Drinking water in a future climate

The supplier of drinking water in Norrköping, Sweden, needs decision support in order to adjust the treatment processes at the water plant in accordance with climatic related changes in raw water quality in the lake Glan. At present, only limited information is available on what kind of water quality changes can be expected in lake Glan in a future climate.

The aim of this case study has been to develop a method for simulating water quality changes in the lake using indata from the demonstrator.

There are large uncertainties in modelling water quality in the lake. Therefore further development is needed to fully meet the clients need for quantified climatic related changes in water quality in the hydrologic system.

Read Full Technical Report here!

The lake Glan, Sweden

Lake Glan in Sweden

Case Study Description
Data Description
Reference information

Water-management issue to be addressed

The quality of raw water used for producing drinking water is changing with the ongoing climate change. This study aims at describing the climatic changes in water quality parameters in the lake Glan, which affect the processes and the performance of the water treatment plant. The water quality parameters threatening good performance of the water treatment plant are water temperature, retention time, stratification, pH, turbidity, ammonium, nitrate, nitrite, phosphorus, organic matter, carbon, COD (chemical oxygen demand), Coli-bacteria, cyanobacteria, algal toxins and chlorophyll. The overall aim of this case study is to find a method to produce climate related information that will help Norrköping Vatten och Avfall to make decisions on adaptation of the plant in the long run of 20-50 years.

Decision support to client

The final results of the study will be used for helping Norrköping Vatten och Avfall to make decisions on measures needed in the water treatment plant to prepare and adapt to climate change.

Decision support in terms of quantification of changes in water temperature, organic matter, bacteria, turbidity and other water quality parameters may be used to quantify the amount of chemicals to be used, to assess the need to change treatment strategy or the need for new treatment process steps in the plant. The results may be used to calculate benefits and costs for investment strategies and present to the municipality politicians. The results may also be used as decision support on initiating campaigns of environmental monitoring needed. Finally, the more detailed results, the better support for prioritizing different measures needed.

"The results from the model have given us (Norrköping Vatten och Avfall) new knowledge that changed our discussions, pre-conditions and in some cases goals for our investigations about Norrköpings drinking water in the future ... results have made the decisions in a way harder, because they added new aspects to consider, but in the end they will give more reliable strategies for the future".

Temporal and spatial Scale

The timescale of adaptation measures in the water treatment plant varies depending on which parameter that are of interest.  This study will mainly focus on adaptations that are implemented over the timescale of 20-50 years. The spatial scale for decision making is mainly local to regional. Lake Glan and its surroundings is approximately 200-500 km2

Knowledge Brokering

Meetings have been held frequently with the client during the different project stages. Both face-to-face as well as telephone conferences. Regular updates have been sending to the client for their information as well as their continuous input to the case study.

A meeting was held 2017-02-14 including a presentation of the first results from modelling of E-coli for a reference period and for different climate scenarios. During the meeting interesting discussions were held regarding the element of uncertainty in the study, for example the different sources that contribute with E-coli to the lake. There was also a discussion about the next step in the study. The next face-to-face meeting was held 2017-04-25. During the meeting there were discussions about results from the modelling of important factors for the quality of the raw water, for example the stratification in the lake and the turnover time.

We found that it is important to hold frequent dialogues with the client regarding the usefulness of results and to understand the client's daily- and long term issues.

Climate Impact Indicators

Pan-European Indicators

The chosen indicators are listed below, with the belonging category stated within brackets.

  • Water temperature [water quality]
  • Soluble P conc. [water quality]
  • Particulate P conc. [water quality]
  • Precipitation (seasonality) [Precipitation]
  • River flow (seasonality) [Water quantity]
  • Cloud cover (seasonality) [Air]

The main objective is to set up different scenarios for describing present and future conditions in the local hydrodynamic model of the lake.  

Local indicators

Local water quality indicators will be produced by setting up a 3D hydrodynamic model for the lake Glan. The model will use Pan European indicators as input data. Water quality variables such as water temperature, nutrients, organic matter and Coli-bacteria will be calculated.

Pan-European data to local scale

Read Full Technical Report here!

drinking-water_annaStep 1: Collect Pan European meteorological, hydrological and/or water quality data/indicators for present and future conditions.

Step 2: Collect local water quality data.

Step 3: Set up a local 3D hydrodynamic model  for the lake Glan using Pan Eurpoean Climate Indicators as indata.

Step 4: Simulate and quantify local water quality changes in the lake in present and future climate.

Step 5: Validate the local model and evaluate the results of the simulations in relation to uncertainties in indata and in relation to local data and knowledge from the client on how much the processes in the water treatment plant might be affected.

Step 6: Propose actions to be done and measures needed in the water treatment plant to prepare and adapt to climate change.

Lessons learnt

Lessons learnt, so far, are that there are large uncertainties to handle when modelling water quality. One limiting factor for a successful result is the lack of indata and information on boundary conditions for the hydrodynamic model, when it comes to water quality indicators such as organic matter/carbon, turbidity, bacteria/algae, viruses, pH and more.

The demonstrator offers a good overview of what indicators are available and where. Resolution on catchment area has been another successful factor. However, we would like a better explanation and understanding of the metadata for the downloaded excel files from the demonstrator.

Importance and Relevance of Adaptation

Traditional methods for assessing variability in the processes of the water treatment plant are based on local measurements of pH, turbidity, pressure, temperature, chlorine and indicator bacteria. Measurements of water quality variables are made throughout the process chain: from raw water intake to the distribution system to the end user’s tap. This method is not enough for controlling the long term variability, but more of a here-and-now-method for effects and adaptation. For adapting to climate related changes in the long run, of 20-50 years or more, the traditional method used is to follow the research done in the field of drinking water generally and evaluate qualitatively how the general changes might affect local conditions. The need for more local studies is therefore critical when it comes to adaptation to climate related changes. 

Borg is a surface water treatment plant, which means that surface water (0-1,2m) is used in the production. Surface water is very sensitive to the changes and disturbances in the surroundings. The optimization of our processes has to follow the changes to maintain the production of drinking water. For some parameters in this study, the predicted influence due to climate change can tell us in which direction we should change our process to meet the future.

Pros and Cons or Cost-Benefit analysis of climate adaptation

If no climate adaptation measures were implemented in the treatment plant the supply of drinking water in the community would be threatened, with possibly serious effects on the society. The risk of outbreaks of water related infections would increase and the costs for alternative water distribution (water tanks) would increase.  As an example: In winter 2010-2011 the Swedish city Östersund were affected by an outbreak of the parasite Cryptosporidium that was spread by drinking water in the community. About 27000 inhabitants were infected by the parasite. An analysis of the consequences of the outbreak showed that the societal costs were SEK 141-221 million (“Parasitutbrottet i Östersund”, Ramböll & MSB, MSB794 - December 2014, ISBN 978-91-7383-523-7).

Policy aspects 

The most important result of this case study will be water quality changes on a local scale. The results of the study will serve as one, among several, decision support systems when it comes to policies and implementation strategies to meet climate related changes in raw water quality. The indicators will work as an early warning system: What do we have to do to maintain our drinking water quality tomorrow and in twenty years.

More about these policies can be found here.


Walter Gyllenram
E-mail: walter.gyllenram‹at›smhi.se
Tel dir: +46 031-751 8995

SMHI / Swedish Meteorological and Hydrological Institute
Sven Källfelts gata 15




Relevant EU Policy


Purveyor: SMHI / Swedish Meteorological and Hydrological Institute

Anna Edman
E-mail: Anna.Edman<a>smhi.se
Tel vx: +46 (0)11 495 80 00

SMHI / Swedish Meteorological and Hydrological Institute
Sven Källfelts gata 15

Value added by Copernicus Climate Change Service: 

Norrkoping vatten

Client: Norrköping Vatten och Avfall

Interior of Borgs Water Treatment Plant in Norrköping, Sweden

Interior of Borgs Water Treatment Plant in Norrköping, Sweden

Slow sand filters for water purification at Borgs Water Treatment Plant in Norrköping, Sweden
Slow sand filters for water purification at Borgs Water Treatment Plant in Norrköping, Sweden


Borgs Water Treatment Plant in Norrköping, Sweden

Borgs Water Treatment Plant in Norrköping, Sweden