cities it supplies in the states of Delaware, New Jersey, New York and Pennsylvania, should achieve metering of all services within a ten year period. A survey by the Asian Development Bank (ADB) in 1996 showed that, of 27 Asian cities serving over 1 million people, only 15 were fully metered and six metered less than 7% of their connections (Calcutta 0%: Karachi 1%).
Trade and industrial supplies are usually metered because they are a major source of income to a water undertaking. In the UK many small shops and offices occupied only in the daytime used not to be metered, but now generally are, even though their consumption is small. Overseas standpipe supplies are not metered and are usually given free. The city of Bombay, for instance, supplies 400 M1/day (megalitres per day) to some 6 million people in its slums. In many countries large quantities of water are used for watering public parks and green areas and supplying government offices and military establishments, etc. They are often not metered nor paid for if the government (or state or city) supplies the water. The result is that reported water distribution losses depend on the accuracy with which the unmetered consumption is estimated.
1.2 Levels of total consumption
The usual measure of total consumption is the amount supplied per head of population; but in many cases the population served is not known accurately. In large cities there may be thousands of commuters coming in daily from outside; in holiday areas the population may double for part of the year. Other factors having a major influence on consumption figures are:
• whether the available supplies and pressure are sufficient to meet the demand, 24 hour or intermittent;
• the number of population using standpipes;
• the extent to which waterborne sanitation is available;
• the undertaking's efficiency in metering and billing, and in controlling leakage and wastage;
• how much of the supply goes to relatively few large industrial consumers;
• the climate.
Some countries, such as India, rarely have any cities with a 24-hour supply. The ADB survey of 1996 already referred to showed that 40% of 50 Asian cities surveyed did not have a 24-hour supply, and about two-thirds had street standpipe supplies. Hence comparison of average total consumption between undertakings is not informative. High consumption can be caused by large industrial demand and low consumption by a shortage of supplies. However the general range of total supplies per capita is:
• from 600 to 800 led (litres per capita per day) in the big industrial cities of USA;
• from 300 to 550 led for many major cities and urban areas throughout the world;
• from 90 to 150 led in areas where supplies are short or there are many street standpipes, or many of the population have private wells.
In England and Wales the average total supply was 288 lcd (1998/99). In Scotland it was 460 lcd and in Northern Ireland 407 led in 1997/98, mainly because these areas have a high rainfall providing plentiful supplies of good quality water.
1.3 Consumption surveys
A consumption survey is necessary when losses appear to be large, or consumers in some areas cannot get an adequate supply, or metering and billing practices appear to be inefficient. This situation often occurs on many undertakings throughout the world where lack of money and technical resources has resulted in water supply systems where leakage and consumer wastage is high, and records of consumption are unreliable. In such a situation, it should be noted that:
Total supply = Total legitimate potential demand plus consumer wastage and distribution losses minus unsatisfied demand.
Hence the total supply can seem to be adequate when expressed as the water available per head of population, but this may conceal the fact that, due to excessive wastage and leakage, there is much unsatisfied demand because some consumers do not get the water they need. It is then necessary to conduct a consumption survey to find the state of the system. The steps involved are the following.
(1) Log the initial state of the undertaking before any remedial work is started, by marking on a map of the distribution system areas where water pressure is too low for consumers to get what they need (a) at peak demand times, and (b) during the whole of the daytime.
(2) Check the accuracy of source meters, e.g. by diverting the source output over a temporary measuring weir, or by measuring the input to a tank.
(3) Check the general validity of supply meter readings by, for instance, check-reading a number of meters over a period and comparing with the readings billed.
Find the typical number of supply meters found stopped at any one time, and investigate what billings are made when meters are found stopped.
Find the average age of meters and how frequently they are brought in for testing and repair.
Test some meters of typical size and age for accuracy. Where more than 15% of meters are found stopped at any one time, or many meters are over 10 years old and not brought in for testing and repair, a substantial amount of under-recording must be suspected.
(4) Assess typical domestic consumption per capita by classifying dwellings into five or six classes, and test metering 30-35 dwellings typical of each class. (See Section 1.4 below.)
(5) Test meter a few typical standpipes and, by estimating the population reliant on each, estimate the typical standpipe consumption per capita.
(6) Examine the supply meters on all large trade and industrial supplies; check the accuracy of those found in poor condition.
List the largest potential trade consumers and check their billing records to see whether they seem reasonable having regard to the size of their supply pipe, their hours of take, and amount of water likely to be used for their production. It can be found that some major consumer is missed from the billings, or an establishment may have two supply feeds of which only one is metered.
(7) Meter, or by some other means estimate, the amount of water supplied unmetered to such as government or municipal offices, and also to public parks and gardens to get a measure of their probable consumption.
From the foregoing an estimate can be made of the probable total potential demand on the system in the following manner. On a map of the distribution supply districts, mark areas of the different classes of housing. Using appropriate population densities per hectare and measuring the areas of each class of housing within a district, estimate the total domestic demand per district by using the appropriate consumption per capita derived from Step 4. Add an allowance for unavoidable consumer wastage which will not have been registered by the test metering under Step 4.
In each district any standpipe consumption should be added, and the trade and other non-domestic demands apportioned according to the character of the district and the location of major trade consumers. An allowance for a reasonable degree of unavoidable distribution leakage should be added, usually expressed as a percentage addition to the total domestic and trade demand for each district. This gives the total average daily demand on the whole system, broken down into sufficiently small supply districts for the demand in each district to be distributed to 'nodal points' of the mains layout, i.e. key junctions of mains. These 'nodaldemands' can then form the basis for an hydraulic analysis of flows in the distribution system, as described in Section 14.14. From this can be ascertained the adequacy of the distribution system to meet the demands.
1.4 Test metering in-house domestic consumption
To assess average domestic consumption per person when supplies are not metered or records of metered domestic consumption are not reliable, it is necessary to test meter a sample of properties. It is best to rely on the results of test metering 30-35 households from each of five or six classes of households. Larger samples are difficult to conduct accurately because of the need to ensure all meters work accurately, all properties are leakfree, and the difficulty of keeping check of the number of people in each household. The households test metered should be typical of their class; they should not be chosen at random because the sample size is too small for random selection and could result in a bias of the sample mean towards the higher or lower end of the range within the class. Only about five or six classes of household should be adopted because it is difficult to distinguish between a larger range of households with any certainty. The test period should be 2–4 weeks, avoiding holiday times and, if possible, extremes of weather. Meter readings and occupancy rates should be ascertained weekly. Theoretically a sample size of at least 30 households is required to provide a reasonable estimate of the mean consumption in a given class of households. In practice 35 properties per class will need to be test metered to get a minimum of 30 valid results because of mishaps – a stopped meter, a leak discovered, or occupants gone away, etc.
Unfortunately all such tests show a wide scatter of results, as illustrated in Fig. 1.1. The mean of a sample can therefore be substantially influenced by a few households where the consumption seems extraordinarily high. Provided such high consumption is not due to meter reading error, these figures should not be excluded because domestic consumption is so highly variable. The problem is, however, that the sample size is too small to evaluate the incidence of such high consumers which may be, for example, one in 20, so that a sample size of 30 may contain no such high consumers, or one or two. However, if five classes of housing are adopted there will be at least five separate samples from which to judge, roughly, the frequency of such exceptional consumption.
(Continues...)
Excerpted from WATER SUPPLY by Alan C. Twort Don D. Ratnayaka Malcolm J. Brandt Copyright © 2000 by A. C. Twort, D. D. Ratnayaka and M. J. Brandt. Excerpted by permission of Butterworth-Heinemann. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.