In Denmark more than 98% of the drinking water originates from groundwater. The amount and quality of the groundwater resource is therefore a recurrent topic. A 2003 study from GEUS (Geological Survey of Denmark and Greenland) estimates the available amount of groundwater at only 1.0 billion m³ per year.

A survey carried out 11 years ago calculated this quantity at 1.8 billion m³. The difference between the two figures is mainly explained by the fact that the new assessment includes details of the freshwater cycle.

Effects on streams and nature limit the amount that can be extracted from the groundwater when the existing environmental objectives for streams are considered. The calculation from GEUS also shows that, with respect to the existing environmental goals, too much water is being pumped to larger cities like Copenhagen, Odense and Århus.

These facts explain why there is increasing focus on groundwater resources. Water consumption has previously decreased because of water-saving campaigns and increases in water prices and taxes and consumption is still decreasing. Minimizing the loss of water from the water supply networks is an element in this context. The main objective of this study has been to investigate how much water could possibly be saved between the waterworks and places of consumption.

The project objectives are:

• To quantify the amount of water lost in the water distribution system.
• To quantify the relationship between the loss of water in the public water supply system and the loss of water in the consumer-owned part of the system.

The focus is on the loss of water in the supply system and not on the possible savings that could be made. The water supply system is divided into public and private sections. The definition of the private section is that it is the water supply system within the boundary of a plot registered with the Land Registry . The rest of the water supply system is publicly owned.

In recent years, water consumption in Denmark has decreased considerably. However, in some parts of Denmark, especially around the larger cities, the resource is considerably overused. This leads to environmental and supply-safety issues. At the same time, in some areas large water loss occurs because of water leakage and cracks etc. in the water distribution network.

This project investigates the impact of water loss for the utilization of groundwater as a resource. Firstly, the quantitative value of the water loss is assessed and the possibilities for reducing it are reviewed. Secondly, the loss of water in the private section of the water distribution network and in the public section is estimated. Finally, consideration is given to methods of preventing the loss of water.

The following conclusions have been made based on the data collected and analysed:

• The true water loss is about 75% of the total registered loss of water.

From the waterworks to the consumer a part of the distributed water is lost. This loss is on average about 75% of the total loss. This loss comes from leakage and cracks in the water distribution network. Therefore focus on limiting leaks and cracks is important. The remaining registered loss is not actually a loss, but it results from activities such as water consumption through fire hydrants, flushing of water mains and sewers, unmeasured consumption at construction sites, water-measurement device inaccuracy and emptying of pipes in order to prevent frost damage.

• Water consumption and loss have decreased for many years.

The water consumption in Denmark has decreased year by year since 1986. Water loss has also decreased, but not as consistently as water consumption. Losses had decreased to about 6% of water distributed (25 million m³) in 2002. A tax on water was introduced in 1994 and, in 1999, the regulations made water meters compulsory in every property.

• The minimum possible loss has been reached in many places.

Many municipalities claim that the limit for minimum loss has been reached in their water distribution network, even without continued systematic renovation. Some water supplies have losses as low as 0.3%, and further reduction is not possible. A loss of about 3-4% is seen as a realistic and a good target for most water supplies.

• Some waterworks have large losses.

Some waterworks experience a permanent loss of about 30%.

• The loss is not dependent upon the size of the water supply.

It has little importance if the water supply is private or publicly owned.

• There is no geographic difference.

No significant geographical variation in water loss within the counties has been found.

• The age of the pipes is important.

The age of the pipe has a considerable influence. There is a clear relationship between increasing pipe age and specific water loss.

• Minimizing water loss can be optimized.

Registration of leakages in terms of location, causes, estimated loss etc. in a database is a helpful tool to minimize losses. Larger cities maintain such a database. Water supplies with high losses can use the experience from water supplies with a database of leakages. Surveillance of leakages e.g. by registering consumption at night or by monitoring the network at different places is important when the losses are to be reduced. Sudden bursts in the system are normally traced quickly. Most of the water loss comes from the large number of small leakages, especially leaks in connections and leakage due to corroded pipe material.

• The overuse of the water resource changes only little by reduction of water loss.

Because water loss is generally limited, reducing water loss in the water distribution network does not really help prevent overusing the groundwater resource. Actions that reduce the loss in the water supply network should therefore be compared with other methods to reduce groundwater extraction.

• The uncertainty in quantifying loss of water.

The uncertainty in quantifying water loss is high because reporting is based on a volunteer arrangement and it is uncoordinated. The waterworks can provide statistics one year but not the next year. The data that lie behind the official loss in Denmark are calculated from 50% of the publicly supplied water. The loss would probably be higher if all water suppliers had to report and if an external control existed.

• The loss in public compared to private sections is difficult to quantify.

It is not possible to distinguish quantitatively between water losses in public and private service connections. The boundary between public and private sections is in a place where no water meters are installed. Often the loss in the private section is included in the public loss because the water meter is situated at the buildings well within the private section.

• The service connections are in poor condition.

The general experience is that the service connections from mains to properties in private areas are in a poorer state than the public network because the responsibility of the municipality stops at the Land Registry boundary. However, the private section is in a less disturbed soil environment and this can increase its lifetime considerably compared to public sections situated in an environment submitted to traffic loads etc. The number of leaks in private and public sections of the service connection can be used to estimate where major loss occurs. An investigation from a leakage surveillance company suggests that of the total number of leaks, 61% come from the public section and the remaining 39% from the private section. Corresponding data from Rødovre municipality shows a distribution of 76% and 24%, respectively, i.e. again more leaks in public networks.

• More water is lost from the public section.

The loss in the public section of the distribution network is higher than the private section, also when expressed as absolute water amounts. Data from Århus shows that 75% of the estimated loss is from public distribution mains, while only 10% of the loss is from service connections that are either public or private.

In order to decrease the loss of water within distribution networks, the project points to the following recommendations:

• It is recommended that water supplies report outgoing water amounts from the waterworks in the same way as they do for indrawn water. The reporting should be exposed to external verification.

• Further investigations should be considered. This should include an evaluation of unmonitored water use, i.e. a detailed water balance for the water distributed from the waterworks. The most recent extensive survey is 25 years old. It is also recommended that the loss from the private section is investigated more thoroughly.

Because the losses are most often relatively small, it is recommended that the possible reduction in the use of the groundwater resources as a result of initiatives to decrease losses should be compared with other approaches to reduce the groundwater extraction.