PIARC 2002 XI International Winter Road Congress 28–31 January 2002 – Sapporo (Japan)

PIARC 2002 XI International Winter Road Congress 28–31 January 2002 – Sapporo (Japan)
VTI särtryck 350 • 2002
PIARC 2002
XIth International Winter Road Congress
28–31 January 2002 – Sapporo (Japan)
Reprints from Proceedings of Oral Presentations:
COST Action 344: Improvements to Snow and Ice Control on European Roads
and Bridges
Marilyn Burtwell and Gudrun Öberg
Winter Maintenance Standards in Cycleways
– Appropriate Road Condition for Increased Cycling during Winter
Anna Bergström
De-Icing Salt and Roadside Environment
– Strategies for Impact Analyses
Göran Blomqvist
Weather Descriptions and Compensation Model for Winter Road Maintenance
Staffan Möller and Carl-Henrik Ulegård
Predicting Steady State Concentrations of Chloride in Groundwater and
Surface Water
Eva-Lotta Thunqvist
Non-Exhaust Particles in the Road Environment – A Literature Review
Mats Gustafsson
Winter Index by Using RWIS and MESAN
Jan Ölander
VTI Särtryck 350 · 2002
PIARC 2002
XIth International Winter Road Congress
28–31 January 2002 – Sapporo (Japan)
Reprints from proceedings of Oral Presentations:
COST Action 344: Improvements to Snow and Ice Control on European Roads
and Bridges
Marilyn Burtwell and Gudrun Öberg
Winter Maintenance Standards in Cycleways
– Appropriate Road Condition for Increased Cycling during Winter
Anna Bergström
De-Icing Salt and Roadside Environment
– Strategies for Impact Analyses
Göran Blomqvist
Weather Descriptions and Compensation Model for Winter Road Maintenance
Staffan Möller and Carl-Henrik Ulegård
Predicting Steady State Concentrations of Chloride in Groundwater and
Surface W
ater
Water
Eva-Lotta Thunqvist
Non-Exhaust PPar
ar
ticles in the Road En
vironment – A Literature Re
vie
w
articles
Environment
Revie
view
Mats Gustafsson
Winter Index by Using RWIS and MESAN
Jan Ölander
Contents
Topic VI: Development of Snow-Removal and Ice-Control Technology
COST Action 344: Improvements to Snow and Ice Control on European
Roads and Bridges
Marilyn Burtwell, TRL, United Kingdom
Gudrun Öberg, VTI, Sweden
Topic III: Winter Road Issues and Traffic Safety in Urban Areas
Winter Maintenance Standards on Cycleways
– Appropriate Road Condition for Increased Cycling During Winter
Anna Bergström, VTI, Sweden
Topic IV: Environment and Energy
De-Icing Salt and Roadside Environment
– Strategies for Impact Analyses
Göran Blomqvist, VTI, Sweden
Topic I: Winter Road Policies and Strategies
Weather Descriptions and Compensation Model for Winter Road Maintenance
Staffan Möller, VTI, Sweden
Carl-Henrik Ulegård, SNRA, Sweden
Topic IV: Environment and Energy
Predicting Steady State Concentrations of Chloride in
Groundwater and Surface Water
Eva-Lotta Thunqvist, KTH, Sweden
Topic IV: Environment and Energy
Non-Exhaust Particles in the Road Environment – A Literature Review
Mats Gustafsson, VTI, Sweden
Topic II: Snow and Ice Management, and its Costs
Winter Index by Using RWIS and MESAN
Jan Ölander, SNRA, Sweden
COST ACTION 344: IMPROVEMENTS TO SNOW AND ICE CONTROL
ON EUROPEAN ROADS AND BRIDGES
Marilyn Burtwell* and Gudrun Öberg**
*TRL Limited
Old Wokingham Road
Crowthorne, Berks RG45 6AU,UK
TEL: +44-1344-770214/FAX: +44-1344-770748
E-mail address: mburtwell @trl.co.uk
**Swedish National Road and Transport
Research Institute (VTI)
58195 Linköping, Sweden
TEL: +46 1320 4153/FAX: +46 1314 1436
E-mail address: gudrun.oberg@vti.se
1. Abstract
Effective snow and ice control is a vital service provided by European highway authorities in order to
ensure, as far as possible, that road users can travel safely and with minimum disruption in cold and severe
climatic conditions. The need for innovative snow and ice control techniques and processes has continued
to grow as national and European road networks have developed substantially over recent decades. The
demand for improvement, including the sophistication of the techniques and technology used, continues to
be driven by the increasing need for safe and efficient national and international road freight and
passenger transport and by the environmental and other policies affecting highways.
European Commission project, COST Action 344: Improvements to snow and ice control on
European roads and bridges, started in April 1999, is a three-year project with participation from eighteen
European countries.
The project aims are:
1) Review of existing international practices
2) Definition of snow and ice control requirements in different European climatic regions.
3) Specification of ‘Best Practice’ in different European climatic regions.
4) Development of guidelines for the integration of specified snow and ice control methods into
network level road management and maintenance systems.
5) Recommendations for improvements to driver information and traffic management systems.
6) Recommendations for future research.
This COST Action will promote the exploitation of technological advances in the application and
distribution of snow and ice control measures, with a view to providing significant environmental and
safety benefits and lower operational costs. Millions of ECUs could be saved through lower operational
costs and a reduction in adverse effects on the highway infrastructure and the environment. For the road
users, more effective management of winter operations could lead to a reduction in traffic delays and
accidents. For practitioners, implementation of ‘Best Practice’ should enhance standards and lead to Best
Value being achieved. The implementation of Best Value could provide the means to measure the
performance of the winter maintenance service within various road administrations.
Interim results of the COST Action are being disseminated to European and national policymakers,
regional planners, engineers, road and vehicle operators, industry and academia. This approach
ensures maximum dissemination of knowledge. The Internet, a CD-ROM, Email, handbooks and
events such as workshops, conferences and seminars are being used to target a wider audience.
2. Introduction
Effective snow and ice control is vital to ensure, as far as possible that road users can travel safely and
with minimum disruption in cold and severe weather conditions. However, it is important that the winter
maintenance service is provided at an affordable price and that ‘Best Value’ is achieved with minimum
environmental impact and traffic disruption, and with high standards of safety. Information on ‘Best
Practice’ is therefore essential to ensure widespread implementation of appropriate standards of service.
The need for innovative snow and ice control techniques and processes has grown over recent
decades in line with the development of national and European road networks. The demand for improvement,
1
including the sophistication of the techniques and technology used, continues to be driven by the
increasing need for safe and efficient road freight and passenger transport, and by the environmental and
other policies affecting highways.
The COST Action 344: Improvements to snow and ice control on European roads and bridges, started
in April 1999 and is part funded by the European COST (Co-operation in the field of Scientific and
Technical Research) programme (EU, 1999). The Action (www.cordis.lu/cost-transport/home.html) is a
three-year project with participation from eighteen European countries. TRL is the Chair of the COST
Action and represents the UK Highways Agency, which is responsible for the operation and maintenance
of the Trunk Roads and motorways in England. VTI is the Vice Chair and represents the Swedish
National Roads Administration, which is responsible for the operation and maintenance of the Swedish
national road network. These organisations are members of the COST 344 Management Committee.
3. Objectives of the research
The main aim of the COST project is to improve the performance of snow and ice control methods and
operations by defining the requirements for ‘Best Practice’ in different climate domains, across the EU
and other COST member states. This will provide national highway authorities with information on the
best materials, techniques and procedures to meet the changing demands of the European road
infrastructure and, at the same time, harmonise safety and environmental standards. It will thus provide
guidance to decision makers.
A significant contribution will be provided to meet the stated goals of the Transport European Road
Network (TERN) as below:
•
Sustainable mobility of persons and goods within the EU under the best possible social and safety
conditions (Article 2.2a).
• Integration of environmental concerns into the design and development of the network (Article
5d).
• Promotion of network interconnection and inter-operability between the EU and the third world
countries (Article 6).
Assessments of operational practices, employed at national level, are also expected to result in the
development of objective criteria and benchmarks for various aspects of snow and ice control and their
impact.
4. Work programme
The aims of the research project are:
a) To review existing international practices, involving the following elements:
• terminology review and creation of a European glossary;
• literature review covering the years 1990 to 2000 to establish the state-of-the-art
practice and research in snow and ice control methodologies;
• review of current research and development work, in both the public and private
sectors;
• review of current practices by evaluating selected case studies in targeted EU regions;
and
• creation of an inventory of snow and ice control methods, equipment and materials.
b) to define snow and ice control requirements in different European regions;
c) to determine ‘Best Practice’ in different European regions;
d) to develop guidelines for the integration of specified snow and ice control methods into
network level road management and maintenance systems;
e) to make recommendations for improvements to driver information systems and traffic
management systems; and
2
f) to make recommendations for future winter maintenance research, which has potential benefits
for practitioners and road users.
Particular areas where further investigation has been proposed are:
• the most effective and least environmentally harmful de-icing/anti-icing materials, and the most
effective treatments in the various climates encountered across COST member states;
• implications resulting from the introduction of innovative road surfacings to establish
benchmarks for safe and effective winter maintenance;
• innovative Road Weather Information Systems (RWISs), which would benefit from a review
of accuracy, reliability and the introduction of developing capabilities such as residual salt
sensors; and
• road icing information and prognosis systems.
Investigations are also underway on the following:
• Operational procedures:
• driver information systems using existing methods and innovative developments
employing advanced telematics; and
• the impact of methods designed to maximise traffic flows and reduce accident severity
in winter conditions.
Information on many of these research elements has been drawn from the experience and knowledge of
participating member states through detailed assessments and a review of current and ongoing research.
The common interests and general objectives are shared by the member states and the planned work is
drawing upon most of the relevant work currently in progress and planned within all COST countries
together with the results of work undertaken previously.
5. Task Groups
Six Task Groups, TG1 to TG6 with nominated leaders, will run through the three-year life of the
Action. The seventh group, TG7 will start in year 3 of the project. These Groups involve the most
appropriate blend of technical expertise for the tasks from a broad geographical distribution across Europe
to ensure an extensive input and high quality outputs. The Groups are:
TG1 – Information gathering, literature review and glossary
TG2 – Definition of requirements
TG3 – ‘Best Practice’
TG4 – Future research
TG5 – Road management system
TG6 – Driver information systems
TG7 – Final report
Each Group has submitted at least one technical deliverable and, these will form a major part of the
final report of the Action.
5.1 Task Group 1 - information gathering, literature and glossary
A glossary of winter maintenance terms in six languages – Dutch, English, French, German, Swedish
and Spanish has been produced. It is expected that PIARC will adopt the COST glossary, in 2002 at the
end of the Action, to complement its own glossary. A European review of literature from 1990 to 2000,
which includes over 600 research papers and reports, has been divided into topics (weather and climate,
equipment, effects, management, de-icing products, equipment for road users, risk management, strategy,
design and construction of the road, costs of winter maintenance, road user information and overview).
The work has also identified about 150 current research projects throughout Europe on winter
maintenance practice and management issues.
3
The review of literature and current projects has identified the gaps in our knowledge and thus where
future research efforts should be directed.
5.2Task Group 2 – definition of requirements
The objectives of TG2 were to consider safety, environmental and information criteria, the
management and operations of snow and ice control and, to identify improvements that would enable
delivery of a more cost-effective and efficient service. To achieve this it is important to set down the
components of a winter maintenance management system which, on balance, will produce a quality
service. The work of TG2 complements the work carried out in TG3 - ‘Best Practice’.
TG2 members have identified the following generic business areas as being of fundamental
importance to road administrations:
a)
Service levels – Relate to the winter maintenance operation itself and includes the
effectiveness of the treatment in preventing ice and snow adversely affecting the highway. It
does not however include safety and traffic movement considerations, which it is argued, are
secondary effects and can be influenced by factors other than the quality of the winter
maintenance operations.
b)
Environment – Includes the effect of winter maintenance operations on the natural
environment, including flora, fauna and marine life.
c)
Safety – Includes the safety of the winter maintenance operatives and the road users. Care
must be exercised to ensure that the reasons for safety performance are understood since
factors other than the quality of winter maintenance may be relevant.
d)
Traffic movement – Includes traffic flow during winter conditions, which may again be
affected by factors other than effectiveness of the winter maintenance operation.
e)
Cost optimisation – Includes analysis of all the factors that contribute to the delivery of a
cost-effective winter maintenance service.
f)
Information to the administration – Includes the provision and management of information
about the performance of the operation so that proper accountability can be achieved.
g)
Information to the road users – Includes the appropriate level of information to road users
in various forms both before and during the journey made.
These generic issues are set out graphically in Figure 1. They have been disaggregated to a) identify
more detailed issues requiring analysis and b) deliver the appropriate quality of winter maintenance
service. Items (a), (b), (e), (f) and (g) above are those issues over which the administration has a
significant level of control whereas items (c) and (d) are random occurrences influenced by other factors
including driver behaviour.
The type of climate is also a prime factor - this depends on the altitude and geographical location, and is
manifest through the frequency, duration and intensity of the winter weather conditions (COST 309,
1992). Conventional classifications can be made ranging from mild to very cold climates. A winter index
is a given function of the number of days with icy conditions with the minimum and mean temperature.
This determines the frequency and duration of ice on the roads. A Road Weather Information System
(RWIS) determines the adverse winter conditions in order to make the necessary decisions with sufficient
time in hand. Winter weather conditions include snowfalls, ice, freezing rain, fog, snowdrifts, avalanches etc.
Their frequency, duration and intensity depend on the meteorology of each area. The onset of winter
weather triggers the resources needed to re-establish the serviceability of the road.
Important characteristics of the road are the road type (high capacity or conventional), carriageway
4
width, layout, gradient, pavement type, frequency and length of bridges and tunnels etc.
Environment
Road safety
Information
to road users
Service level
quality
Traffic flow
conditions
Information
for road
administrations
Cost optimisation
Figure 1. Schematic diagram of the links in the winter maintenance processes
Key:
Issues over which the administration has substantial control.
Issues over which the administration has significant control.
Issues over which the administration has limited control.
5.3 Task Group 3 – ‘Best Practice’
The objectives of TG3 are to identify ‘Best Practice’ in the field of winter maintenance, including
the impact of operations on the environment and benefits to service providers and road users. The
identification of ‘Best Practice’ will encompass all the needs of the European Community specific to
particular countries and/or climates involved in winter maintenance activities. A questionnaire, in the
form of a detailed subject list, was prepared and distributed to EU member states to determine current
winter maintenance practices. The responses have been compiled and compared for common climate
domains (Scandinavian, Maritime, Central European, Continental, Mediterranean and Alpine). The
5
climate domains differ especially in temperature (daily and yearly), humidity, probability of snow, wind
and expectations of the user. A wide range of practice, environmental issues and benefits have
therefore been compared and evaluated.
When preparing a winter maintenance procedural statement, it is necessary to consider climate and
weather information, methods, resources (eg manpower, equipment and materials) that will need to be
employed. This will include information about chemical de-icers, gritting materials, mechanical snow
and ice removal equipment, and special treatments applicable to certain types of road surfacing
materials, bridges, cycletracks and pedestrian footways. It will also include developments in RWISs,
specifically the measurement of residual chemical on the road surface. The efficiency of the chosen
procedures can be measured using internal performance audit methods. An external audit could
measure the number and severity of accidents, travel time delay, user satisfaction and environmental
impact.
It is also important to have in mind the owner of the road, contract manager, operational staff and
road users before decisions about winter maintenance procedures are taken. Fundamental issues,
which influence winter maintenance, are climatic conditions, standards and legal obligations.
Consideration of the points covered above will enable improvements in ‘Best Practice’ to be made
throughout Europe.
5.4 Task Group 4 – future research
At present, various institutions are carrying out work into improvements in winter maintenance
management, procedures, techniques, treatments, weather and climate, safety and other effects. Whilst
valuable, these are largely uncoordinated initiatives and the COST Action has brought these together to
identify ‘Best Practice’.
The objective of TG4 was to identify the most important topics for future research activities in the
domain of COST 344.
The work of the task group was carried out in three phases:
identification of topics for future research;
•
prioritisation of future research topics; and
•
selection and task description of the most important topics for future research.
•
The topics for future research were collected via an e-mail survey sent to the COST 344
Management Committee and other international experts. About 90 respondents sent proposals for
research topics. TG4 members analysed the list of about 200 different topics received and produced,
by merging, a list of 93 research topics for prioritisation.
This topic list was used as a basis for an Internet survey, where experts from different countries and
representing different organisations (authority, industry, research or academia) were asked to
prioritise the research topics. In all, 57 experts completed the survey.
A number of topics were regarded as very important or important and TG members produced
tentative research task descriptions for these topics. The six most important future research topics are:
1. Forecasting, measuring and modelling the road surface condition.
2. Winter maintenance and management policies and strategies (service performance,
harmonised quality levels etc).
3. Costs and benefits of operational practice in rural and urban areas.
4. Effects of road weather conditions and winter maintenance on traffic flow and safety,
capacity and road user behaviour.
5. More cost-effective, efficient and environmentally friendly de-icing products.
6. Weather-related traffic management and information systems optimal for traffic safety and
efficiency.
5.5 Task Group 5 – road management system
A Winter Maintenance Management System (WMMS) is an important integral part of an integrated
6
Road Management System (RMS) and financial, quality, legal and social aspects need to be
considered.
There are two levels of a WMMS that should be considered - the strategic level where the socioeconomic consequences of a chosen winter maintenance strategy are calculated, and the day-to-day
level used for the management of the winter maintenance activities.
On a strategic level, it is not the objective to define the level of service but to define which
parameters have to be considered when defining the level of service. In practice, it is an optimisation
process between costs and benefits, as far as is practicable, because of the limited funds available. The
efficiency and effectiveness of the service provision and the chosen optimisation process, which must
be continually reviewed, determines delivery. New research ideas need to be fed into this
optimisation process to continually improve it and the subsequent service.
A WMMS on the day-to-day level may consist of several parts/systems such as:
•
•
•
•
•
•
•
administrative information;
route planning;
Road Weather Information System (RWIS);
call-out system;
reporting and documentation of actions;
information to road users; and
follow-up of actions.
Some European countries have a WMMS that includes many of the above parts but many
countries have one or more of the parts as separate systems, eg Road Weather Information System
(RWIS). A RWIS includes outstations, which measure parameters close to the road, eg road surface
temperature, and common meteorological information, eg wind speed, humidity etc.
TG5 members are considering the components and inputs and outputs required for a WMMS
and its compatibility with other modules or systems in a RMS. Comments on the benefits of
introducing a WMMS into a RMS will be included in the final report from the Action.
5.6 Task Group 6 –driver information systems
TG6 members are considering the effectiveness and benefits of driver information and traffic
management systems for road users in adverse weather conditions. Information for drivers is essential
if they are to travel safely on the road network in winter but the nature of the information given needs
to be timely and accurate. Ways of disseminating the information could include telematics (in-driver
vehicle systems), the Internet, radio, telephone, journals, teletext and variable message signs alongside
the road.
It is recognised that road users comprise different driver groups, which have different needs for pretrip and on-trip information. The driver groups have been identified as:
• Professional drivers (eg public transport, haulage, security services)
• Frequent drivers (eg commuters)
• Occasional drivers (eg school errands, tourists)
• Related businesses (eg travel agencies, private information services).
It is important to identify what sort of information each driver group requires. For example,
Finland has carried out a study of the frequent and occasional drivers, and this is being examined in
detail for the purposes of the Action. This work may be considered as a good example of ‘Best
Practice’ and much can be learned from it.
A questionnaire has been compiled by TG 6 members and circulated to all the European members
of the Action to seek answers to a series of questions regarding driver information and related
information systems. The questions include:
• What actions are used now?
7
•
•
•
•
•
What are the effects of these actions?
What are the costs and benefits of driver information systems?
What do road users need?
What could be done better?
What could be provided but is not?
The usefulness of information needs to be considered to avoid information ‘overload’ and the
timing of this information is also important. Three stages of the information process are essential – at
the onset of winter weather, during winter events, and in the case of a crisis. This will ensure that the
drivers have timely information and can plan their journeys in advance or during their travel on the
road network. When faced with exceptional circumstances such as heavy snowstorms and traffic
difficulties, collaboration with the police and other bodies is essential.
Private radio systems utilise the information services of the roads administration in Iceland and
Finland. For example, TRAVEL-GUIDE is a current project undertaken in Finland and is concerned
with traffic management and information services. The approach is to specify a commonly agreed
data exchange interface, via which private service providers have access to public organisation
information and vice versa. The Viking Travel and Traffic Information Service (www.ten-t.com/viking) and
its guidelines propose quality requirements for road weather and road surface condition information.
Systems such as these described above are being investigated further in the COST Action.
5.7 Task Group 7 – final report
The final report will include summaries of the Task Group reports, benefits of the project to different
user groups, a discussion, and conclusions together with overall recommendations.
6. Dissemination of information from the Action
A dissemination plan has been produced to promote the results of the Action to European and national
policymakers, regional planners, engineers, road and vehicle operators, industry and academia. This
approach will ensure maximum dissemination of knowledge. Results of the Action are to be disseminated
to a wider audience by means of events such as workshops, conferences and seminars in the participating
EU countries and member states and by e-mails and the Internet. At the end of the Action, the final report,
a CD-ROM and a series of handbooks will be made available to interested winter maintenance personnel
in the participating EU Countries and member states.
7. Summary
The COST Action will:
• Identify ‘Best Practice’ and emerging developments within and between EU and other COST
member states.
• Investigate necessary improvements to RWISs to introduce any latest available features such
as residual salt sensors.
• Ensure that treatments are carried out to reduce any harmful effects in the environment.
• Assess the impact of methods designed to maximise traffic flows and reduce accident severity
in winter conditions.
• Generate recommendations for the integration of specified snow and ice control methods
into network level road management and maintenance systems.
• Develop recommendations for further improvement in the dissemination of up to date and
reliable information to practitioners and road users.
• Generate recommendations for improving the level and quality of user input information in
snow and ice control decision making.
• Identify future research.
8. Benefits
The Action has promoted exploitation of technological advances in application and distribution of snow
8
and ice control measures leading to significant environmental benefits. With the application of the
knowledge gained, millions of ECUs could be saved through lower operational costs and a reduction in
adverse effects on highway infrastructure and the environment.
For the road users and communities, more effective management of winter operations will lead to a
reduction in traffic delays and accidents.
9. Acknowledgements
The authors wish to thank the members of COST Action 344 Management Committee for their
contributions to the project.
The work described in this paper forms part of the UK Highways Agency’s research programme carried
out by TRL and is published by permission of the Chief Executives of the UK Highways Agency and TRL.
The work described in this paper forms part of the Swedish National Roads Administration research
programme carried out by VTI and is published by permission of the Swedish National Roads
Administration and VTI.
10. References
COST 309 (1992). Road Weather Conditions, 1992 - 146 pp - ISBN 92-826-3244-X - EUR 13847.
European Commission (1999). COST Action 344: Improvements to snow and ice control on European
Roads and Bridges - Memorandum of Understanding.
Copyright TRL Limited 2001. This paper has been produced by TRL Limited, under a contract
placed by the UK Highways Agency. Any views expressed in it are not necessarily those of the UK
Highways Agency.
9
WINTER MAINTENANCE STANDARDS ON CYCLEWAYS
- Appropriate Road Condition for Increased Cycling During Winter
Anna Bergström
Infrastructure Maintenance, Swedish National Road and Transport Research Institute
S-581 95 Linköping, Sweden
TEL.: +46 13 20 40 48 / FAX: +46 13 20 41 45
E-mail: anna.bergstrom@vti.se
1. Abstract
From an environmental perspective, a reduction in motor traffic would be desirable. In urban
regions, this could be achieved by increasing cycling as a means of personal travel. Improved winter
maintenance of cycleways could lead to more winter cyclists. In this paper, the results of a Swedish
PhD project, with the purpose of studying the effects of winter road maintenance on cycling, will be
summarised. Included in the paper are the results from a literature review focusing on winter
maintenance of cycleways, an introductory questionnaire survey to improve the knowledge about
travel behaviour during winter, and a field study to see whether it was possible to attain an improved
service level. Focus is set on the field study where “new” equipment for snow clearance and de-icing
of cycleways was tested.
Results are presented indicating that there seems to be a prevailing discontent among the public
concerning winter maintenance of cycleways, and that better winter maintenance could lead to
increased cycling. Although slippery surfaces are of great importance for the safety of cyclists,
cycleways not cleared from snow seem to be more important for the mode choice.
The results presented from the field study will show that a test method using a broom for snow
clearance and brine for de-icing, provided a higher maintenance service level than methods normally
used on cycleways.
2. Introduction
Car-based transport has a wide range of impacts upon society and the broader environment. Air
pollution, congestion, noise, road accidents, and extensive land use for parking facilities and road
constructions, are some of the effects generating large costs for the society. A reduction in motor
traffic especially in urban regions would therefore be desirable. This could be achieved by increasing
cycling as a means of personal travel, leading to a more economical use of resources such as materials
and energy. Regular cycling also contributes to keeping people fit and healthy. From an environmental
perspective it is especially important to reduce the number of short car trips since they are responsible
for a relatively large amount of the emissions caused by traffic. This is particularly true in winter due
to all the cold starts.
Half of all the car trips made in Sweden are shorter than 5 km (Riks-RVU, 1998), and since most
people consider there is no difficulty in cycling distances less than 5 km (Herrstedt et al., 1995;
Nilsson, 1995), there is a potential for an increased cycling. However, in Sweden, the cycling
frequency during the winter is only about a third of that during the summer (Öberg et al., 1996). This
decrease during winter is probably largely due to the less favourable weather conditions; low
temperatures, strong winds, and precipitation all have a negative influence on cycling (Emmerson,
Ryley and Davies, 1998). But, road conditions are also of importance. A cycleway with poor snow
clearance means limited accessibility, and a slippery cycleway increases the risk of fall accidents,
which deters many from cycling during winter.
In seeking to promote cycling in winter, it is important to know the significance of maintenance
service levels of cycleways for travel behaviour. Even though bad road conditions affect cycling
negatively, it is not certain that improved winter maintenance standards could lead to more winter
cyclists. If it could, it would be desirable to identify the potential for winter cycling. There is also a
1
need to identify current winter maintenance service levels of cycleways, and the possibilities for
making improvements at a reasonable cost.
Considering this, a PhD project with the objective of studying the effects of winter road
maintenance on cycling was initiated in 1997 by the Centre for Research and Education in Operation
and Maintenance of Infrastructure (CDU). The project is presently being conducted at the Swedish
National Road and Transport Research Institute (VTI), and supported financially by the Swedish
National Road Administration The project will result in a doctoral dissertation, which is planned for
December 2001.
The objective of this paper is to summarise the results of the PhD project, including a literature
review, an introductory questionnaire survey, and field studies. The main focus is set on the field study
where “new” equipment for snow clearance and de-icing of cycleways was tested. The field study
included a pilot study, and a two-year large-scale study, and was evaluated through road condition
observations, measurements of friction, traffic censuses, and a questionnaire survey. In particular the
results from that questionnaire survey, which aimed to get the users’ opinion of the method tested, will
be presented in this paper.
3. Literature Review
Throughout the PhD project relevant literature has been reviewed. The literature review focused
on winter maintenance of cycleways, such as methods for snow clearance and skid control,
requirements of road operation service levels during winter, and methods of monitoring road condition
and evaluating the level of service. Other factors associated with winter cycling were also of interest,
in particular those related to the mode choice, but even more general topics, such as accidents
involving cyclists, were included. Reports representing results not relevant to Swedish conditions were
excluded, implying that most of the literature studied was Swedish. Unfortunately, there was not much
to be found concerning winter maintenance of cycleways. Most studies in relation to cyclists and
cycleways involve accident studies or travel surveys from a summer conditions perspective.
Nevertheless, in the literature it was found that the methods and equipment used for cycleway
maintenance are usually the same as for roads and streets (NVF, 1984). Therefore, in many cases, the
equipment is too large and heavy for this purpose, and can cause damage to cycleways; it is also
difficult for it to pass through tunnels and narrow passages. Its usability is also reduced to a certain
degree by low speed. In recent years, however, a new generation of vehicles, for example the Multicar
and the Mercedes Benz UX 100 (Figure 1) have become available on the market. These vehicles are
light, manoeuvrable, and fast (although engendering high safety), and can be easily equipped for a
variety of applications (NVF, 1999). The possibility of changing the application of the vehicle by
alternating the equipment makes for good economy, since it enables the same vehicle to be used for
both winter and summer maintenance operations. Consequently, these smaller vehicles are becoming
more and more popular for municipal use, although they are not yet common in all Swedish
municipalities. The new vehicles are rather expensive to purchase, and functioning old equipment is
not exchanged simply because it is old fashioned. The most common vehicles used for snow clearance
on cycleways today are several kinds of tractors such as the Volvo BM 650 or bucket chargers such as
the Lundberg 341 (Lindmark and Lundborg, 1987; NVF, 1984).
There seems to be no specific methods of monitoring road conditions on cycleways. The
methods available were developed for roads and street (Gabestad, 1988; Möller and Öberg, 1990), and
although some of them can be used for cycleways, they are not well adapted for it. During winter the
road condition changes continuously with the weather, as well as being influenced by traffic.
Therefore, a visual inspection is almost the only suitable method of monitoring road conditions during
winter, although the assessment is subjective and entails considerable manual efforts. Measurement of
friction is one of few objective methods of evaluating the level of service on roads during winter.
However, the friction measurement devices available, like the methods for maintenance, are usually
too large and heavy to be suitable for the use on cycleways.
In the literature review it was also found that there seems to be a prevailing discontent among
the public concerning winter maintenance of cycleways. In a survey performed among citizens in 12
Swedish municipalities (SALA, 1998), only 29% of the respondents thought that snow clearance and
2
skid control of facilities for cyclists and pedestrians were “very good“ or “rather good”, while 68%
were satisfied with winter maintenance of motor traffic roads in central areas. This indicates that there
is a need to improve winter maintenance on cycleways. However, it is unclear if the dissatisfaction is
due to insufficient service level requirements, or if the requirements in reality are poorly met.
According to Möller, Wallman and Gregersen (1991) the accident risk for cyclists increases 5 to
10 times during icy and snowy road conditions compared to bare surfaces. Single accidents in
particular are more prevalent during winter. Besides ice and snow, grit from winter maintenance also
constitutes a safety hazard for cyclists. According to Binderup Larsen et al. (1991), 10% of all single
accidents are caused by loose grit on the road surface. Although slippery surfaces are of great
importance for the safety of cyclists, cycleways not cleared from snow seem to be more important in
the choice to cycle or not during winter (Giæver, Øvstedal and Lindland, 1998).
4. Questionnaire Survey
To improve the knowledge about travel behaviour during winter, a questionnaire survey was
conducted in the PhD project, in 1998 (Bergström 1999, and 2000). The survey focused on journeys to
work, and questionnaires were answered by a total of 499 employees at three large companies in two
Swedish cities, Luleå and Linköping. The survey aimed to clarify the importance of winter
maintenance service level of cycleways for the choice of mode, and to get the respondents opinion
concerning the current service level of cycleways.
In the survey it was found that the total number of bicycle trips to work decreased by 47% from
the summer period, April to October, to the winter period, November to March. During summer 36%,
and during winter 19%, of all the trips to work were bicycle trips. At the same time the number of car
trips increased by 27% from 53% during the summer period to 68% during the winter period. In total,
38% of the respondents stated that they would cycle more during winter if the maintenance service
level of cycleways was improved. A majority of the respondents, 57%, thought that winter
maintenance on cycleways needed to be improved, 9% thought that it was satisfactory, and 30% were
uncertain or had no opinion. The survey also concluded, in accordance with the literature review, that
snow clearance is more important than skid control for the choice of mode.
Another conclusion from the questionnaire survey worth mentioning is that distance seems to be
more important for the mode choice during winter than in summer. In summer, one can hope to
transfer some of the car trips up to 5 km to bicycle, while it seems that the critical distance is
shortened to about 3 km during winter.
5. Test of Unconventional Methods for Winter Maintenance on Cycleways
Both the literature review (SALA, 1998) and the questionnaire survey (Bergström, 1999;
Bergström, 2000) indicated that the public is unsatisfied with the service levels provided on cycleways
during winter, and that improved winter maintenance on cycleways could lead to increased cycling.
However, it is uncertain whether it is possible to improve the service level of cycleways at a
reasonable cost, what maintenance methods are to be used, and how much they are able to affect the
choice of mode during winter. Further studies are therefore needed, and in the PhD project it was
decided to conduct field studies to test unconventional methods of snow clearance and skid control of
cycleways. The methods tested were compared to traditional maintenance methods with respect to
service levels achieved, such as the degree of snow clearance and the surface friction. Interviews and
questionnaire surveys were also done, to see if the road users noticed any difference in the level of
service achieved with the equipment tested. To see if an improved standard would lead to an increase
in cycling, bicycle censuses related to different road conditions were also conducted.
5.1 Method
To find a winter maintenance method that could improve the service level of cycleways, and to
get experience, with a view to a large-scale study, of the problems resulting from certain maintenance
methods, a pilot study was carried out in Linköping (Sweden) in 1999. Traditionally, in Linköping
cycleways are cleared through ploughing and skid control is attained by abrasives in 4 to 8 mm size. In
the pilot study, two different and unconventional methods of snow clearance and skid control were
3
tested on two selected cycleways. One of the methods used a traditional steel plough for snow
clearance and graded gravel for skid control. The graded gravel consisted of natural granular stone
particles washed and processed to obtain a size of between 2 and 5 mm. This test method was similar
to the method normally used on cycleways in Linköping, but was still meant to produce a higher
service level by having a tougher starting condition and by using the graded gravel with the purpose of
reducing cyclists’ problems with punctured tyres. The other test method used a front-mounted broom
for snow clearance combined with a brine spreader for de-icing. Using the snow broom was meant to
reduce any remaining layer of ice and snow so that the salt dosage needed to achieve a bare surface
could be minimised. The idea of this “brine method” originated from Odense in Denmark (Mikkelsen
and Prahl, 1998), where a similar method had been used for winter maintenance on cycleways for
several years. However, it was uncertain if this method was applicable to the Swedish winter climate.
The results achieved in the pilot study were limited and uncertain, since the test was performed
for only a little more than a month. However, it was concluded that the method of using a broom for
snow clearance and brine for de-icing produced a higher level of service compared to a traditional
method, and was therefore considered of sufficient interest for further research in a large-scale study.
The method using the graded gravel did not notably improve the service level, and although graded
gravel might reduce cyclists’ problems with punctured tyres, it may also increase the problem with
poor friction on bare surfaces. Therefore it was decided not to go on with that method.
The large-scale study was carried out during two winters, between October 1999 and March
2001. In this study a housing area, Ekholmen, within cycling distance of a large workplace, Saab AB,
in Linköping, Sweden, was used as a test area. In addition to all the cycleways within Ekholmen, three
major routes from Ekholmen to Saab AB were included in the test area, resulting in a total of about
23 km of cycleway. In the test area the cycleways were given a higher level of service than usual in
Linköping by using the front-mounted broom for snow clearance and brine, or on some difficult
occasions pre-wetted salt, for de-icing. The equipment used was almost the same as that used in the
pilot study, but instead of a Multicar used in the pilot study a new vehicle, a Mercedes Benz UX 100,
was purchased for the large-scale study (Figure 1). Another modification before the large-scale study,
was the use of a spinner, instead of a spraying boom, for brine spreading. As in the pilot study, snow
clearance and skid control were performed more frequently than on other cycleways, starting snow
clearance at a snow depth of 1 cm loose snow and de-icing on every occasion ice, snow, or hoarfrost
occurred. In Linköping snow clearance is normally started at a depth of 3 cm.
In the large-scale study, as well as during the pilot study, observations of the road surface
conditions were conducted after each occurrence of snowfall or hoarfrost. For these observations, a
method for roadways (Möller and Öberg, 1990) modified to better describe the prevailing conditions
on cycleways (Bergström, 2000) was used. Observations were done on both cycleways included in the
test and maintained with the “brine method”, and on cycleways used as controls and maintained
traditionally. As a complement to the observations, measurements of friction were conducted on a few
occasions. These measurements were performed with a Portable Friction Tester (PFT), developed at
the Swedish National Road and Transport Research Institute (VTI) to measure friction on road
markings in wet conditions (Lundkvist and Lindén, 1994). Since the PFT is reasonably small and
handy, it was considered practicable in this case when measuring friction on cycleway surfaces where
it can be difficult to use other measuring devices (Bergström, 2001).
To get the users’ opinions of the test method, interviews were carried out on a few occasions,
especially in the pilot study. The large-scale study was also evaluated through a questionnaire survey,
performed in 2000. A total of 570 questionnaires were answered by employees at Saab AB living in
the test area of Ekholmen, and by reference groups. The large-scale study was also evaluated by
counting cyclists, particularly during the second winter of 2000/2001.
4
Figure 1. The Mercedes Benz UX 100 Used in the Large-scale Study, Equipped with a Front-mounted
Broom for Snow Clearance and a Spinner for Spreading Brine, or Pre-wetted Salt.
5.2 Results
In the pilot study, and in the first winter of the large-scale study, the weather conditions were
not ideal for the purpose of testing new winter maintenance methods since it was fairly mild, with high
average temperatures and less snow than normal. During the second winter of the large-scale study,
there were periods of high snow intensity, but overall one could say that this winter was also milder
than normal. Unfortunately, this means that the results cannot apply to the typical winter conditions in
this region.
The large-scale study has not yet been fully evaluated. Bicycle censuses related to different road
conditions are not yet analysed, and a financial evaluation of the test method still remains to be made.
Nevertheless, both in the pilot study and in the large-scale study, the observations of road surface
conditions showed that there was almost always a dry, moist, or wet bare surface on cycleways in the
test area, no matter what the conditions were on other cycleways in the municipality. This implies that
the test method using a broom for snow clearance and brine for de-icing provides a higher
maintenance service level than the methods traditionally used in Linköping. At the end of each study
period, the effect of the midday thaw in combination with the “brine method” showed it to be very
efficient for clearing the cycleways. If brine had been spread in the morning during a day of sunshine,
the road condition on the cycleways in the afternoon was almost always dry bare surface.
During the pilot study, and the first winter of the large-scale study, on occasions with a snow
depth over 2–3 cm of loose snow, and if the snow was very wet, the broom had problems clearing the
snow. The effect of the broom was therefore improved by adding an extra hydraulic engine before the
second winter of the large-scale study. This improved the snow clearing results considerably and at
almost any snow depth, the snow could swiftly be swept away. Still, the operator had to maintain a
slower speed than during traditional ploughing. Also, in a few stretches in the test area, where the
pavement was in very bad condition, it was difficult to get good snow-clearing results, although the
broom was likely more effective on such stretches than traditional ploughing.
The friction measurements, performed both in the pilot study and in the large-scale study,
showed that the friction level on the cycleways maintained with the “brine method” was considerably
higher than on cycleways maintained traditionally. At the time of the measurements, the surface on the
cycleways included in the test was bare and wet and there was snow on the cycleways used as control.
It is not surprising that a snowy surface is more slippery than a bare surface. Nevertheless, this showed
that the test method using a broom for snow clearance and brine for de-icing resulted in a surface less
slippery than would be the case with the maintenance methods normally used.
5
In the questionnaire survey conducted within the large-scale study (mainly to evaluate the
winter of 1999/2000), 43% of the respondents stated that they would cycle more during the winter if
the maintenance service level of cycleways was improved. A total of 62% thought that winter
maintenance on cycleways needed to be improved, 12% thought that it was satisfactory, and 25% were
uncertain or without an opinion. Naturally, most of those who were uncertain or without an opinion
were those who did not cycle to work. This also applied to those who were satisfied with the winter
maintenance. However, there were a number of winter cyclists who thought that winter maintenance
on cycleways did not need to be improved. In the questionnaire the respondents were given the
opportunity to specify how winter maintenance on cycleways should be improved. Most of the
answers (162) suggested improved skid control, for example: “gritting should be done more often”,
“prevent slush from creating frozen tracks”, and “use salt on cycleways”. Many (141) also suggested
better snow clearance, such as “clear the cycleways more often”, and “clear the cycleways earlier in
the morning”.
According to their mode choice for journeys to work in summer and winter, the respondents
were divided into different categories of “cyclist”: “winter cyclist”, “summer-only cyclist”,
“infrequent cyclist”, and “never cyclist”. A winter cyclist is defined as a person who uses a bicycle for
travelling to work in at least two cases out of five during the period from November to March. A
summer-only cyclist is defined as a person who uses a bicycle for travelling to work in at least two
cases out of five during the period from April to October, but less during the period from November to
March. An infrequent cyclist is a person who cycles only occasionally, fewer than two cases out of
five, when travelling to work, no matter the season; and a never cyclist is a person who never uses a
bicycle for a journey to work. In the survey, 51% were winter cyclists, 24% summer-only cyclists, 9%
infrequent cyclists, and 16% never cyclists. It should be noted that the large number of winter cyclists
in this survey is probably a lot higher than for an average Swedish workplace.
Of the 570 respondents, 214 lived within the test area of Ekholmen, and of those 128 were
classified as winter cyclists. Winter cyclists within the test area were found to be more satisfied with
the maintenance service level of cycleways during the winter of 1999/2000, compared to winter
cyclists in the control areas (Table 1). This indicates that, in accordance with the measurements of
friction, and the road condition observations, the test method did produce a higher maintenance service
level than traditional methods.
Table 1: Respondents Satisfied with the Maintenance Service Level of Cycleways Concerning Different
Road Conditions in the Test Area Compared to the Control Areas.
Road Condition:
Slush
Loose Snow
Black Ice
Packed Snow/ Thick Ice
Total Average:
Satisfied respondents in the:
Test Area
Control Areas
49%
28%
62%
44%
50%
25%
50%
24%
53%
30%
In addition, a majority of the winter cyclists in the test area thought that the maintenance service
level during the large-scale study in 1999/2000 was higher compared to earlier winters (Figure 1).
Also in the control areas, many winter cyclists thought that the service level of cycleways had
improved during the test winter of 1999/2000. However, the number was not as striking as for the test
area, in Ekholmen. It should be mentioned that winter cyclists in the control group in Hjulsbro, the
first control group in Figure 2, were to some extent affected by the test, since the last part of their
cycle route to Saab AB was located within the test area.
6
678
Control Areas
Test Area
100
90
80
Percent
70
60
No Opinion/Unsure
50
A Lot Worse
40
Slightly Worse
30
No Difference
20
Slightly Better
10
A Lot Better
0
Ekholmen
Hackefors
Hjulsbro
Vasastan
Figure 2. The Respondents’ Evaluation of the Maintenance Service Level of Cycleways During the Test
Winter of 1999/2000 Compared to Earlier Winters.
Even though most of the winter cyclists living in the test area, Ekholmen, were satisfied with the
maintenance of cycleways provided during the winter of 1999/2000, and thought that it had been
better compared to earlier winters, 44% were against the use of salt on cycleways. However, the
attitude towards the use of salt on cycleways to combat icy conditions was more positive within the
test area compared to the control areas. Of winter cyclists living in the test area of Ekholmen, 43%
were positive to the use of salt compared to 23% of winter cyclists in the control areas. In total, all
respondents included, 26% were positive to the use of salt on cycleway, 53% were against its use and
20% were unsure. A majority (52%) of those who were positive lived in Ekholmen, and thus had
experienced the use of salt on cycleways, which was not the case for those who lived in the control
areas.
The results from the questionnaire survey, concerning the use of salt on cycleways, can be
compared to interviews conducted in the pilot study. Of the 122 people interviewed, on five different
occasions, a majority (53%) thought that it was acceptable to use brine on cycleways, while 30% were
against its use and the remainder were unsure.
6. Conclusions and Discussion
There seems to be a prevailing discontent among the public concerning winter maintenance of
cycleways. This indicates that there is a need to improve winter maintenance on cycleways. However,
it is unclear if the dissatisfaction is due to insufficient service level requirements, or if the
requirements in reality are poorly met. If we want people to use their bicycles whenever possible, they
have to be provided with safe and accessible cycleways. Wet snow freezing and creating icy tracks is
the road condition cyclists fear most, and slippery surfaces of all kinds, including grit on bare surfaces,
create a safety hazard for cyclists. Although slippery surfaces are of great importance for the safety of
cyclists, cycleways not cleared from snow seem to be more important in the choice to cycle or not
during winter. Further studies need to be carried out to clearly define a good road standard from a
cyclist’s point of view. When striving for good winter maintenance standards, the structural standards
of the pavement should not be forgotten. Potholes or other irregularities that create an uneven surface
can negatively affect the results of snow clearance.
Surveys presented in this paper indicate that improved winter maintenance on cycleways could
lead to increased cycling. Since distance seems to be more important for the mode choice during
winter than in summer, the critical distance of which one can hope to transfer some of the car trips to
bicycle is shortened from 5 km in summer to about 3 km during winter.
7
Winter maintenance methods used on cycleways today are often adapted to the prevailing
conditions on motor traffic roads. Consequently, they are not necessarily the best methods for bicycle
traffic. However, there are equipment and methods available that are better adapted to cycleways.
Since the surface conditions are very important for the safety and accessibility of cyclists, it is
important that these methods are more widely used. It is also important to improve the methods
available to better suit their purpose and also to become more cost effective. A combination of
different methods adjusted to present weather and road conditions is likely to be the best solution.
Measurements of friction, road condition observations, and a questionnaire survey, presented in
this paper, showed that a method using a front-mounted broom for snow clearance and brine for deicing produced a higher maintenance service level than methods normally used on cycleways. In
particular during spring, in combination with the midday thaw, this method proved to be efficient for
clearing cycleways. Thus, the method using a front-mounted broom for snow clearance and brine for
de-icing, is probably a good method for regions with low snow accumulations but with major ice
formation problems. Linköping and many other municipalities in southern Sweden have winter
conditions of this kind. Also in regions with a colder climate such as northern Sweden, this method is
probably advantageous during spring and fall when the temperatures are higher and the amount of
snow is less; during winter, however, other methods are likely to be better suited. A drawback with the
method using a front-mounted broom for snow clearance was that the operator had to maintain a
slower speed than during traditional ploughing. This increases the time to operate and hence increases
the cost.
A majority of the winter cyclists living in the test area were satisfied with the maintenance
service level achieved with the method using a front-mounted broom for snow clearance and brine for
de-icing, and thought that it was improved compared to earlier winters. Nevertheless, many were still
against the use of salt on cycleways. The fact that the attitude towards the use of salt on cycleways was
more positive within the test area compared to that in the control areas indicates that the advantages of
using salt become more evident for the road users when experienced directly. However, if the common
opinion of the public is that salt should not be used on cycleways, it can be difficult introduce such a
method. The use of salt should always be as moderate as possible due to its environmental side effects.
Its advantages and drawbacks need though to be compared with alternative methods such as the use of
abrasives. On some occasions the use of salt can be more cost effective, even when the environmental
effects have been taken into consideration. Further studies comparing the impact of abrasives and salt
on the environment with security and economy are necessary to be able to make the right decisions
concerning winter maintenance of cycleways and footways.
7. Acknowledgements
The financial support given by the Swedish National Road Administration through the Centre
for Research and Education in Operation and Maintenance of Infrastructure is gratefully
acknowledged.
8. References
Bergström, A. (1999). “Winter cycling — The importance of road condition in selection of
transport mode.” [in Swedish, English summary]. VTI meddelande 861, Swedish National Road and
Transport Research Institute, Linköping, Sweden.
Bergström, A. (2000). “Winter maintenance service levels on cycleways.” Licentiate thesis,
TRITA-IP FR 00-80, Div. of Highway Engrg., Dept. of Infrastructure and Planning, Royal Institute of
Technology, Stockholm, Sweden.
Bergström, A. (2001). “Friction Measurements on Cycleways Using a Portable Friction Tester.”
To be published.
Binderup Larsen, L. et al. (1991). “Single accidents among cyclists.” [in Danish]. UlykkesAnalyseGruppen, Odense, Denmark.
Emmerson, P., Ryley, T. J., and Davies, D. G. (1998). “The impact of weather on cycle flows.”
Transport Research Laboratory, Berkshire, England.
8
Gabestad, K. (1988). “A manual for planning and carrying out road surface condition studies in
the wintertime.” [in Norwegian]. Volume 1, TØI report 0013, Norwegian Institute of Transport
Economics, Oslo, Norway.
Giæver, T., Øvstedal, L., and Lindland, T. (1998). “Geometric design of bicycle facilities —
Interviews and route choice studies.” [in Norwegian]. SINTEF report STF22 A97615, SINTEF Bygg
og miljøteknikk, Trondheim, Norway.
Herrstedt, L., Lei Krogsgaard, K. M., Nilsson, P. K., and Jensen, O. K. (1995). “The Potential of
the Bicycle in City traffic.” [in Danish, English summary]. Vejdirektoratet, Trafiksikkerhed og miljö
R17, Copenhagen, Denmark.
Lindmark, M., and Lundborg, G. (1987). “Maintenance and operation of footways during
winter.” [in Swedish]. Swedish Association of Local Authorities (SALA), Report 11, Stockholm,
Sweden.
Lundkvist, S.-O., and Lindén, S.-Å. (1994). “Road marking friction — A comparison between
the SRT pendulum and the VTI Portable Friction Tester.” VTI notat 65A-1994, Swedish National
Road and Transport Research Institute, Linköping, Sweden.
Mikkelsen, L., and Prahl, K. B. (1998). “Use of brine to combat icy bicycle lane surfaces.” Xth
PIARC International Winter Road Congress, 16–19 March in Luleå, Linköping, Sweden.
Möller, S., Wallman C.-G., and Gregersen, N. P. (1991). “Winter road maintenance in urban
areas — Road safety and trafficability.” [in Swedish]. TFB & VTI forskning/research No. 2, Swedish
Transport Research Board and Swedish National Road and Transport Research Institute, Linköping,
Sweden.
Möller, S., and Öberg, G. (1990). “Instructions for road condition observations.” [in Swedish].
VTI notat T 83, Swedish National Road and Transport Research Institute, Linköping, Sweden.
Nilsson, A. (1995). “The potential for replacing cars with bicycles for short distance travel.” [in
Swedish, English summary]. Thesis 84, Lund Institute of Technology, Dept. of Traffic Planning and
Engrg., Lund, Sweden.
Nordic Road Association (NVF). (1984). “Maintenance and operation of facilities for cyclists
and pedestrians.” [in Norwegian, Danish, and Swedish]. NVF Report No. 24:1984, Oslo, Norway.
Nordic Road Association (NVF). (1999). “Equipment for maintenance and operation of
cycleways and footways — Test of vehicles and equipment.” [in Norwegian]. NVF Report No.
4:1999, Oslo, Norway.
Riks-RVU, Svenskarnas resor 1998. (1998). “The Swedish Travel Survey of 1998.” [In
Swedish]. Statistiska centralbyrån, Stockholm, Sweden.
Swedish Association of Local Authorities (SALA). (1998). “Evaluating municipal services —
presentation of a questionnaire survey in public administration in 1997”. [in Swedish]. Stockholm,
Sweden.
Öberg, G., Nilsson, G., Velin, H., Wretling, P., Berntman, M., Brundell-Freij, K., Hydén, C.,
and Ståhl, A. (1996). “Single accidents among pedestrians and cyclists.” [in Swedish, English
summary]. VTI meddelande 799, Swedish National Road and Transport Research Institute, Linköping,
Sweden.
9
DE-ICING SALT AND ROADSIDE ENVIRONMENT
– STRATEGIES FOR IMPACT ANALYSES
Göran Blomqvist
Swedish National Road and Transport Research Institute, VTI, SE-581 95 Linköping, Sweden
TEL. +46-13204147/FAX +46-13204145
1. Abstract
Society needs to maintain road safety and accessibility of the road network at acceptable levels
during the winter season. The use of sodium chloride as de-icing medium can lead to several impacts
on human health and nature, as for instance damage to ground water resources and vegetation. The
question of whether the political goals of accessibility, transport quality and safety can be fulfilled at
the same time as the goal of a good environment is fulfilled, must be seen as a delicate matter of
conflicting interests.
In order to be able to evaluate countermeasures taken against the undesired impacts, the system
needs to be monitored with indicators at several levels within the system. An integrated environmental
assessment framework that is suitable for such evaluations is the DPSIR-approach. It is for instance
used by the Swedish Environmental Protection Agency for the follow-up of the national
environmental quality objectives in Sweden. According to this framework there is a chain of causal
links, from the societal need for transportation as driving force (D) of the system, over the pressure (P)
of roadside exposure to salt, to an altered state (S) of the roadside environment leading to different
kinds of impact (I), which may require some kind of societal response (R).
In most cases it is important to find useful indicators as early in the system as possible, especially
when the environmental effect is delayed in time, as for instance regarding contamination of ground
water resources. In that case an early warning could be reached.
By assigning adequate indicators to the different levels of the DPSIR model, the road keeper will
not only strengthen his scientific understanding of the ecological effects, but also increase his
possibilities to take appropriate measures to improve the sustainability of the system and finally
increase the knowledge of the environmental utility of the strategic actions taken.
2. Introduction
In June 1998, the Swedish Parliament adopted a new transport policy on the basis of the
Government Bill “Transport policy for sustainable development” (1997). The overall goal of the
transport policy is defined to be a transport system that is environmentally, economically, culturally,
and socially sustainable. The overall goal was divided into five sub-goals: an accessible transport
system, a high transport quality, a safe traffic, a good environment, and a positive regional
development. In addition to that, the Swedish Roads Act (1971, section 23) states that roads shall be
held in a satisfactory state by maintenance and other measures. Therefore, in order to maintain road
safety and accessibility of the road network at acceptable levels also during the winter season, the
roads are kept free from ice and snow by ploughing and by the use of chemical de-icing. The winter
road maintenance regulations of Sweden (Drift 96…, 1996) prescribe sodium chloride as the only
allowed chemical de-icing agent to be used. Unfortunately the salt solution does not stay on the road
surface where it has its desired effects, but will by different mechanisms be dispersed into the roadside
where it may lead to undesired environmental impacts (Blomqvist, 1999; Thunqvist, 2000). The
question of whether the goals of accessibility, transport quality and safety can be fulfilled at the same
time as the goal of a good environment is fulfilled, must therefore be seen as a delicate matter of
conflicting interests.
The Swedish National Road Administration (SNRA) is responsible for the winter road
maintenance of about 98 000 km of state roads in Sweden (Ölander, 2000). Twenty-five per cent of
the SNRA appropriation for road maintenance and operations is spent on winter road maintenance
works, such as snow ploughing and de-icing (Ölander, 2000). The de-icing salt use on the national
1
Salt use (tonnes per season)
road network has approximately doubled since the 1970’s and has for the last six seasons varied in the
size between 200 000 and 300 000 metric tonnes (figure 1).
500 000
450 000
400 000
350 000
300 000
250 000
200 000
150 000
100 000
50 000
1990’s
1999 / 00
1980’s
1989 / 90
1979 / 80
0
Figure 1
The Seasonal Salt Use On The Swedish National Road Network (Metric Tonnes
Per Season). The Arrows Depict The Changes In The Winter Road Maintenance Regulations
During The Last Decade. Before That The Winter Maintenance Was Regulated In Five-year
Plans.
The SNRA has a constant ambition of improving the requirements of the winter road maintenance
regulations. Therefore the regulations have changed several times during the last decade (figure 1). In
1996 a limit value of 200 000 tonnes salt per year was set up as a goal to be reached by the year 2000
(Kretsloppsanpassad väghållning… 1996). This goal was indeed reached as the salt-use in the calendar
year of 2000 was 196 700 tonnes (Pettersson, personal communication). For the future, the strategy
for decreasing the salt use is the development of a salt index that allows the actual salt use to be
compared to the salt needed according to the weather conditions prevailing during the entire winter
(Ölander, personal communication). In that way the actions of the contractors can be compared to the
requirements of the regulations.
A key issue when taking management decisions in the road sector is to ensure that the limited
funds are spent to greatest effect within the various constraints that pertain (Robinson et al., 1998).
Knowledge of the different interrelationships within the system is therefore of great importance. Since
1999 the Swedish Environmental Code (1998, chapter 2, section 2) states that those who pursue an
activity or take a measure, or intend to do so, must possess the knowledge that is necessary in view of
the nature and scope of the activity or measure to protect human health and the environment against
damage or detriment. After decades of investigations, however, we still have to deal with the problem
of environmental effects of the use of de-icing salt in the winter road management. The regulations of
the winter road management have changed four times during the 1990’s (figure 1) but, since we don’t
have useful indicators of the environmental pressure, states and impacts, we still don’t know the
environmental utility of these changes. The objective of this paper is to describe the system of de-icing
practices and their environmental effects with special reference to the salt exposure of the roadside
environment and damage to Norway spruce seedlings. The objective is also to describe a monitoring
system and discuss the importance of indicators for the follow-up, which ultimately will increase the
knowledge of the environmental utility of the strategic actions taken by the road administrator.
3. Describing the system
Full understanding of the total system is probably not possible, but by simplifying the system of
the real world into a model to start with, a conceptual understanding of the total system could be
reached.
2
Atmospheric deposition
Splash and
ploughing
Wind
1
2
2
Run-off
Spray
4
5
10
3
7
6
9
8
Groundwater transport
Figure 2
A Conceptual Model Of The Transport Mechanisms And Pathways From The
Road.
The system of de-icing salt can be described in several ways. This can be illustrated e.g. as in the
picture above (figure 2) showing the transport mechanisms and pathways from the road, or as a box
model divided into several compartments as shown below (figure 3). The de-icing salt is either
transported through the compartments or accumulated within them.
4
Salting
vehicle
5
Air
Vegetation
9
1
5
Road
surface
3
Road
construction
5
2
5
Surface &
Snow layer
Roadside
7
2
Drainage
system
5
10
6
Surface
water
7
Soil
8
Ground
water
Figure 3
A Box Model Of The Physical System Of De-icing Salt Migration From The
Vehicle That Carries On The De-icing Actions To The Compartments On, In Or Around The
Road Which Are Involved In The Migration Or Accumulation Of De-icing Salt. The Boxes
Depict The Compartments Where The De-icing Salt Either Is Passing Through Or – To Some
Extent – Is Retained Or Accumulated. The Arrows Depict The Mechanisms That Govern The
Migration Of The De-icing Salt, Such As: Deposition, Run-off, Infiltration, Percolation, and
Root Uptake.
The de-icing vehicle applies the salt on to the road surface1. This is the action where the entire
system origins. The salt will then leave the road surface by itself (by gravity) or by the action from
traffic in the following ways. By run-off,2 the salt will reach the roadside or the drainage systems.
Some parts may infiltrate3 the road surface and reach the interior of the road construction. By being
forced into the air4 by the vehicles or by ploughing, the salt leaves the road as splash, spray or dry
crystals to be deposited5 onto the road surface or roadside (roadside cover, ditch, etc.) of the
technosphere or on the vegetation, soil surface, snow layer or surface waters in the surrounding
3
ecosphere. By leaving the drainage system6 or percolating7 from the roadside or soil surface through
the soil the salt solution may reach the groundwater8. Where the soil solution and groundwater are in
contact with the root zone of vegetation, uptake9 through the roots may occur. Some part of the salt
deposited onto the foliage, stem and branches of the vegetation will enter the interior of the plant, but a
large portion will be transported as throughfall10 and stemflow10 to the soil surface beneath the
vegetation.
The pathways by which the de-icing salt may reach the different plant parts have been discussed
extensively in the literature. There is no doubt that damage may occur either when the salt is deposited
onto the foliage and when it reaches the root system. This has been shown both in field studies and in
laboratory studies under controlled conditions (Dobson, 1991; Brod, 1993; Pedersen and Fostad,
1996). When the salt is deposited on to the foliage it may either stay on the exterior parts (needles,
leaves, stem etc.) or be transported to the interior of the plants either through the leaf cuticle or
through the bark on the branches or stem. Also the stomata have been suggested to be a pathway to the
interior parts (Burkhardt and Eiden, 1994). Different kinds of particles deposited on the exterior parts
of needles have been suggested to lead to additional damage (Flückiger et al. 1977).
The symptoms of salt damage in coniferous trees are often described as needle browning and
needle loss (Pedersen and Fostad, 1996). Some trees are able to compensate for damage by producing
new shoots, but when the damage is too great, this is not possible (Pyykkö, 1977; Blomqvist, 1999).
The consequences of this damage are many. One is the impact on biota in itself; another is the
effect on the landscape. The impact of de-icing salt on conifers is a result of a complex interplay of
many causal relationships (e.g. loss of needles: lower photosynthetic capacity; increased amount of
salt in soil water: osmotic stress: inhibition of water uptake; stress avoidance of the plant: energy
expenditure). Most of these effects will in the end result in diminished growth of the tree stand and can
also predispose the tree to damage from fungi or insects. Such effects have been described by e.g.
Pedersen and Fostad (1996). It is often difficult to separate between different stress factors since one
factor may have predisposed the tree to damage, another factor can have triggered the damage, and a
third factor may have contributed to the actual killing of the tree (Aronsson et al., 1995).
The extent of damage is governed by some kind of dose-response function. For some species, the
function has been suggested to be S-shaped (Figure 4) (Dragsted, 1990). Many investigations of the
amount of chloride in e.g. needle tissue compared to the extent of visible damage symptoms have been
performed and comprehensive compilations have been published by e.g. Dobson (1991) and Brod
(1993). The theoretical extent of damage can be calculated by combining the distance–exposure and
exposure–damage functions (Figure 4). One should, however, keep in mind that the roadside
environment is also exposed to many other stress factors (Viskari, 1999).
Exposure–Damage
Distance–Exposure
Distance–Damage
Distance
=
Exposure
Damage
+
Damage
Exposure
(Dose–Response)
Distance
Figure 4
Theoretical Functions Of Exposure To Salt And Susceptibility To Salt Damage
Give The Pattern Of Damage In The Roadside Environment
Forestry should be thought of as a process rather than a steady state. While one effect of the
roadside exposure to de-icing salt is a lower yield from forestry, a possibly more important long-term
consequence may manifest itself when reforestation is to take place. The seedlings and young plants
are much more sensitive to salt exposure than are older trees. Reforestation may therefore be virtually
impossible in a zone of up to several tens of metres from the road. This extends in many places beyond
the road reserve area and, hence, may affect the land next to the road. If reforestation is made
impossible the landowner is subjected to a forced change of land-use (Figure 5), which probably will
4
Growth
lead to a different situation concerning the legislative possibilities to claim for damages, than do the
impact of diminished growth in the roadside.
Diminished
growth
Change of
land-use
Time
Reforestation
Figure 5
Implications for forestry, seen as a process
Sweden has a long tradition of monitoring salt concentrations in groundwater. Starting already in
the late 1970’s, Bäckman has been monitoring sodium and chloride concentrations in observation
wells as influenced by de-icing salt. Long-term increases have been documented (Bäckman 1980,
1997). The same result has been obtained in long-term monitoring of chloride concentrations in
municipal groundwater supplies (Knutsson et al. 1998; Rosén et al. 1998). Likewise, a long-term
increase trend was documented by Thunqvist (2000) compiling data from up to 23 Swedish municipal
groundwater plants for the period 1954–1999. By compiling data from 13 000 private drilled wells,
Olofsson and Sandström (1998) found that wells located close to major roads had increased
concentration of chloride.
The salt that has percolated through the road construction, or roadside and reached the soil or
groundwater will to a large extent be transported with the groundwater to the surface waters and then
follow the water cycle to the sea. In that sense this is – in a very long time-scale – a geochemical cycle
of salt, extracted from the seas or rocks, placed on the roads and then returning to the seas again. On
the road surface, most effects are desired and in the seas salt is at least not harmful. The important
issue is what happens in between (Thunqvist, 2000).
4. Monitoring the system
Robinson et al. (1998) have stated that: “A key challenge for the road manager is to find ways in
which to describe the problems and impacts of road maintenance that can be understood by the
politicians and the general public”. One could also state that a crucial challenge for the scientific
community would be to find key parameters and indicators of the system’s different compartments
that can be understood and utilised by the road managers (Blomqvist, 2001b). A system that is used by
the European Environmental Agency (EEA) as a generic tool to support understanding of complex
environmental systems is the DPSIR model (Towards a transport and environment reporting
mechanism ‘TERM’ for the EU, 1999). This system is used for facilitating communication and is
based on indicators of the different compartments. Societal needs and activities can be viewed as
driving forces (D) that lead to a pressure (P) on the environment. The pressure may change the state
(S) of some compartments of the environment. This, in turn, can lead to impacts (I) on a system such
as human health or nature. Finally, the society will respond (R) in some way to combat the problem in
one or several of the earlier stages in the model (Figure 5).
5
Driving
forces
Response
Pressure
Impact
State
Figure 5
The DPSIR Framework For Reporting On Environmental Issues
Using the DPSIR-model (figure 5), the use of de-icing salt and damage to vegetation could be
described as follows. The need for transportation (D) leads to a roadside exposure to salt (P), which
alters the state of the vegetation (S), thereby leading to different impacts (I), which may require some
kind of response (R) from society (Figure 6).
Laws, directives, policies...
Impacts:
Need for
transportation
- Public (road users) -
D
EU directives
Landscape
scenery
De-icing
action
The Environmental
Code
Environmental
quality objectives
- - - Land-owner - - Diminished
tree growth
Transport
policy
Change of
land-use
Roadside
exposure
P to salt
etc...
- - - Ecology - - Stress
Species
composition
State of
vegetation
S
Figure 6
Response
I
R
The System Of De-icing Action And Damage To Vegetation As Illustrated By The
DPSIR model.
The impacts have been divided into different spheres of interest where the interest of the public as
road users is threatened by the impaired landscape scenery. The interest of the rural landowners can be
threatened by the diminished tree growth or the forced change of land-use that may occur if
reforestation is made impossible in the salt exposed environment. The vegetation as part of the
ecosystem can be influenced by being stressed and as a result there may be a change in the species
composition of the roadsides.
By adding two new compartments to the original DPSIR model, the model is made suitable to be
used for identifying the involved processes. The two new compartments are the first oval between the
driving force and pressure (figure 6) which represents the activity that is induced by the driving force,
causing the pressure. In this case it is the actual de-icing measure taken. The other new compartment
in figure 6 is the box of laws, directives, policies, regulations, contract conditions, etc. These are partly
a result of the needs within the society (e.g. need for transportation is manifested in some of the goals
6
of the transport policy), but it can also be used as a toolbox of the society to respond to all of the
stages in the DPSIR model.
In most cases it is important to find useful indicators as early in the system as possible, especially
when the environmental effect is delayed in time, as for instance regarding contamination of ground
water resources. In that case an early warning could be reached.
By assigning adequate indicators to the different levels of the DPSIR model, the road keeper will
not only strengthen his scientific understanding of the ecological effects, but also increase his
possibilities to take appropriate measures to improve the sustainability of the system. The road keeper
needs also to know the environmental utility of the responses he has taken.
6. Ongoing research
Research at the Swedish National Road and Transport Research Institute (VTI) in Linköping,
Sweden is currently addressing the issue of roadside exposure to salt and the development of
indicators of different components within the system as described above, both regarding groundwater
contamination and damage to vegetation.
7. Acknowledgement
This paper is based on the results of the licentiate thesis “Air-borne transport of de-icing salt and
damage to pine and spruce trees in a roadside environment”(Blomqvist, 1999) and PhD thesis “Deicing salt and the roadside environment: Air-borne exposure, damage to Norway spruce and system
monitoring” (Blomqvist, 2001). They are results from the project ”Influence of De-icing salt on
Roadside Vegetation” (VTI Project No 80131) performed at VTI during 1996–2001. The work has
been financed by the Swedish National Road Administration through the Centre for Research and
Education in Operation and Maintenance of Infrastructure (CDU) at the Royal Institute of Technology
(KTH) in Stockholm and by the Swedish National Road and Transport Research Institute (VTI) in
Linköping, Sweden.
8. References
Aronsson, A., Barklund, P., Ehnström, B., Karlman, M., Lavsund, S., Lesinski, J.A., Nihlgård, B., and
Westman L. (1995) Skador på barrträd, Skogsstyrelsen, Jönköping. (In Swedish).
Bäckman, L., (1980) Vintervägsaltets miljöpåverkan, VTI Rapport Nr 197, National Road and Traffic
Research Institute, Linköping, 62 pp. + app. (In Swedish).
Bäckman, L., (1997) Vintervägsaltets miljöpåverkan – Resultat av jord- och grundvattenprovtagningar
vid observationsområden i Skaraborgs län 1994–1996, VTI-notat Nr 25-1997, Swedish National
Road and Transport Research Institute, Linköping. (In Swedish).
Blomqvist, G., (1999) Air-borne transport of de-icing salt and damage to pine and spruce trees in a
roadside environment, Licentiate Thesis, TRITA-AMI-LIC 2044, Division of Land and Water
resources, Department of Civil and Environmental Engineering, Royal Institute of Technology,
Stockholm, Sweden. ISBN 91-7170-475-2.
Blomqvist, G., (2001a) De-icing salt and the roadside environment: Air-borne exposure, damage to
Norway spruce and system monitoring, PhD thesis, TRITA-AMI-PHD 1041, Division of Land and
Water resources, Department of Civil and Environmental Engineering, Royal Institute of
Technology, Stockholm, Sweden. ISBN 91-7283-081-6.
Blomqvist, G., (2001b) Long-term effects of deicing salt on the roadside environment. Part I: Forestry,
Transportation Research Board Conference Proceedings 23:179–185.
Brod, H.G., (1993) Langzeitwirkung vos Streusalz auf die Umvelt, Verkehrstechnik Heft V2,
Bundesanstalt für Straβenwesen, Bergisch Gladbach. (In German).
Burkhardt, J., and Eiden, R. (1994) Thin water films on coniferous needles, Atmospheric Environment
28(12):2001–2017.
D’Itri, F.M., (1992) Chemical Deicers and the Environment, Lewis Publishers, Inc., Chelsea,
Michigan, USA, 585 pp.
Dobson, M.C., (1991) De-icing salt damage to trees and shrubs, Forestry Commission Buletin 101,
Department of the Environment Arboriculture Contract, Forestry Commission, Farnham, Surrey.
Dragsted, J., (1990) Some results from Danish investigations in salt stress on trees. Aquilo, Series
Botanica 29, 21–23.
7
Drift 96, Allmän teknisk beskrivning av driftstandard, (1996) Publ 1996:016, Swedish National Road
Administration, Borlänge. (In Swedish).
Flückiger, W., Flückiger-Keller, H., Oertli, J.J., and Guggenheim, R., (1977) Versmutzung von Blattund Nadeloberflächen im Nahbereich einer Autobahn und deren Einfluβ auf den stomatären
Diffusionswiederstand, European Journal of Forest Pathology 7:358–364. (In German).
Holldorb, C. (2000) Winterterm – Dictionary of Terms for Winter Maintenance, draft version,
European COST 344 Action “Improvements to Snow and Ice Control on European Roads and
Bridges”.
Knutsson, G., Maxe, L., Olofsson, B., and Jacks, G., (1998) The origin of increased chloride content in
the groundwater at Upplands Väsby, In: Nystén, T, and Suokko, T., Deicing and dustbinding –
risk to aquifers, Proceedings, NHP Report No 43, pp 223–231, Finnish Environment Institute,
Helsinki.
Kretsloppsanpassad väghållning – Handlingsplan. (1996) VV Publ 1996:29, Swedish National Road
Administration, Borlänge. (In Swedish).
Ölander, J., (2000) Winter Road Maintenance – The Swedish way, Proceedings, Talvitiepäivät Winter Road Congress, Finnish National Road Administration, Feb 2–3, 2000, Tampere, Finland.
Ölander, Jan, Swedish National Road Administration, 2001-04-02, personal communication.
Olofsson, B., and Sandström, S., (1998) Increased salinity in private drilled wells in Sweden – Natural
or Man-made?, In: Nystén, T, and Suokko, T. (eds), Deicing and dustbinding – risk to aquifers,
Proceedings, NHP Report No 43, pp 75–81, Finnish Environment Institute, Helsinki.
Pedersen P.A., and Fostad, O., (1996) Effects of deicing salt on soil, water and vegetation. Part I:
Studies of soil and vegetation, MITRA Nr 01/96, Statens Vegvesen, Oslo. (In Norweigan, English
summary).
Pettersson, Ola, Swedish National Road Administration, 2001-04-03, personal communication.
Pyykkö, M., (1977) Effects of salt spray on growth and development of Pinus sylvestris L. Ann. Bot.
Fennici 14, 49–61.
Robinson, R., Danielson, U., Snaith, M. (1998) Road Maintenance Management, Concepts and
Systems, The University of Birmingham and The Swedish National Road Administration,
Macmillan Press Ltd, 291 pp.
Rosén, B., Lindmark, P., Knutz, Å., and Svenson, T., (1998) Municipal well along highway damaged
by de-icing – a local case study at Brännebrona, Sweden, In: Nystén, T, and Suokko, T., Deicing
and dustbinding – risk to aquifers, Proceedings, NHP Report No 43, pp. 245–251, Finnish
Environment Institute, Helsinki.
System med indikatorer för nationell uppföljning av miljökvalitetsmålen, (1999) Rapport 5006,
Swedish Environmental Protection Agency, Naturvårdsverkets förlag, Stockholm.
The Environmental Code (1998), SFS 1998:808, Statute book of Sweden. (In Swedish).
The Roads Act (1971), SFS 1971:948, Statute book of Sweden. (In Swedish).
Thunqvist, E.-L., (2000) Pollution of groundwater and surface water by roads – with emphasis on the
use of deicing salt, Licentiate Thesis, TRITA-AMI-LIC 2054, Division of Land and Water
Resources, Department of Civil and Environmental Engineering, Royal Institute of Technology,
Stockholm.
Towards a transport and environment reporting mechanism (TERM) for the EU, (1999) Technical
report No 18, Environmental Environment Agency, Copenhagen.
Transport policy for sustainable development (1997), SFS 1997:652, Statute book of Sweden. (In
Swedish).
Viskari, E.-L., (1999) Disperson, deposition and effects of road traffic-related pollutants on roadside
ecosystem, Doctoral dissertation, Kuopio University Publications C, Natural and Environmental
Sciences 88, Department of Ecology and Environmental Science, University of Kuopio, Kuopio,
Finland.
8
WEATHER DESCRIPTIONS AND COMPENSATION MODEL FOR
WINTER ROAD MAINTENANCE
Staffan Moller and Carl-Henrik Ulegard
Swedish National Road and Transport Research Institute and Swedish National Road Administration
Contact address: Swedish National Road and Transport Research Institute,
SE- 581 95 Linkoping, Sweden
TEL.: + 46 13 20 41 61 / FAX: + 46 13 20 41 45
E-mail address: staffan.moller@vti.se
Abstract
A good compensation model for regulating costs for winter road maintenance between client and
contractor requires two well-functioning sub models.
•
•
A sub model that describes the weather during the winter season.
A sub model that links the weather descriptions to the need to take measures/set in resources.
The basis for the weather descriptions are data collected from the individual stations in the Swedish
National Road Administration system for road weather information, RWiS. Through using special
definitions, the data are translated into eight weather situations at an hourly level. Examples of
weather situations are snowfall, drifting snow and risk of slipperiness caused by rain or sleet on a cold
roadway.
The hour-by-hour weather descriptions are then summarised into clearly defined weather
occurrences, for instance drifting snow during 6 hours or a snowfall lasting 20 hours with a snow
depth of 10 cm. The final result of weather descriptions for a winter is a number of clearly defined
weather occurrences.
The compensation model is based on the number of weather occurrences for each RWiS station
chosen as representative for a certain maintenance area. Starting from each weather occurrence the
number of so-called resultant weathers is calculated being the basis of compensation. In this step the
connection is made between weather and the need to take measures.
Background
A good compensation model (payment model) for regulating costs for winter road maintenance
between client and contractor, based on winter conditions, requires two well-functioning sub models.
• A sub model that describes the weather during the winter season.
• A sub model that links weather descriptions to the need to take measures/set in resources.
Weather Situations
Weather descriptions are based on raw data from individual stations in the Swedish National Road
Administration system for road weather information, RWiS. The following raw data are used.
•
•
•
•
•
•
•
Air temperature.
Road surface temperature.
Relative humidity.
Dew point temperature.
Precipitation type.
Precipitation quantity.
Wind speed.
1
With the aid of specific definitions, these raw data are translated from the RWiS to the eight weather
situations below on an hourly basis.
•
•
•
•
•
•
•
•
Snowfall (S).
Drifting snow (D).
Slippery surface due to rain or sleet on a cold road (SSP).
Slippery surface due to damp/wet roads freezing (SSF).
Slippery surface due to light hoar frost (SSH1).
Slippery surface due to heavy hoar frost (SSH2).
Specific weather, type 1, i.e. drifting snow with high wind speed (SW1).
Specific weather, type 2, i.e. snowfall with high snow intensity (SW2).
The following are two examples of definitions for translating raw data into weather situations:
Drifting snow (D).
• Snow likely to drift should be present (as in a specific definition).
• The average wind speed should be 5 m/s or above.
Slippery surface due to light hoar frost (SSH1).
• Road surface temperature must be at least 0.5 °C lower than the dew point temperature.
• Road surface temperature must be below + 1.0 °C.
Weather Situations On An Hourly Basis
An example of hourly weather situations is given in table 1 below. The situations are for an operating
area where three RWiS stations, 307, 312 and 320, have been selected as representative.
Station
Weather
Hour
Snow
depth (mm)
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23
307
Weather situation
307
Snow depth (mm) 0
0
0
0
0
0
0
0
0
0
0
312
Weather situation
Date: 28 Nov.
312
Snow depth (mm) 0
320
Weather situation
320
Snow depth (mm) 0
S
S
S
S
0
0
0
0
0
0
0
0
0
0
S
0 0.2 0
0
0
0
0
0
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
D
D
S
S
S
S
S
S
S
S
62
D
0 0.2 1,0 1.6 4.8 4.8 9.0 7.1 5.9 6.6
S
0
S
0 0.5 0.2 0 3.5 4.8 5.8 8.5 5.8 8.8 9.9 8.2 5.8
41
S
0
0
0
0
0
0 2.1 0 1.3 4.0 5.6 7.1 5.1 2.4 3.5 9.8 9.7
307
Weather situation S
307
Snow depth (mm) 5.0 4.6 5.5 5.9 6.0 3.8 0.6 0
0
0
0
0
0 0.3 0.8 1.6 0.2 0.2 1.5 0
312
Weather situation D
51
Date: 29 Nov.
D
S
S
S
S
S
S
S
312
Snow depth (mm) 5.0 3.1 3.4 4.2 2.3 6.1 2.3 0
320
Weather situation S
320
Snow depth (mm) 6.6 9.5 5.8 7.8 9.8 5.6 1.5 0
S
S
S
S
S
0
0
0
0
S
S
S
S
S
S
S
S
S
S
S
S
S
0
0 1.4 2.4
S
S
S
S
S
S
S
S
S
S
0
0
0 0.5 0.5 1.2 0.5 1.7 0.5 1.4 2.8 0
0 0.4 1.5 2.1 0
0
0 0.7 0.3 0.2 0
40
S
0 0.2 0.6 1.1 0.8 0.5 0.3 0.8 0.2 0 0.4 1.0
S
0
S
32
S
0 0.3 2.4
58
Date: 30 Nov.
307
Weather situation S
307
Snow depth (mm) 1.1 0
S
312
Weather situation S
0
0
S
S
312
Snow depth (mm) 1.0 0.3 0
0
320
Weather situation S
S
320
Snow depth (mm) 1.0 0
Table 1
0
0
S
S
S
S
S
S
S
S
S
S
0
0
0
0
0
0
0
0
6
0
0
0
0
0
0
0
0
5
0 1.0 0.8 0
0
0
0
0
0
0
0
8
S
0 1.0 0.8 1.6 0.3 0
S
S
0
0
S
0 0.2 0
S
0 0.5 0 0.2 0.8 1.1 2.1 0.2 0
0
0
S
72-hour Weather Situations At Hourly Intervals For An Operation Area.
2
The weather description table reveals that a relatively heavy snowfall (S) started at approximately
12.00 on 28 November and continued for around 20 hours, i.e. until 07.00 the next morning. At station
312 drifting snow (D) occurred for 5 hours around midnight at the same time as it snowed.
After a break of around 6 hours, it started snowing again. This time in the form of light snowfall
that continued for around 24 hours, although this stopped several times. When this snowfall had
ended, between 8 and 12 cm of snow had fallen, depending on the individual RWiS stations.
Payment Model
The payment model is based on weather occurrences, not on specific hours with certain weather
situations. Examples of weather occurrences are a snowfall between 06.00 and 12.00 or slippery
surface due to hoar frost between 01.00 and 05.00. The payment model is based on weather
occurrences because for example, an occurrence of hoar frost for four hours is not equivalent to four
separate hours of hoar frost on four different days. The first four-hour hoar frost may perhaps call for
just one salting run while the four separate hours would probably need four runs.
The starting point for payment calculations is the weather situations on an hourly basis set out
above. These weather situations at hourly intervals are first merged together into weather occurrences.
Each weather occurrence then generates one or several so-called resultant weathers that form the basis
for payment. The calculations are done for one RWiS station at a time and then summarised for the
whole maintenance area. It is at this stage that a relation is set up between weather and need for action.
It must be stressed that one resultant weather does not equal one action to be taken, e.g. a salting run
or a ploughing run.
The following rules apply when resultant weather calculations are performed.
Calculation Order For Different Weather Situations
1.
2.
3.
4.
5.
Specific weather, type 1 (SW1).
Specific weather, type 2 (SW2).
Drifting snow (D).
Snowfall (S).
Slippery surfaces of all types (SSP, SSF, SSH1, SSH2).
Demarcation Of Weather Occurrences
The following method for the demarcation of weather occurrences applies for all weather situations
except SW2.
The first hour during winter with the current weather situation, generally called W, is identified.
The last hour during this first occurrence of W is identified. This is found when there is a break of at
least 6 hours until the next hour of W. The following weather occurrences with W during the winter
season are demarcated in the same way. See illustration 1 below.
W
<-------------------->
≤ 5 hrs
W
≥ 6 hrs
W
<--------------->
<--------------------------->
first occurrence of W
<------------------------------------------------>
second occurrence of W
<--------------------------->
Illustration 1 Demarcation Of Weather Occurrences Of Weather Type W.
3
Specific Weather Type 1 (SW1)
1. All hours with SW1 during the current calculation period are demarcated as weather occurrences
according to the method given in illustration 1.
2. How long the SW1 occurrences found last is determined in the following way. If SW1 occurs ≥
HSW1 hours consecutively then specific weather type 1 arises. Otherwise there are just intermittent
hours of SW1.
3. The SW1 occurrences that meet the time period requirements are shown under the heading
“Resultant weather, SW1”. This states the start and finish times for each occurrence.
4. During the time period when an SW1 occurs and HSW1 after hours thereafter no calculations are done
for weather of types drifting snow, snowfall and slippery surface, (payment for the SW1 weather
plus the subsequent HSW1 after hours takes precedence). If however several SW1 or SW2 occurrences
arise within HSW1 after hours after an SW1 occurrence ends, the end time point is extended.
Specific Weather Type 2 (SW2)
1. Weather occurrences of type SW2 are calculated and how long they lasted is determined for each
RWiS station with the aid of a specific procedure. The calculations are done such that the snow
intensity shall amount to not less than ISW2 cm/hr for not less than HSW2 hrs. The SW2 occurrences
are given under the heading “Resultant weather, SW2” where the start and end time point is given
for each occurrence.
2. During the course of an SW2 occurrence and HSW2 after hours afterwards no calculations are done for
weather of types drifting snow, snowfall and slippery surface (payment for SW2 weather and the
following HSW2 after hours takes precedence). However if there are several SW2 or SW1 occurrences
within HSW2 after hours after an SW2 occurrence ends, the end time point is extended.
Drifting Snow (D)
1. All hours with D during the current calculation period are demarcated as weather occurrences in
accordance with the method in illustration 1.
2. If D arises ≥ HDRIFT hours in succession during the course of the D-occurrence the time period
requirement for drifting snow is met, i.e. drifting snow occurs. For instance, as SW1 is a heavier
type of D, a 4-hour combination of situations D, D, SW1, D can be counted as drifting snow.
3. Each D-occurrence that meets the duration requirement is divided into 4-hour intervals (the last 4hour interval can be between 1 and 4 D hours) and the number of intervals is calculated. The
amount of snow in each 4-hour interval is calculated. The intervals that have both ≤ 0,3 cm of snow
and do not have D or SV1 hours are discounted. The remaining intervals are shown in four classes
under the heading “Resultant weather, drifting snow”.
The classes are defined according to the following amounts of snow d (cm).
0,0 ≤ d ≤ 0,3
0,3 < d ≤ 1,0
1,0 < d ≤ 2,5
2,5 < d.
4
Snowfall (S)
1. All hours with S during the current calculation period are demarcated as weather occurrences
according to the method in illustration 1.
2. Each occurrence of S is divided into 4-hour intervals (the last 4-hour interval can be between 1 and
4 S hours) and the number of intervals is calculated. The amount of snow in each 4-hour interval is
calculated. The intervals that have ≤ 0,3 cm of snow are discounted. The remaining intervals are
shown in three snow quantity classes under the heading “Resultant weather, snowfall”.
The snow quantity classes are defined by the following amounts of snow d (cm).
0,3 < d ≤ 1,0
1,0 < d ≤ 2,5
2,5 < d.
Slippery Surface Occurrences Of Type SSP, SSF, SSH1 And SSH2
1. When snowfall or drifting snow occurs and up to 6 hours after such weather, no demarcation is
done for slippery surfaces. Nor during the occurrence of SW1 or SW2 and up to HSW1 after resp.
HSW2 after hours thereafter is any demarcation made for the occurrence of slippery surfaces. The
following tests are done for the remaining periods.
2. When demarcating occurrences of slippery surfaces, all types of slippery surface, i.e. SSP, SSF,
SSH1 and SSH2 are classed the same.
3. All hours of slippery surface weather during the current calculation period are demarcated as
occurrences of slippery surfaces according to the method in illustration 1.
4. The duration of the resultant weather is linked to the type of slippery surface weather, HSSP, HSSF,
HSSH1 and HSSH2 hours.
5. The slippery surface type during the first hour in the first occurrence of slippery surface (generally
called type SS1) is identified. The duration of this type is HSS1 hours. An interval of HSS1 hours is
set out from the first hour of such inclusive. A test is done to see if some type of slippery surface
weather shorter than HSS1 hours occurred within the interval. If no such type is found then a
resultant weather of type SS1 has occurred. This is shown under the heading “Resultant weather,
slippery surface type SS1”.
6. Otherwise the interval is shortened by one hour at a time and new tests are done until no slippery
surface type shorter than the length of the interval is found within the interval. Then a resultant
weather of the slippery surface type with the shortest duration within the interval (generally called
type SS2) has been found. This is shown under the heading “Resultant weather, slippery surface
type SS2”.
7. The slippery surface type in the first hour after the resultant weather found and shown according to
the above is identified and steps 5 and 6 are repeated until the first slippery surface occurrence is
resolved. The next occurrence of slippery surface is dealt with in the same way.
Implementation
Regulating the costs between client and contractor for winter road maintenance has previously been
done in accordance with three principles.
1. Payment has been based on the resources used such as ploughing hours, amount of salt spread etc.
2. The contractor has been paid for the stretch cleared, e.g. number of kilometres ploughed and
number of kilometres salted/sanded etc.
3. Payment has been based on different types of so called weather days, e.g. days with snowfall and
days with icy roads.
5
With the aid of the payment model described above, payment will be based on the same basic data that
the contractor uses to decide what action to take.
Before the start of the tendering procedure of winter road maintenance in an operating area, a
number of RWiS stations are selected that are representative for the operating area in question and that
will form the basis for payment. Three or four stations are normally selected. These can be within or
outside the boundaries of the operating area.
Data from different RWiS stations can be combined. For instance, one can choose to get
temperature and precipitation from one RWiS station and wind speed from another.
A number of parameters are then set that are included in the eight different weather situations (see
page 1) and also govern the resultant weather. Examples of such parameters are:
•
•
•
•
•
Lowest snow intensity for SW2 to arise, ISW2 [cm/hr].
Shortest duration for triggering SW2 payment, HSW2 [hrs].
Lowest wind speed for drifting snow to be declared. VDRIFT [m/s].
Shortest duration to trigger drifting snow payment, HDRIFT [hrs].
Length of resultant weather SSP, HSSP [hrs].
The resultant weather is determined for a “normal” winter in the operating area with the help of
historic data from a number of winter seasons. The number of SW1, S, SSP, SSH1 etc. is then listed.
As a part of the tendering procedure the contractor sets prices for each of them.
It should be added that in first hand the payment model described above covers snowploughing and
salting actions. The need for sanding measures is normally regulated according to some other payment
model.
6
PREDICTING STEADY STATE CONCENTRATIONS OF CHLORIDE
IN GROUNDWATER AND SURFACE WATER.
Thunqvist, Eva-Lotta
Royal Institute of Technology, Marinens väg 30, SE-136 40 Haninge, SWEDEN, Tel. +46 8 7073106,
Fax. +46 8 7073127, lotta@haninge.kth.se
Abstract
A road in operation with its traffic can pose a serious pollutant threat to groundwater and surface
water in its vicinity. Examples of pollutants are salt for deicing and dustbinding; metals from corrosion
of vehicles and wear of road surface and tires; hydrocarbons from the wear of road surface, tires,
exhaust, oils; and hazardous goods discharged in the case of an accident. In Sweden about 300000
tonnes of sodium chloride are used annually by the Swedish National Road Administration for deicing
purposes. In addition the local municipalities also use salt for deicing purposes. The use of studs
improve the friction but increase the wear and the grinding effect on winter roads. The wear of a wet
surface is reported to be two to seven times the wear of a dry surface, and hence, the grinding effect
may be further increased by the use of deicing salt.
The movement of pollutants from the road to the surrounding environment will involve run-off
from roads, airborne spreading, infiltration from road construction and road area. The chloride ion is a
good tracer. It is conservative and highly soluble and not subjected to retardation or degradation. A
small part of the sodium may be retained in soil but almost all of the deicing salt will be either
infiltrated and found in groundwater or form runoff and be found in surface water. Other, nondegradable road-related substances may be retained in soil to a greater extent. Eventually, all
pollutants from roads, which are not subjected to degradation, will be transported either to surface
water or to groundwater.
In this paper a method is presented by which the steady state concentration of chloride in
groundwater and surface water due to the use of deicing salts can be calculated. The calculations are
based on digital data for catchment areas, net recharge (precipitation with the deduction of
evapotranspiration), background deposition and road network with deicing salt application rates. All
data are processed and presented with the GIS-tool Arcview. The method makes its possible to scan an
area, e.g. a country, in order to make the decisions on what areas to protect and which measures to
adopt. The method can also be used to predict the steady state concentrations of other road related
pollutants.
Introduction
A road in operation with its traffic can pose a serious pollutant threat to groundwater and surface
water its vicinity. Examples of pollutants are salt for deicing and dustbinding; metals from corrosion
of vehicles and wear of road surface and tires; hydrocarbons from the wear of road surface, tires,
exhaust, oils; and hazardous goods discharged in the case of an accident. In Sweden about 300000
tonnes of sodium chloride are used annually by the Swedish National Road Administration for deicing
purposes of national roads (Thunqvist 2000). In addition the local municipalities also use salt for
deicing purposes.
The major roads in Sweden are deiced with 10-20 tonnes of sodium chloride per kilometre
annually. On the road the effects of the salt are desired, and in the ocean a high salt concentration is
1
natural. However, on their way the sodium and the chloride ions will pass through an environment
where the natural concentration of salt is low, involving an impact on the environment (Figure 1).
salt
desired
groundwater
surface
water
salt
natural
sea
Figure 1. The movement of deicing salt from road to sea.
The chloride ion is a good tracer. It is conservative and highly soluble and not subjected to
retardation or degradation. A small part of the sodium may temporarily be retained in soil but almost
all of the deicing salt will by infiltration or runoff reach groundwater and surface water. Several
Swedish investigations show that the chloride concentrations in both groundwater and surface water
have increased in the vicinity of roads (e.g. Bäckman & Folkeson 1995, Olofsson & Sandström 1998,
Thunqvist 2000). The movement of pollutants from the road to the surrounding environment will
involve run-off from roads, splash, airborne spreading and infiltration from road construction and road
area. Other, non-degradable road related substances may be retained in soil to a greater extent, but
they will eventually reach groundwater or surface water. There are also several investigations which
show that heavy deicing salt application increases metal mobilisation (Amrhein et al 1994, Bauske &
Goetz 1993, Norrström & Jacks 1998).
The Swedish Environmental Code (SFS 1998:808, ch.2) states that everyone is required to
possess the knowledge of the impact of one’s activities and to implement protective measures in order
to prevent impact on the human health and the environment. A simple and robust model to estimate
the concentration of chloride in recharge water based on different salt application rates would be a first
step towards that knowledge. It is of importance that the data are easily accessible and the processing
tool well known.
This paper shows how such a simple model can be used to predict chloride concentrations in
water due to application of deicing salt. If it is possible to scan an area, e.g. a county, it is possible to
make the decisions on what areas to protect and which measures to adopt. The limitations of the model
and how the results can be refined by the use of more indata or by the use of a more sofisticated model
is discussed.
Background
There are different ways of combining the data which give different precision in identifying water
in the vicinity of roads where there is a risk for high chloride concentrations. The simplest way is to
combine the maps of the road network with maps of the hydrogeological conditions and maps of lakes
and watercourses in order to identify risk areas. The result is a rough indication of where conflicts of
interest exist between roads and groundwater/surface water. This method has been used by the
Swedish National Road Administration in order to identify where the consequence of a road accident
would be serious for a water supply (Vägverket 1995). However, the investigation has been limited to
national roads and to municipal water supplies of a certain capacity. The actual number of important
aquifers and lakes/watercourses in conflict with deiced roads is much greater. Furthermore, only the
different sites of conflicting interest are listed - the increase in chloride concentration for different
areas is not calculated. In Figure 2 the digital maps of the main roads are combined with digital maps
showing the main aquifers in the county of Västmanland. Historically roads in Sweden were built on
2
good soil material high in the landscape. Thus, the roads are often built on the major eskers (which are
also important aquifers) since they provided the necessary requirements.
The winter road maintenance categories are A1 to A4, B1and B2. The A-roads are deicied
regularly, where category A1is the largest road with the most frequent recurring deicing operations.
The B1-roads are deiced occasionally, although sand with a small amount of deicing salt added is used
more frequently, and on the B2-roads even less deicing salt is used.
Winter road maintenance
A2
A3
B1
B2
Groundwater
5 - 25 l/s
25 - 125 l/s
> 125 l/s
N
W
E
S
10
0
10
20 Kilometers
Figure 2. Intersections between major roads and major aquifers in Västmanland, Sweden.
A more general method is to apply the annual amount of deicing salt within the area and the
annual net recharge to the catchment areas within the investigated area (Huling & Holocher 1972,
Howard & Haynes 1993, Thunqvist 2000). In Sweden the National Swedish Meteorological and
Hydrological Institute has estimated the catchment areas for the whole of Sweden and they are
available in digital form. If it is assumed that deicing salt application has occurred sufficient time for
steady-state conditions to be established, the average chloride concentration in the discharge will be
the same as in the net recharge. During the first years of deicing salt application the chloride
concentration in the recharge will be much higher than in the discharge. Hence, chloride will
3
concentration
accumulate in the storage and the concentration will increase. On conditions that the salt application is
invariable the chloride concentration in discharge will eventually be the same as in recharge (steady
state). The increase in concentration as a function of time is an exponential curve (Figure 3). The
calculated chloride concentration from road salt is then added to the natural background deposition for
the area.
time
Figure 3. The exponential function for a complete-mix box model.
Methods and Materials
The average concentration in net recharge can be calculated as the annual average amount of
chloride applied divided by annual net recharge for the area plus the natural background deposition.
[Cl ]tot
M Cl * m
salt
M NaCl
=
+ [Cl ]dep
(P − E) A
In order to make these calculations for a region the necessary digital data are:
-
Maps showing the catchment area obtained from the Swedish Meteorological and Hydrological
Institute (SMHI) or digital elevation data from which the catchment areas are obtained by the use
of a Geographical Information System e.g. Arcview.
-
Net recharge obtained from maps made by the Swedish Meteorological and Hydrological Institute.
-
Natural deposition of chloride (considering the difference in deposition for different land use) in
the area from the Swedish Environmental Protection Agency (SEPA 2000).
-
National road network, the different road categories and the average amount of deicing salt
applied for each category obtained from the National Swedish Road Administration.
All data are processed and presented with the GIS-tool Arcview.
Information on net recharge from SMHI is based on the annual average for the period 1961-1990.
The natural background deposition values are calculated from deposition values 1985-1989 (SEPA
2000). Deicing salt application is based on annual average values for the period 1995-1999.
4
Catchment area
N
W
E
S
10
0
10
20 Kilometers
Figure 4. Catchment areas of Mälaren in the county of Västmanland
The calculations have been made for the Swedish county of Västmanland. In Figure 4 the
catchment areas of Mälaren are shown with a mean sub catchment area of 34 km2. The calculated
chloride concentration is an average value for the recharge water in every sub catchment area.
However, the chloride concentration in discharge water for a sub catchment area is a function of the
concentration in recharge in the area and the concentration in water from areas upstream the basin [1].
[Cl ]i
=
M Cl
M NaCl
 m

∑  ( P − E ) A + [Cl ]
salt

dep
∑A
i
Ai
[1]
i
Result
In Västmanland the annual net recharge varies between 200 and 300 mm, and the background
deposition contributes to a chloride concentration of 2 mg/l in recharge water. In Figure 5 the average
calculated chloride concentration in recharge is shown for all sub catchment areas of Mälaren within
the county of Västmanland. The amount of deicing salt applied (in tonnes per kilometre and season)
for the different road categories is 12 tonnes for A2, 11 tonnes for A3, 4 tonnes for B1and 1 tonne for
B2. There are no A1 or A4 roads within the county.
5
Winter road maintenance
A2
A3
B1
B2
Chloride conc
0 - 10 mg/l
11 - 20 mg/l
21 - 30 mg/l
31 - 40 mg/l
41 - 50 mg/l
51 - 60 mg/l
N
W
E
S
Figure 5. Average chloride concentration in recharge water in sub catchment areas in Västmanland,
Sweden.
Discussion
The method shows the environmental effect of deicing salt application on a larger scale. The
seasonal variations and the local spatial variations may influence the actual concentration. To calculate
the average steady state concentrations is to simplify a complex event in order to estimate the effects.
Originally the calculations were county based. However, in order to calculate concentrations in a
particular sub catchment area the calculation must begin at the watershed and hence, the calculations
were change to be catchment area based. As mentioned above the calculated chloride concentration is
an average value for the recharge water in every sub catchment area and the chloride concentration in
discharge water for a sub catchment area is a function of the concentration in the recharge in the area
and the concentration in water from areas upstream the basin. In reality the closest representation of
this value would be the chloride concentration in a lake or in a large spring close to the outflow from
the sub catchment area. As a part of the National Monitoring Programme, the SLU University in
Uppsala investigates the status of some lakes every fifth year. In the three most recent investigations
the chloride concentration has been measured. In order to evaluate the method, the calculated chloride
6
concentration in each sub catchment area was compared to the measured chloride concentrations in
lakes close to the outflow from the respective sub catchment area. The calculated chloride
concentration was consistent with the measured values.
Another possible way to appreciate the effects of deicing salt is to start with measured values and
estimate the contribution from other sources. How large is the contribution from road salt compared to
other sources of chloride (e.g. relict salt, saltwater intrusion, sewage, landfills, fertilisers etc). The
presented method only considers deicing salt application and background concentration. Other sources
may be of equal or larger importance locally. However, the greatest impact of deicing salt may be for
areas where the chloride concentration already is increased due to other factors.
In order to estimate the chloride concentration with more precision it is necessary to consider the
direction of the water flow. Upstream a road only fresh water will infiltrate or form runoff.
Immediately downstream the road the chloride concentration in infiltrating water or surface water will
be the highest. The location of the road within the catchment area will be of importance for the
obtained chloride concentration. If the road is located further downstream the dilution effect will be
much greater than if the road is located “higher up” in the system with only a small amount fresh
water contribution.
The contamination will be quite high fore small areas with many major roads. If the area is
“higher up” in the system, the concentration in discharge water from the area will be high. If the area
is closer to the outlet the dilution effect may reduce the concentration to lower levels.
When calculated chloride concentrations are compared with the measured concentrations
especially in urban areas it is important to remember the differences in urban and rural environment.
In the rural environment the road can be considered a line source. In the urban environment the use of
deicing salt is more evenly distributed since not only the Swedish National Road Administration but
also municipal and private property owners use salt for deicing purposes. Hence is it more accurate to
consider the average application in g/mm2 (Howard & Haynes 1993). Furthermore the amount of
paved surfaces and the drainage water system in an urban environment will affect the water
distribution within the catchment area.
Aknowledgement
The project was financed by the Swedish Road Administration through the Centre for Research
and Education in Operation and Maintenance of Infrastructure (CDU) and by the Royal Institute of
Technology (KTH) in Stockholm.
References
Amrhein, C., Mosher, PA., Strong, JE., Pacheo, PG., 1994. Heavy Metals in the Environment. Trace
Metal Solubility in Soils and Waters Receiving Deicing Salts. J. Environ. Qual: 23, 219-227.
Bauske, B. & Goetz, D., 1993. Effects of Deicing-Salts on Heavy Metal Mobility. Acta hydrochim.
hydrobiol. 21, 38-42.
Bäckman, L., Folkeson, L., 1995. The influence of de-icing salt on vegetation, groundwater and soil
along Highways E20 and 48 in Skaraborgs County during 1994. Swedish National Road and
Transport Research Institute (VTI) meddelande Nr 775A.
Howard, K. & Haynes, J., 1993. Groundwater Contamination Due To Road Deicing Chemicals - Salt
Balance Implications. Geoscience Canada, Vol 20, 1-8.
Huling, EE. & Hollocher, TC., 1972. Groundwater contamination by Road Salt: Steady-State
Concentrations in East Central Massachusetts. Science, Vol 176, 288-290.
Norrström, AC & Jacks, G., 1998. Concentration and Fractionation of Heavy Metals in Roadside Soils
Receiving Deicing Salts. The Science of the Total Environment 218, 161-174.
Olofsson, B. & Sandström S., 1998. Increased salinity in private drilled wells in Sweden – Natural or
manmade? In Deicing and Dustbinding – Risk to Aquifers, NHP Report 43, 75-81.
7
Swedish Environmental Protection Agency (SEPA), 2000. Environmental Quality Criteria:
Groundwater. Report 5051.
Thunqvist, E-L., 2000. Pollution of groundwater and surface water by roads, Licentiate thesis,
Division of Land and Water Resources, Royal Institute of Technology, Stockholm, Sweden, ISBN
91-7170-600-3.
Vägverket (Swedish National Road Administration), 1995. Yt- och grundvattenskydd. VV Publ
1995:1. (In Swedish).
8
NON-EXHAUST PARTICLES IN THE ROAD ENVIRONMENT
– A LITERATURE REVIEW
Mats Gustafsson
National Swedish Road and Transport Research Institute (VTI)
SE – 581 95 Linköping
Sweden
E-mail: mats.gustafsson@vti.se
1. Abstract
Non-exhaust particles in the road environment originate from wear of asphalt road pavement,
mainly caused by the use of studded tyres, and corrosion of vehicle components such as tyres and
brakes. Other sources are road maintenance, road equipment and particles originating in the road
surroundings. This literature survey aims at giving an overview of the current knowledge about
airborne particles from these different sources in the context of characteristics and emissions as well as
health and environmental effects.
2. Introduction
Particles related to road traffic have over the last few decades become an important issue among
scientists, the reason being the often-recurrent indications of relationships between airborne particles
and effects on health and the environment. A recently published study shows that particles related to
road traffic are responsible for about 3 % of the total mortality in France, Austria and Switzerland
(Kunzli et al., 2000).
Due to their small size and their ability to adsorb toxic components in the exhaust gas, exhaust
particles have been the main focus for research. In the road environment though, particles from several
other sources occur. Particles generated through wear of vehicles and pavement, added through
maintenance measures or transported to the road environment from the surroundings, make up a large
fraction of road dust. All these particles accumulated on the road surface can be re-suspended by the
action of passing vehicles (fig.1.).
Particles in
suspension above the
road surface
Deposition
Surroundings
Re-suspension
Wear, corrosion, spill
and exhaust particles
Pavement wear
Surroundings
Winter maintenace
(salt and sand, ploughing)
Accumulated road dust
Figure 1. Schematic illustration of sources and fluxes of particles in the road environment.
1
The concentration of particles in the air is usually measured with respect to mass or number per
volume of air. Mass distribution, chemistry and physical properties are also important aspects when
describing the characteristics of particles. When the importance of particle size became apparent a
standardised measure for inhalable particles was constructed called PM10. Roughly, this is the fraction
of particles smaller than 10 µm in diameter. This standard is commonly used throughout the world and
also in the relatively few Swedish cities that measure particles on a regular basis. As interest has
shifted towards even smaller particles, the standard has been supplemented with PM2.5, to measure
particles smaller than 2.5 µm.
This literature survey aims to summarise current knowledge of sources, emissions and health and
environmental effects of non-exhaust particles in the road environment. These are generally larger
than exhaust particles, but a significant fraction occur within PM10 and also PM2.5.
3. Particle sources
Wear particles in the road environment have mainly three sources; tyres, brakes and pavement,
but also incorporate wear from other movable parts in vehicles. For Swedish conditions, pavement
wear during the winter months, when studded tyres are used, is the main source of wear particles.
3.1 Tyres
Depending on quality demands and area of use, tyres consist of a variety of mixtures of rubber
polymers. Latex, as well as synthetic rubber, such as e.g. isoprene rubber, is used to obtain the desired
properties of elasticity, heat resistance and friction (Ahlbom and Duus, 1994). More latex is used in
bus and truck tyres, due to higher friction demands. Rubber mixtures vary greatly between
manufacturers and also between summer, winter and studded tyres, making it difficult to generalise the
chemistry of tyre wear particles (Johansson, 2000).
Tyres also consist of a large number of chemicals added during manufacture. Reinforcing agents,
vulcanisers, accelerators, activators, colour pigments, softeners, dispersing agents, anti-oxidants, antiozonants, stabilisers etc are used during production (Rogge et al., 1993). In literature, tyres are often
mentioned as a main source of zinc in the road environment (Rogge et al., 1993). This is due to the
relatively large amount of zinc oxide used as an activator to make the accelerators more efficient
during manufacture.
Also, PAH (polycyclic aromatic hydrocarbons) occurs in relatively large amounts in tyres. Very
different information on the concentration of substituted and non-substituted PAH in tyres is reported
in literature. (Ahlbom and Duus, 1994) report 7000 µg g-1, while in a Swedish survey, (Lindgren,
1998) has measured between 33 and 93 µg g-1 in three ordinary tyres. Measured in wear particles,
(Takada et al., 1991) reports 31-71 µg g-1.
The size distribution of tyre wear particles is difficult to generalise. (Kobriger and Geinopolos,
1984) and (Noll et al., 1987) claims a mean diameter of 20-25 µm, while (Kumata et al., 1997) reports
a bimodal distribution with peaks at 0,4-0,5 µm and 5-7 µm. These results imply that a considerable
proportion of the wear particles are airborne and inhalable.
Studded tyres, commonly used in Sweden during winter, are made of the same types of rubber
mixtures as other tyres. The studs were initially made of steel, but due to the large wear they caused to
road pavements, they are nowadays made of light metals or plastic. The number of studs and the stud
force have also been decreased (Jacobson and Hornvall, 1999a).
3.2 Brake linings
Similarly to tyres, brake linings are very heterogeneous in composition and manufacturerdependent. The friction materials contain binders, fillers, fibres of glass, plastic, steel, organic or
inorganic material or metals. Metals are also used as heat conductors (Rogge et al., 1993). Table X
shows the metal content of the most common brake linings for cars making up approx. 60% of the
Swedish car fleet (Westerlund, 1998).
An important feature of brake lining particles is their small size. Using a brake dynamometer
(Garg et al., 2000) showed that 63% of the wear particles were smaller than 2.5 µm (PM2.5), i.e.
respirable.
2
3.3 Pavement
The most important wear associated with the use of studded tyres is pavement wear. This depends
on the weight, number and composition of the studs, the flow, composition and speed of the traffic,
climatic conditions, road geometry, pavement composition to mention a few factors. The percentage
and quality of the stone material and the properties of the asphalt itself are of great importance. In
Sweden, high quality pavements with a high percentage of very hard porphyry and quartzite (about
95% in the surface) have gradually replaced pavements with less resistant local rock material on the
heavily trafficked roads (Jacobson and Hornvall, 1999b).
The high cost of maintenance related to pavement wear has caused many countries to prohibit the
use of studded tyres. In Japan and Norway, regulations have also been based upon the health aspects
of the road dust. In Japan studs are prohibited. Before the restrictions, the concentration of airborne
dust could vary between 30 µg m-3 in summer and 400 µg m-3 in winter (Takishima et al., 1987). The
current restrictions are questioned though, since the climate of Hokkaido involves very icy roads,
which has affected the number of traffic accidents negatively (Norem, 1998). In Norway, the “Road
grip project” (Krokeborg, 1997; Larssen and Haugsbakk, 1996) has so far resulted in stud restrictions
in Oslo.
The particles formed through pavement wear reflect the asphalt composition. In Sweden about
95% is rock material and 5% bitumen. This oil product contains asphaltenes (5-25%), saturates (520%), cyclic compounds (45-60%) and resins (15-25%) (Gonzàles Arrojo, 2000). The PAH content is
very small and not considered a main source of PAH in the road environment (Lindgren, 1998). The
size of the pavement wear particles varies, but is generally regarded as being fairly large. According to
(Bækken, 1993) only about 2% are smaller than 36 µm. Japanese studies on the other hand state a size
interval of about 5-50 µm (Amemiya et al., 1984). This particle size depends mainly upon the
properties of the rock material in the pavement, so a large variation is to be expected between
countries and regions.
3.4 Salt and sand
About 200,000 – 400,000 tons of de-icing salt was added to Swedish roads annually between
1991/92 and 1995/96. The salt is re-suspended as wet or dry aerosol by passing vehicles and can be
transported hundreds of meters from the road. The effects of these salt droplets and particles have been
described exhaustively in literature (Blomqvist, 1999), but information about their characteristics is
unfortunately very rare.
3.5 The road as source for PM10 and PM2.5
Many studies make no attempt to distinguish between particle sources, but concentrate on
describing concentrations of the PM10 and/or PM2.5 fractions.
In Norway, before the regulation of studded tyre use, (Larssen, 1987) measured PM10
concentration to 55 µg m-3 in dry conditions and 10 µg m-3 in wet and therefore concluded that road
wear particles contributed 45 µg m-3 in dry conditions. In a later study, (Larssen and Haugsbakk,
1996) found that in dry conditions, the road dust depot does not grow due to a balance between
produced and re-suspended particles. The contribution of road dust to mean annual PM10 and PM2.5
concentrations in Norwegian cities has been shown to be high along roads and streets. The
contribution to highest mean diurnal concentration is significant both in city centres and along roads
and streets (tab. X) (Larssen and Hagen, 1997).
Results from countries where studded tyres are not allowed shows that in spite of this, road dust
often contributes a substantial proportion of PM10 or PM2.5. (Schauer and Cass, 2000) determined the
concentration of road pavement dust to 0.5-1 g m-3 during severe air pollution episodes in California,
USA. Also in the USA, (Noll et al., 1987) found that tyre rubber particles made up approximately
35%, limestone 54% and silicates 10%, both important road building materials, of road dust collected
in business districts in Argonne and Chicago. (Chow et al., 1996) estimated that the road dust
contributed to 25-27% of urban PM10 concentrations as compared to a 30-42% contribution from
exhaust particles.
3
4. Emissions
4.1 Non-exhaust Particles
Emission factors for tyres in literature range from 0.006 to 0.36 g km-1. Using a road simulator
(Rogge et al., 1993) estimated the wear to 0.006-0.09 g km-1 and tyre. Swedish investigations in cooperation with the police authorities, a local traffic company and tyre manufacturers calculated the
emissions to 0.09 g km-1 for a car and 1.0 g km-1 for a bus (Table 1) (Lindström and Rossipal,
1987).
Table 1. Emissions from tyres (Lindström and Rossipal, 1987).
Component
Car
Bus
g km-1
g pbkm-1
Rubber
0,05
0,7
Carbon black
0,03
0,3
Process chemicals,
Activators,
0,011
0,1
Accelerators
Sulphur
0,002
0,02
Total
0,09
1,0
Total emissions from tyres in Sweden has been calculated by (Ahlbom and Duus, 1994), see
Table X). These figures has been criticised by STRO (Scandinavian Tyre and Rim Organisation) who
claim that the calculations do not account for tyre tread and therefore overestimates the emissions
(Johansson, 2000). (Ahlbom and Duus, 1994) also calculated the PAH emissions connected to tyre
wear and concluded that these, 28 µg km-1, are about six times as high as the contribution from
exhaust from a car with a catalytic converter, 5 µg km-1. For Sweden the total amount is 14 tonnes
PAH per year. STRO on the other hand, have calculated the emissions to 284-470 kg y-1.
Table 2. Emissions from tyres (Ahlbom and Duus, 1994)
Component
Total annual emissions in Sweden (tonnes)
Polymeres
5 000
Black carbon
2 500
Oil
2 000
Zinc oxide
150
Stearin acid
70
Sulphur
100
Accelerators
50
Anti oxidants
100
Other
30
Totalt
10 000
In a recently published brake dynamometer investigation (Garg et al., 2000), brake lining
emissions were calculated to 3.2-8.8 mg km-1. Early work by (Cha et al., 1984) supports these figures.
(Westerlund, 1998) estimated the contribution from brake lining wear to metals in the Stockholm
environment and found that about 3,900 kg of copper, 900 kg of zinc, 560 kg of lead as well as a few
kg of chromium and nickel were added each year from cars, buses and trucks. About 80% of the brake
lining wear could be attributed to cars.
Emissions of airborne, inhalable particles from pavements are difficult to handle from literature.
The pavement source contribution is often hidden in terms like “road dust”. Even in Norway, where
great efforts are made to measure PM10 concentrations in i.e. Oslo, the contributions from long range
4
transport and local wood burning disturbs the possibilities to calculate pavement emissions (Larssen,
2000). The pavement wear caused by studded tyres has in Sweden decreased from about 30 g vkm-1 in
the 80ies to about 10 g vkm-1 today. This sums up to about 110 000 tonnes each year.
4.2 Exhaust Particles
The main particle emissions related to vehicle exhaust come from diesel engines. As the attention
paid to the health effects of these particles has increased, the development of cleaner diesel engines
has accelerated. (Lenner and Karlsson, 1998) compiled particle emission figures from 19 different
sources to be used in a quantitative model (Table 3). The figures presented here might therefore be
somewhat out of date due to engine development.
Table 3. Emissions of exhaust particles (mg km-1) (Lenner and Karlsson, 1998). W cat. = without
catalytic converter (figures in brackets are standard deviations).
Car
Heavy truck
W cat.
Cat.
New cat.
Diesel
< 16 t
> 16 t
Bus
16 (2)
2,4 (0,5)
1,4 (0,3)
279 (56)
630 (227)
1080 (430)
830 (274)
5. Health and Environmental Effects
A very extensive literature deals with the relationships between airborne particle concentrations
and public health. This study mainly concentrates on literature that discusses the coarser fraction of
inhalable particles (> 2.5 µm), but some information about smaller fractions has been included for
comparison. Of special importance for human health is the inhalable fraction. Both their ability to
enter more or less deep into the respiratory system and their ability to adsorb toxic substances, such as
PAH and heavy metals, to their surface make them highly interesting to health scientists.
During the late 1990s, American studies have shown that latex particles from tyre wear contain
allergens which might be coupled to an increased risk of latex allergy and asthma (Miguel et al., 1996;
Williams et al., 1995). The studies have their background in the increasing over-sensitivity to latex in
society, the causes of which are not yet clear. (Williams et al., 1995) found that 53% of the latex
particles found in Denver air were inhalable and that their mean size was 6-7 µm. Chemical analyses
suggested the particles originate from tyre wear. (Glovsky et al., 1997; Miguel et al., 1996) extracted
latex allergens from tyre wear particles in Los Angeles and suggested the particles were a potentially
important factor for latex allergy and asthmatic symptoms associated to air pollution.
The health effects of particles related to the use of studded tyres have mainly been studied in
Japan. As early as the mid-1980s, (Morikawa, 1985) related the road dust concentrations to respiratory
symptoms among asthmatic children and (Ikeda et al., 1986) to the frequency of upper respiratory
symptoms. (Watanabe et al., 1990) studied the concentration of elements in the lungs of feral pigeons
and found significantly higher concentrations of Si, Al, Pb and Ti in pigeons living in cities where
studded tyres were used.
Many highly resistant Swedish pavements are based upon rocks containing high concentrations of
quartz (mainly quartzite and porphyry). Quartz dust is well known to induce silicosis among for
example miners, and quartz is regarded as one of the most toxic minerals. In Norway, ongoing
research on PM10 particles from road tunnels shows that particles containing the minerals quartz,
amphibole, chlorite and epidote induced a much higher production of interleukin-6 and –8 in human
lung epitel than did particles containing plagioclase (Hetland et al., 2000). (Murphy et al., 1998)
compared particles of crystalline quartz, amorphous quartz, from diesel exhaust and black carbon and
their impact on rats’ lungs. Somewhat surprisingly, he found more damages from crystalline quartz
and amorphous quartz no effects from diesel or black carbon particles. This should imply a surface
structure or chemical effect. As opposed to this, many studies rather imply a particle size effect, i.e.
the chemistry or structure is not important (Camner, 2000).
Epidemiological studies relate inhalable particle concentrations to mortality, morbidity, lung
cancer, asthma, respiratory symptoms and coughs, usually in urban areas and to a greater extent
among sensitive populations such as children, asthmatics and elderly people. Extensive compilations
of current knowledge has been made by e.g. (Vedal, 1997) and (Areskoug, 2000).
5
Commonly, literature implies that fine and ultra-fine particles (< 1 µm) show stronger relationships to
health effects than do coarser fractions. The increase in mortality is generally 0.5-1.0% per 10 µg m-3
increase in PM10 concentration. Hospital admissions due to short term exposure increases by 0.5-3.0%,
which confirms a relationship to particle concentration. The relationship is often stronger for
symptoms in the lower respiratory tract than in the upper and also stronger for elderly people but also
for children (Areskoug, 2000).
Despite the fact that most studies stress the importance of the fine fractions, quite a large number
of exceptions exist. (Pekkanen et al., 1997) found that ultra-fine particles were not more strongly
related to variations in peak expiratory flow rate (PEFR) than were PM10 or black smoke particles. A
study made in Cochella valley in USA where coarse particles with a geologic origin contribute to a
very large fraction of PM10 show that PM10 was significantly associated to all used measures of
mortality (Ostro et al., 1999). On the other hand, (Schwartz et al., 1999) saw no such signs in a similar
study. In Mexico City, (Castillejos et al., 2000) found that PM10-2.5 had a stronger effect on mortality
(4.07% associated with a 10 µg m-3 concentration increase) than PM10 (1.83%) and PM2.5 (1.48%). A
probable explanation might be the presence of biogenetic material in the PM10-2.5 fraction.
Except for the health effects, non-exhaust particles affect public comfort by dirtying cars,
sidewalks, house fronts, windows and even insides of houses.
Literature on the environmental effects of road dust particles as such is scarce. Most of it deals
with the road as a source of pollution for the roadside environment (Bækken, 1993; Bækken and
Jörgensen, 1994; Bjelkås and Lindmark, 1994; Gjessing et al., 1984; Kobriger and Geinopolos, 1984;
Lygren and Gjessing, 1984; Sansalone et al., 1995). Particles from pavement wear contribute to the
structure and composition of the roadside soils. The accumulation of material might amount to as
much as 1.5 cm y-1. Roadside soils diverge strongly from adjacent soils both regarding size
distribution and chemical properties. High pH, high content of base cat-ions and heavy metals are
characteristic features of these soils (Bækken, 1993).
Some studies concern the effects of particles on vegetation. Particles on the surface of leaves and
needles have been shown to cause stress and therefore reduced growth due to increased temperature,
blocked stomata and the hygroscopic properties of some particles (Farmer, 1993; Flückiger et al.,
1978). Effects on limnic systems include high, and sometimes toxic, concentrations of PAH and heavy
metals in lake sediments (Bækken and Jörgensen, 1994), but also first flush effects, where the particle
depot accumulated on a road causes high concentrations of toxic compounds in streams during
rainfall.
De-icing salt, which can be transported to the roadside environment as an aerosol or as dry dust,
also affects vegetation negatively, which is a very visible problem in Sweden as Norway spruce and
Scots pine along salted roads often turn brownish in spring due to the salt (Blomqvist, 2001). Salt
deposited on leaves and needles causes osmotic stress causing dessication. The salt has also been
shown to accumulate in ground water reservoirs with a hydrologic connection to road environments
(Thunqvist, 2000).
6. Discussion
The sources for non-exhaust particles, judging from this literature survey, are many and their
interplay complicated. Information about emissions and characteristics of more diffuse sources, like
corrosion and biogenic material deposited on the road, has not been found in literature.
A critical report on tyre wear in Sweden (Ahlbom and Duus, 1994) has caused a debate, where
the STRO claim the report to be incorrect and exaggerated in many respects. Nevertheless the debate
has led many tyre manufactures to develop winter tyres without HA-oils, which is used as an argument
in commercials. In summer tyres, the HA oils are more difficult to omit, since they are responsible for
a large part of the grip properties (Johansson, 2000). Very varying information is reported about the
properties of tyre particles. Both small quantities of inhalable particles as well as rather large
proportions of respirable particles are reported, which might be a result of the large variation in
materials, wear conditions and measurement methods.
Studded tyres and the related pavement wear are the particle source most thoroughly investigated
in conditions similar to those in Sweden. In our neighbouring country, Norway, studded tyres have
6
been the subject of much debate and major research for two decades. The focus has been on health
effects, and an economic investigation during the “Veggreppsprosjektet” (Road grip project) showed
that restrictions in the four largest cities in Norway were profitable to society (Krokeborg, 1997). So
far, only Oslo has introduced the restrictions since 1999. Studded tyre use is now about 30% as
compared to 70% before the restrictions. Due to large variations in seasonal weather it is too early yet
to determine any effect on PM10 concentrations (Hagen and Haugsbakk, 2000).
There are a few studies on brake linings and their contribution to pollution. Most of these show
that particulate heavy metals are the main cause of concern. An important aspect from a health point of
view is that brake lining particles are very small and therefore potentially more dangerous to health.
For Swedish conditions the three most important emission sources for non-exhaust particles are,
in order of magnitude, pavement wear (about 110,000 tonnes), tyre wear (about 10,000 tonnes) and
brake lining wear (about 1,000 tonnes). The total amount of these particles emitted during a year is in
the same order of magnitude as that of exhaust particles, but seasonal effects of climate, local sources,
maintenance actions etc make the emissions very uneven both temporally and spatially.
Health surveys dealing with particle effects are common and usually based upon measurement of
PM10 and/or PM2.5, which are measures produced for this specific purpose. The results of these studies
show a rather scattered picture of how PM relates to toxicological or epidemiological effects. There is
a consensus that particle size matters and that there is a specific particle effect, but what this effect
really involves is still not clear. A part of the problem definitely lies in the PM methodology. PM10 or
PM2.5 says nothing about size distribution below 10 or 2.5 µm and nothing about chemical or physical
properties, i.e. surface area, of the particles. According to (Camner, 2000), there is a gap between
toxicology and epidemiology, since the effects shown in epidemiological studies, with effects on
populations at rather low PM concentrations can not be verified in toxicological experiments where it
takes much higher concentrations to cause the same medical symptoms. Regarding environmental
effects of particles, these are seldom related to particles as such, but rather to pollution from PAH or
heavy metals.
7. Conclusions and research needs
The international literature on non-exhaust particles and their effects is quite extensive. The
information about particle emissions and particle characteristics displays a very large variation
though, depending on investigation quality, methods and extent as well as geographical variations. The
material is often based on short-term measurements seldom valid for other geographic locations or
during other time intervals. For Swedish conditions, the following approximations might be
considered:
Wear
Total in Sweden
• Pavement
• Tyre
• Brake linings
Emissions
•
•
Pavement wear
Tyre
•
•
•
Brake lining wear
Re-suspension
Exhaust
110 000 t y-1
10 000 t y-1
1 000 t y-1
from studded tyre use <10 g km-1
car 0,006 - 0,36 g km-1
truck, bus approx. 1 g km-1
0,0032 - 0.0088 g km-1
0,13 - 6 g km-1
car with catalytic converter 0.0014 - 0.0024
g km-1
car (diesel) 0,279 g km-1
heavy truck (>16 tonnes, diesel) 1,08 g km-1
7
Health effects
Toxicological aspects
• Tyres
•
Pavement wear
•
Brake lining wear
allergy and asthma from latex particles?
relatively high PAH content
relatively large percentage of PM10
important source for PM10 in the road environment
mineral composition and surface properties?
large percentage of PM10
especially important in cities?
Epidemiological aspects
• Few other measures than PM10 and PM2.5 are studied
• Wear particles mainly in the coarse fraction PM10-2.5
• Many studies show a higher correlation of health effects to PM2.5 but there are also
studies indicating higher correlation to PM10-2.5
• Not only a particle effect. Chemistry and surface properties probably also
important.
To be able to thoroughly investigate the relations between non-exhaust, as well as exhaust-,
particles and health- and environmental effects it is essential to complete PM measurements with
measurements giving more information on chemical characteristics, particle size distribution and
maybe also surface characteristics. Characterisation should be made on source specific particles to be
able to make reliable source apportionments in field measurements as well as risk assessments. More
field data covering a greater temporal and spatial variation are needed to improve knowledge about
variations in particle concentration, composition and characteristics.
At VTI (National Swedish Road and Transport Research Institute) efforts to characterise
particles from pavement and tyre wear using the VTI road simulator are being made. The road
simulator offers the possibility to study “pure” wear particles since it is situated inside a building.
Ongoing projects deal with emission factors for inhalable wear and re-suspension particles to be used
in emission models and an inventory of road cleaning methods effective for particle removal. Future
particle research efforts at VTI are planned to include spatial and temporal variations and model
validation studies in the field.
8
References
Ahlbom, J. & Duus, U.; Nya hjulspår - en produktstudie av gummidäck. 6/94, Kemikalieinspektionen, 1994.
Amemiya, S., Tsurita, Y., Masuda, T., Asawa, A., Tanaka, K. & Katoh, T.; Investigation of environmental problems caused
by studded tires of automobiles using PIXE. Nuclear Instruments and Methods in Physics Research, B3: 516-521,1984.
Areskoug, H.; Particles in the ambient atmosphere. Scandinavian Journal of Work, Environment and Health, 26 Suppl 1: 522,2000.
Bækken, T.; Miljøvirkninger av vegtrafikkens asfalt og dekkslitasje. O-92090, NIVA, 1993.
Bækken, T. & Jörgensen, T.; Vannförurensninger fra veg - langtidseffekter. 73, Statens vegvesen, 1994.
Bjelkås, J. & Lindmark, P.; Föroreningar av mark och vägdagvatten på grund av trafik. 420, SGI, Linköping, 1994.
Blomqvist, G.; Air-borne transport of de-icing salt and damage to pine and spruce trees in a roadside environment. Licentiate
Thesis, Royal Institute of Technology, Stockholm, 1999.
Blomqvist, G.; De-icing salt and the roadside environment: Air-borne exposure, damage to Norway spruce and system
monitoring. Doctoral Thesis, Royal Institute of Technology, Stockholm, 2001.
Camner, P.; pers. com., Institute of Envrionmental Medicin, Karolinska institutet, 2000.
Castillejos, M., Borja-Aburto, V.H., Dockery, D.W., Gold, D.R. & Loomis, D.; Airborne coarse particles and mortality.
Inhalation Toxicology, 12 (Supplement 1): 61-72,2000.
Cha, S., Carter, P. & Bradow, R.L.; Simulation of automobile brake wear dynamics and estimation of emissions. Society of
Automotive Engineering , Inc., Technical paper: 3502-3522,1984.
Chow, J.C., Watson, J.G., Lowenthal, D.H. & Countess, R.J.; Sources and chemistry of PM10 aerosol in Santa Barbera
County, Ca. Atmospheric Environment, 30(9): 1489-1499,1996.
Farmer, A.M.; The Effects of Dust on Vegetation - A Review. Environmental Pollution, 79: 63-75,1993.
Flückiger, W., Flückiger-Keller, H. & Oertli, J.J.; Der Einfluss von Strassenstaub auf den stomatären Diffusionswiederstand
und die Blatt-Temperatur - ein antagonistischer Effekt. Staub - Reinhalt. Luft, 38(12): 502-505,1978.
Garg, B.D., Cadle, S.H., Mulawa, P.A., Groblicki, P.J., Laroo, C. & Parr, G.A.; Brake wear particulate matter emissions.
Environmental Science and Technology, 34(21): 4463-4469,2000.
Gjessing, E., Lygren, E., Berglind, L., Gulbrandsen, T. & Skaane, R.; Effect of highway runoff on lake water quality. The
Science of the Total Environment, 33: 245-257,1984.
Glovsky, M.M., Miguel, A.G. & Cass, G.R.; Particulate air pollution: possible relevance in asthma. Allergy Asthma Proc,
18(3): 163-6,1997.
Gonzàles Arrojo, M.; Pavement wear caused by the use of studded tyres. 6A-2000, VTI, Linköping, 2000.
Hagen, L.O. & Haugsbakk, I.; Måling av luftkvalitet ved to sterkt trafikkerte veier i Oslo vinteren 1999/2000. OR 32/2000,
NILU, 2000.
Hetland, R.B., Refsnes, M., Myran, T., Johansen, B.V., Uthus, N. & Schwarze, P.E.; Mineral and/or metal content as critical
determinants of particle-induced release of IL-6 and IL-8 from A549 cells. Journal of Toxicology and Environmental
Health, Part A, 60: 47-65,2000.
Ikeda, M., Watanabe, T., Hishamichi, S., Shimizu, H., Fujisaku, S., Ichinowatari, Y., Ida, Y., Suda, M. & Kato, K.; Upper
respiratory symptoms presumably due to studded tire-generated dust. Environment International, 12: 505-511,1986.
Jacobson, T. & Hornvall, F.; Dubbslitage på asfaltbeläggning. VTI meddelande 862, VTI, Linköping, 1999a.
Jacobson, T. & Hornvall, F.; Polering av asfaltbeläggning: friktionsmätningar 1997-1998. VTI notat 6-1999, VTI, 1999b.
Johansson, T.; pers. com. Scandinavian Tyre and Rim Organisation, 2000.
Kobriger, N.P. & Geinopolos, A.; Volume III. Sources and migration of highway runoff pollutants - research report.
FHWA/RD-84/059, Federal Highway Administration, Washington, DC, 1984.
Krokeborg, J.; Veg-grepsprosjektet Delprosjekt 5.4: Vegstøv - helseskader og -kostander, Økonomiske konsekvenser av
endret piggdekkbruk, helse og trivsel. 1990, Statens vegvesen, Vegdirektoratet, 1997.
Kumata, H., Takada, H. & Ogura, N.; 2-(4-Morpholinyl)benzothiazole as an Indicator of Tire-Wear Particles and Road Dust
in the Urban Environment. In: R.P. Eganhouse (Editor), Molecular Markers in Environmental Geochemistry. American
Chem. cop., Washington, D.C., pp. 291-305, 1997.
Kunzli, N. et al.; Public-health impact of outdoor and traffic-related air pollution: a European assessment. Lancet, 356(9232):
795-801,2000.
Larssen, S.; Støv fra asfaltveier, karakterisering av luftbåret veistøv. OR 53/87, NILU, 1987.
Larssen, S.; pers. com. NILU, 2000.
Larssen, S. & Hagen, L.O.; Partikkelforurensning fra piggdekk. OR 16/97, NILU, 1997.
Larssen, S. & Haugsbakk, I.; Veg-grepsprosjektet, Delprosjekt 5.3: Vegstøv - sammansetning og effekt på fysisk miljø.
Vejstøvdepot og svevestøvkonsentrasjon. Intern rapport nr. 1911, Statens vegvesen, Vegdirektoratet, Veglaboratoriet,
1996.
Lenner, M. & Karlsson, B.O.; Kvantitativ beräkningsmodell för trafikens utsläpp av cancerframkallande ämnen i svenska
tätorter. VTI meddelande 847, Väg- och transportforskningsinstitutet, Linköping, 1998.
Lindgren, Å.; Road Construction Materials as a Source of Pollutants. Doctoral Thesis, Luleå University of Technology,
Luleå, 129 pp, 1998.
Lindström, A. & Rossipal, B.; Emissioner från landsvägs- och järnvägstrafik. Trita-Kut 3039, Dept. of Land Improvement
and Drainage, Stockholm, 1987.
Lygren, E. & Gjessing, E.; Highway pollution in a Nordic climate. 0-79024, NIVA, 1984.
Miguel, A.G., Cass, G.R., Weiss, J. & Glovsky, M.M.; Latex allergens in tire dust and airborne particles. Environ Health
Perspect, 104(11): 1180-6,1996.
9
Morikawa, T.; The influence of road dust on asthmatic children. Japanese Journal of Allergology, 34: 297-304,1985.
Murphy, S., BéruBé, K.A., Pooley, F.D. & Richards, R.; The response of lung epithelium to well characterised fine particles.
Life Sciences, 62(19): 1789-1799,1998.
Noll, K.E., Draftz, R. & Fang, K.Y.P.; The composition of atmospheric coarse particles at an urban and non-urban site.
Atmospheric Environment, 21(12): 2717-2721,1987.
Norem, H.; Høy pris for forbud mot piggdekk, Våre Veger, pp. 26-28, 1998.
Ostro, B.D., Hurley, S. & Lipsett, M.J.; Air pollution and daily mortality in the Coachella Valley, California: a study of
PM10 dominated by coarse particles. Environmental Research, 81(3): 231-8,1999.
Pekkanen, J., Timonen, K.L., Ruuskanen, J., Reponen, A. & Mirme, A.; Effects of ultrafine and fine particles in urban air on
peak expiratory flow among children with asthmatic symptoms. Environ Res, 74(1): 24-33,1997.
Rogge, W.F., Hildemann, L.M., Mazurek, M.A. & Cass, G.R.; Sources of fine organic aerosol. 3. Road dust, tire debris and
organometallic brake linig dust: Roads as sources and sinks. Environmental Science and Technology, 27: 18921904,1993.
Sansalone, J.J., Buchberger, S.G. & Koechling, M.T.; Correlations Between Heavy Metals and Suspended Solids in Highway
Runoff: Implications for Control Strategies. Transportation research record, 1483: 112-119,1995.
Schauer, J.J. & Cass, G.R.; Source apportionmnet of wintertime gas-phase and particle-phase air pollutants using organic
compounds as tracers. Environmental Science and Technology, 34: 1821-1832,2000.
Schwartz, J., Norris, G., Larson, T., Sheppard, L., Claiborne, C. & Koenig, J.; Episodes of high coarse particle concentrations
are not associated with increased mortality. Environmental Health Perspectives, 107(5): 339-42,1999.
Takada, H., Onda, T., Harada, M. & Ogura, N.; Distribution and sources of polycyclic aromatic aromatic hydrocarbons
(PAHs) in street dust from the Tokyo Metropolitan area. The Science of the Total Environment, 107: 45-69,1991.
Takishima, T., Nakamura, M., Sasaki, M., Miyano, M., Yamaya, M. & Sasaki, H.; Inhalation of road dust by human subjects.
American Review of Respiratory Disease, 136: 1278-1280,1987.
Thunqvist, E.-L.; Pollution of groundwater and surface water by roads - with emphasis on the use of deicing salt. Licentiate
Thesis, Royal Institute of Technology, Stockholm, 2000.
Watanabe, T., Nakatsuta, H., Kasahara, M. & Ikeda, M.; Silica and aluminum in the lungs of feral pigeons and experimental
rats exposed to studded tire-generated road dust. Environmental International, 16: 23-29,1990.
Vedal, S.; Ambient particles and health: lines that divide. J. Air & Waste Manage. Assoc., 47: 551-581,1997.
Westerlund, K.-G.; Metallemission från trafiken i Stockholm - slitage av bromsbelägg. 2:98, Environment and Health
Protection Admisnistration, Stockholms Luft- och Bulleranalys, Stockholm, 1998.
Williams, P.B., Buhr, M.P., Weber, R.W., Volz, M.A., Koepke, J.W. & Selner, J.C.; Latex allergen in respirable particulate
air pollution. J Allergy Clin Immunol, 95(1 Pt 1): 88-95,1995.
10
WINTER INDEX BY USING RWIS AND MESAN
an Operational Mesoscale Analysis System
Jan ÖLANDER, Sweden
Head of Strategies and Planning - Winter Road Maintenance
National Road Management Department
Swedish National Road Administration (SNRA)
SE-781 87 Borlänge, Sweden
TEL. +46 243 752 23 / FAX. +46 243 754 20
E-mail address: jan.olander@vv.se
1. Abstract
In today’s road administrations it is very important to keep track of costs and the amount of
chemicals and abrasives used in winter road maintenance. Variations in snow and ice conditions from
year to year make it difficult to compare figures. In order to tackle this problem, the Swedish National
Road Administration (SNRA) has developed a new Winter Index.
The Winter Index is based on data from our 680 Road Weather Information System (RWIS)
stations as well as data provided by the Swedish Meteorological and Hydrological Institute (SMHI).
Basically, we use the former for information on the air and road surface temperatures, humidity, wind
and type of precipitation and SMHI data for the amount of precipitation. The RWIS data is collected
every half-hour and the SMHI data every third hour. To calculate the amount of precipitation, SMHI
uses a model called Mesan, which is an operational Mesoscale Analysis System. This model subdivides Sweden into a 22 by 22 kilometre grid net, and calculations are performed for each grid
individually.
The system provides data on slippery roads, snowfall and snowdrifts expressed in number of
occasions; e.g., two icy road surface occasions will be registered if it is known that a skid control
measure will be effective for 5 hours, and the RWIS data shows that there still is a risk of slippery
roads after 6 hours. The same principle applies to snowfalls and snowdrifts. The system can detect
four kinds of slippery surfaces ranging from light frost to freezing rain, and three kinds of snowfall
and snowdrift, from light to heavy.
The amount of material used and the cost involved is entered in the final step of the calculation.
This provides a good basis for comparing salt consumption from year to year. When calculating the
Salt Index in this model, the length of road treated with salt, type of road (standard class) and our
“Guidelines for Salt” are used. A Salt Index of 1.0 indicates that the contractor (or county, regional, or
national road manager) has used the optimum salt dosage.
The system is also a good tool for benchmarking costs.
2. Background
When following up winter road maintenance, an important parameter
for being able to compare costs and material consumption is knowing what
the winter was like compared to other winters. Formerly, salt consumption
was recorded in terms of tonnes per kilometre, and costs in terms of SEK
per kilometre. Needless to say, figures varied from winter to winter, but it
was difficult to determine exactly why. Previously, we used SMHI weather
statistics in Sweden, but these statistics are based on atmospheric
measurements, and as such are not directly connected to road climatology.
Typical parameters obtained through this type of follow-up are days of
snowfall and days of predicted hoar frost formation.
The SNRA owns 681 RWIS stations located throughout the country.
An original idea was to use the data that has been stored in these stations
1
over the years. There would have been two advantages to this: firstly, since RWIS data is used by
contractors as a trigger factor for initiating action, road maintenance measures and follow-up
parameters coincide well. Further, no additional cost is involved since this data is the property of the
SNRA.
However, what we want at the SNRA are parameters that reflect the predicted number of times
when snow ploughing or skid control action is needed. We have therefore chosen follow-up
parameters to show this. The new Winter Index can be used both for cost and salt consumption
follow-up, and as a payment basis for contracted winter road maintenance. The procedure for paying
contractors is described in paper no. 22 under topic #1, which also gives a more detailed description
of how the Winter Index is calculated.
3. Bases and theories behind the Winter Index
As mentioned earlier, we want an index that reflects reality, i.e., one that is related to the
measures actually carried out. Hence, every weather situation that would entail winter maintenance
action has been divided into time periods. The weather situations studied are snowdrifts, snowfall and
the risk of slippery road surfaces (hoar frost formation). This is done in hierarchical order. In other
words, the system first searches for snowdrifts. In the absence of such, it searches for occasions of
snowfall. If neither of these situations is found, the system then searches for occasions of slippery
road surface conditions.
Snowdrifts are divided into four categories (d = snow depth in cm)
0.0 ≤ d ≤ 0.3
0.3 < d ≤ 1.0
1.0 < d ≤ 2.5
2.5 < d.
Further, the duration of a measure is calculated as 4 hours; i.e., if a snowdrift lasts between 0.5
and 4 hours, it is counted as one occasion, and as two if it lasts between 4 and 8 hours, as three if it
lasts between 8 and 12, etc.
Snowfall is divided into three categories (d = snow depth in cm)
0.3 < d ≤ 1.0
1.0 < d ≤ 2.5
2.5 < d.
Four hours is the presumed duration of a measure here as well, and the calculation procedure for
the number of occasions is the same as for snowdrifts.
There are four different categories of slippery surface.
Slippery surface due to rain or sleet on a cold road (HN).
Slippery surface due to damp/wet roads freezing over (HT).
Slippery surface due to light frost (HR1).
Slippery surface due to heavy frost (HR2).
A measure is presumed to last between 3 and 6 hours, depending on the category of slipperiness
and the time of year.
4. Calculating occasions of snowfall conditions
Snowfall is divided into 4-hour periods. The reason behind this is that our winter road
maintenance specifications stipulate different action completion times for different types of road
(standard classes), and 4 hours was considered to be a suitable average. The division into different
snow depth categories also complies with the specifications.
2
To exemplify this, class 0.3 < d ≤ 1.0 signifies a snowfall where it is expected that it will only be
necessary to use salt to melt the snow, whereas class 2.5 < d signifies snowploughing on the low
volume traffic network.
5. Calculating occasions of slippery road conditions
Occasions of slippery surface conditions are those where RWIS data would theoretically mean
hoar frost formation. This means that the figures are directly related to road climatology and not
atmospheric measurements, which is the usual way of measuring.
6. Weather presentation
Weather situations can be presented in tables or as graphs at a resolution that shows either the
entire country, each winter maintenance region or each production and maintenance district. The
following graphs are examples of weather situations in Sweden during the winters between 1996/97
and 2000/2001.
Diagram 1, Total number of weather situations requiring action in the entire country per season
Diagram 2, Snow occasions per region and season
Diagram 3, Slippery surface occasions per region and
season
3
7. Presentation of the salt index
The salt index is calculated on the basis of the weather data above, taking into consideration the
length of the road network treated with salt and the recommended salt dosage for different weather
conditions. The salt dosage values used in the calculation model are based on the SNRA’s Guidelines
for De-Icing, which is an official document that describes different salting methods and dosages based
on different conditions. For instance, a salt solution shall be used in connection with light frost (HR1).
During other weather situations, pre-wetted salt is permitted. The table below shows the
recommended salt dosage in grams per metre 2-lane road for different weather situations.
Snowfall
amount of snow (cm)
SNOW1
SNOW2
SNOW3
0.31-1.00 1.01-2.50
2.5136
90
120
Type of slipperiness
HR1
HR2
HT
HN
24
36
48
60
Table 1. Recommended salt dosage in grams per metre 2-lane road during different weather situations.
Based on this, the salt index is calculated as follows:
⇒ salt consumption during the period for the area in question.
⇒ km of road treated with salt. For motorways and 4-lane roads, both directions are counted.
⇒ number of HR1, HR2, HT, HN, snowfall and snowdrift occasions as per the following table
for the area in question.


 ∑ Salt consumptio n, kg 




Length
of
road
salted,
km
 ∑


Salt index = 
 ( HR1 * 24) + (HR2 * 36) + (HT * 48) + ( HN * 60) + (SNOW1 * 36) + (SNOW2 * 90) + (SNOW3 * 120) 




The salt index can be presented in tables or as graphs at a resolution that shows either the entire
country, each winter maintenance region or each production and maintenance district. The following
graphs are examples of weather situations in Sweden during the winters between 1996/97 and
2000/2001.
4
Diagram 4. Salt index for the entire country
Diagram 5. Salt index per region
Diagram 6. Predicted number of occasions for using a salt
solution or pre-wetted salt
Diagram 6 is intended to show that it probably is also possible to follow up how the contractor
has complied with the recommendations concerning the use of different methods for different
conditions. Since more and more skid control vehicles are being equipped with GPS, which also
records salt dosage and method, this follow-up can probably be done automatically in the future.
8. The future
In preparation for the coming winter, 2001 – 2002, further testing will be done to determine if
we have used the right parameters in our evaluation. There is a some uncertainty concerning the
figures in the SNRA’s Guidelines for De-Icing.
Moreover, we will be developing a similar system to follow up costs for winter road
maintenance. At present, 100% of winter road maintenance is contracted in open competition, and the
SNRA does not have any ploughs or skid control equipment of its own. This would appear to make it
even more important today to follow economic trends.
5
Considering today’s demands on environmentally sound and rational road management, it is
essential to have follow-up tools that indicate whether or not we are making improvement. This is
important for ecologically sound development (it is not only in Sweden that the use of salt on roads
has come into question) as well as to show our principal – the Ministry of Industry Employment and
Communication – as well as road users that we are taking these matters seriously.
6
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising