PROCEEDINGS OF THE 13TH SIRWEC CONFERENCE, TORINO, ITALY (25-27TH MAY 2006)
TOPIC 1: RECORDING AND EVALUATION OF ROAD WEATHER DATA
Measurements with a mobile radiation balance station and simulations with the model system
Alpine3D have been carried out. Two factors were investigated in detail. These are the road
albedo values depending on the surface condition and different horizontal resolutions for the
simulations. We found differences in summer road surface temperatures that can not be
neglected. We expect these differences to be even higher during winter time.
provide information on surface conditions on the pavement surface in addition to standard
weather information such as air temperature, dew point, wind speed, etc. The classic
approach is to install a limited number of RWIS at critical points. In this project a different
method is proposed and studied in order to allow preventive winter maintenance without
installing numerous RWIS. This is achieved by means of a system able to monitor the current
weather conditions directly on the maintenance vehicles. Different systems are already
available for those highway maintenance vehicles. However those systems are either to
expansive to allow for installing them on many vehicles or they are not reliable and accurate
enough for a correct preventive maintenance planning.
spectroscopy and short distance remote sensing. Since the sensor can provide amount of water, ice and
white ice independently of each other, we tested also the feasibility to model directly slipperiness
caused by winter weather on road surfaces. In turned out that it is possible to obtain a fair correlation
between actually measured and modeled friction values in typical winter weather conditions. The field
testing included also a remote infrared surface temperature sensor. Impact of the technology on road
weather information systems is discussed.
mixed with rock salt. Tests have shown that molasses based de-icing products have a unique fluorescence
signal, the peak intensity of which excites and emits at 340 and 420 nm respectively and is unaffected by
changes in temperature. The fluorescence signal is easily identifiable at temperatures as low as -8°C, and the
intensity of the signal has been found to be reasonably constant over the temperature range 0 to 5°C, which is
most associated with the marginal nights in the UK where there is a chance a road surface may freeze. Within
this temperature range the intensity of the fluorescence signal could be used to quantify salt concentrations.
Remote sensing tests of the same fluorescence signal using a fibre optic probe have revealed difficulties in
applying such a technique in a road environment. The potential applications of this research to road weather and
other scientific fields are discussed.
required so as to initiate an appropriate response to the detected environmental conditions. In
particular, any information on road sections having ice or moisture formed thereon may help
to significantly improve traffic safety. These surface data are typically used to forecast
surface conditions and anti-icing. Thus, one of the key components of modern road
information systems (RIS) is the monitoring of the current road surface conditions, wherein
the detection of the presence of water, snow, ice and the beginning formation of ice, is one of
the most important pieces of information for traffic safety. We provide a new ice detection
and forecasting system based on a new innovative ice sensor in order to allow surface
monitoring, with enhanced accuracy and reliability, especially with regard to water, snow,
and ice detection on exposed surfaces and the detection of ice formation at an early stage.
the urban air pollution is an increasing problem. In particular the previsions of NOx, as an
indicator of motor vehicle exhaust, is a useful tool to monitor pollution levels; it allows local
authorities to manage the urban traffic to prevent the air pollution concentration values that
are dangerous for human health. In the cities with heavy vehicular traffic it is possible to
have high concentrations of air pollutants. This model can be useful for administrations to
rationalise the preventive intervention to reduce the occurrence of high pollutants
concentrations and to avoid useless annoyances to people. In this paper, the authors
developed a recurrent neural network (RNN) model to forecast the maximum daily nitrogen
oxides (NOx) concentrations in Palermo. The authors have compared the neural network
results with those obtained from a stochastic model (AR2). The RNN model is different from
the statistic models because it utilizes a pattern recognition approach. The research is based
upon collected data from six monitoring sites during the period 2003-2004. The model
developed is a potential tool for the predictions air quality parameters and it is superior to the
traditional stochastic model.
snowfall, lasting for hours, caused traffic to be blocked and a lot of people to spend nearly
twenty four hours in a dreadful scenery of snowflakes falling from the sky, surrounded by
dark mountains, and suffering from the hunger, the thirst and the cold. In section 2 an
overview of the general geographic characteristics of Italian Apennines is given. In section 3
and 4 the Italian general forecasts and highway road weather network is described. At the
end there is an analysis of the meteorological data collected by the road weather stations
during this extreme event.
TOPIC 2: FORECAST METHODS AND ACCURACY
A technique is developed for diagnosing different types of ice, frost and rime phenomena
based on the classification of processes causing ice generation on the road. Based on this
technique, an algorithm to forecast and diagnose glazed frost on the road has been worked out,
which is realized as a model program demonstrating the potentials of the technique involved.
According to this technique, glazed frost on the road is forecasted in two steps. As a first step,
the road temperature is forecasted with sufficient lead-time. Secondly, based on the forecasted
road temperature values and some other parameters, a diagnostic analysis and forecast of
glazed frost events is performed.
The model program is run with a certain set of sensors available at an automated road weather
station (RWS) and meant for automated information systems to prevent glazed frost by using
A technical problem has been solved, enabling data scanning in an interface suitable for
examination, and software has been developed for personal computer to distinguish the
corresponding types of events in the course of road weather station measurements.
The verification of the model during two winter seasons at two automated road weather
stations has demonstrated satisfactory results. The relative faultless diagnoses of the events
concerned amounted to 83% for the winter of 1996/1997 and 79% for the winter of 1997/1998.
With wintertime comes the potential for icy, slippery roadways, which result in decreasing
the level of traffic performance on roadways. Road authorities regularly take anti-icing and –
slippery procedures to prevent snow or ice from bonding to road surface. However, as the
amount of those uses tends to be increased from year to year, it requires more efficient snow
and ice control with the proactive preventive approach being desirable to apply early enough
to ice from forming. Thus, the more-efficient works needs developing an appropriate method
to predict road surface icing conditions. Our research group works in developing a method
for surface-icing forecast with applying a heat balance model, with the intention of providing
Strategic Snow and Ice Control which takes the proactive winter maintenance strategy.
This paper describes the research framework and attempts to demonstrate how the heatbalance model developed here works to predict road surface temperature.
firstly meteorological conditions, such as temperature, wind speed and direction, short and
long wave radiation and amount of snowfall is calculated by regional weather model with a
fine space resolution. Secondly the change in the property of snow cover is obtained with
SNOWPACK model developed by Swiss Federal Institute for Snow and Avalanche Research
(SLF). Thirdly, based on the above two steps, forecasting of the snow avalanche danger, the
decrease in the visibility due to a blowing snow, and the condition of snow and ice along the
road were carried out.
Development of wind alarm systems for road and rail vehicles: Presentation of the WEATHER project
D. Delaunay, C.J. Baker, F. Cheli, H. Morvan, L. Berger, M. Casazza, C. Gomez, C. Le Cleach, R. Saffell, R. Grégoire, A. Viñuales
Abstract | PDF
presented. The alarm is funded on a risk assessment approach, taking into account wind
modelling and prediction, aerodynamic forces, vehicle dynamics.
incorporation and assimilation of cloud observations from satellites and conventional data are
discussed. Both conventional observations and NOAA and MSG-1 satellite data were
assimilated using a nudging technique and the results were verified for a two weeks period.
Additionally the derived cloud mask from the NOAA satellite was compared with
conventional observations. The results indicate that use of cloud data can be used to improve
forecasts of the road surface temperature in particularly for short term forecasts. It is also
emphasized that NOAA data for some days differ from conventional data and should be used
with care in certain situations.
EuroLM). The main characteristics of the model are 7 km of horizontal resolution, a domain
size of 63648 points, 35 level for vertical resolution and two runs at 00 and 12 UTC.
Apart from the forecasters daily experience in using this model as a modern tool for
forecasting the usual weather parameters for short and very short range forecasts, different
important applications of the model products have been carried out in these years. High
resolution metgrams, vertical profiles, an Automatic Weather interpretation tools (AWI), are
only a few among the most important post-processing outputs. The experience in using high
resolution product for punctual specialised forecasts tuned out for the International
International Cycling Race 88th
Giro dItalia for a month over the entire route, is described
in this presentation, along with a similar one for Winter Olympic Games. Based on that
experience an experiment test concerning some of the main legs of Italian Motorway
Network has started in these last months.
The great majority of natural signals are not stationary and they undergo transients which
include a large range of frequencies in a short period of time. Fourier Transform is not
sufficient to process this kind of signals because all the information on time location of a
certain frequency is lost in the analytical process. The complexity of climate variability on all
time scales requires the use of several refined tools to unravel its primary dynamics from
observations. The analysis of the temporal series has shown that climatic variations are
extremely irregular in the time-space domain. This feature makes them difficult to be
foreseen if no particular mathematical tools are used. The aim of this paper is to describe
how to analyze and represent climatic signals in order to foresee their future short-time
evolution for weather nowcasting. A spectral analysis of the meteorological data is
described. More precisely the purpose of the spectral analysis is the determination of the
optimum sampling frequency for a set of climatic not stationary signals and the description
of an algorithm for the automatic adjustment of such a frequency.
sensor equipment normally consists of sensors for surface temperature, air temperature, relative
humidity, wind speed, precipitation and type of precipitation. This study tries to answer which
variables should be measured at an RWIS-station. This equipment has remained similar since
1979 when the RWIS-stations were first introduced. At a test site outside Göteborg some 100
climate variables, apart from the normal variables of an RWIS-station are measured. A neural
network model is used to select the variables that give the best prediction of the surface
temperature. Thereby recommendations of how to equip an RWIS-station can be made. Some
climatic variables would be difficult to include in the RWIS-system because of high maintenance
level, it may be practically impossible or simply too expensive. Results show that more
temperature sensors in the ground help the neural network model predict the surface temperature.
Ground heat flux and net radiation also improved the output of the model. The temperature
predictions by the model were good when common variables were used as input and were
improved when the additional variables were included. A forecast model from the Swedish
meteorological office (SMHI) was also given as input for the neural network model. While the
model from SMHI alone performed rather poorly, when combining it with the measured
variables and the neural network model a very large improvement was achieved. The neural
network had adapted and improved the output from the SMHI model to the site specific
conditions. The analyzed time series was only two months long, so it was too short for the neural
network model to learn how to predict occasions of special interest for road climate. A next step
is to use a longer time series and more stations to improve the forecasts and so the model can
learn to predict frost events. In the future the neural network model can be used as nowcasting
system to improve the output from forecasting models, such as the one from SMHI.
driver behaviour, vehicle performance, pavement friction, roadway infrastructure and
maintenance, risk management and prevention. Weather events and their impacts on roads
can be predictable, non-recurring incidents that affect safety, mobility and productivity. This
work shows a multi-model high-resolution forecast system which is tailored to transportation
weather-related problems and, particularly, to roads. It is based on leading diagnostic and
prognostic weather research capabilities which are being developed at the Epson Meteo
Centre (CEM). The system is a research and development effort, but some algorithms are
now operationally used at CEM. Two relevant winter case study are presented, showing the
capability of the system to forecast situations with a potential high-impact on road
transportation and safety.
TOPIC 3: TRAFFIC AND TRAFFIC SAFETY IN WINTER
compaction of vehicles on the snow layer on a pavement, a wheel-tracking (WT) test was
carried out. According to the WT test, the heat transfer coefficients between a tire and a dry
pavement or packed snow surface were obtained beside the snow consolidation due to normal
force of vehicle. Consequently, the former was 60W/m
K and the latter was 70W/m
proposed heat balance model well reproduced the snow and pavement temperature and could
estimate the snow height melted by traffic.
the traffic volume, some experiments were performed in the Cryospheric Environment
Simulator (CES). In these experiments, the artificial new snow made by a snowfall machine was
deposited almost levelly on the floor of the cold room. After that, the snow was compacted by
the shuttle action of a vehicle, meanwhile, the density and the hardness of snow and the skid
resistance number (BPN) of the snow surface were measured, under several snow temperatures
of 0, -3, -5, -10 and -15
C. At every snow temperatures, the skid resistance number decreased
with the traffic volume. The skid resistance numbers were more than 45 at the snow
temperatures of -5, -10 and -15
C even on the maximum traffic volume in this experiment of
486. On the other hand, the skid resistance numbers at the snow temperatures of 0
C and -3
were less than 45. In particular, in the case of the snow temperature of 0
C, very slippery snow
surface was formed when the traffic volume exceeded 200. When the snow surface was melted
by the solar radiation and the free water content of snow changed from 0% to 10% and 15%, the
snow hardness decreased suddenly and the skid resistance number increased suddenly
this paper. The research is based on traffic accidents data during the period 2001-2002 in
Vilnius city, Lithuania. The new approach is used to evaluate meteorological phenomena
impact on road collision rates. Traffic Accident Volume (TAV) coefficients are constructed to
estimate this impact. Results revealed that adverse weather conditions impact on road
accidents vary during different phases of meteorological phenomena. 3 different traffic
accident volume phases are marked.
topography and distance from water bodies. Most of the country is covered by mountain ranges
of Alborz and Zagros and some sparce mountains in the interior parts. Most of roads pass
through these mountainous areas. There are 237 dangerous and snow prone passes. Many of
these passes are blocked during cold months by heavy snowfalls, avalanches, blizzards, black
ice, dense fogs. Which causes considerable damage to the economy of the country.
Inclement weather creates a chronic hazard on Mountainous roads in Iran. Past studies indicate
that road collisions rate increase during precipitation & road frosts. The goal of the current study
is assessment of relationship between Road accidents and weather condition with statistical
activities in relation to the climate. In this paper, data from 654 Road Weather Information
System stations in Sweden were used during the five winter seasons, 1998/1999 to 2002/2003.
They where used for studying the distribution of slippery conditions throughout Sweden. The
number of incidents involving slipperiness during a winter season varies according to the
location in Sweden and the manner in which the slippery conditions were formed. Winter
indices were used to calculate the total number of slippery conditions. This was later compared
to Swedens latitudes. It shows a good correlation between the distribution of slipperiness and
Sweden latitudes, both for the total slipperiness and the different types of slipperiness. When
severe hoarfrost was analysed at the regional level, a high degree of correlation between
slipperiness and latitude was found.
variable speed limits. The aim of the study was to estimate the costs and the benefits of a
weather-controlled variable speed limit system that would cover most of the Finnish main roads.
The building costs and annual maintenance costs were evaluated based on a quality, up-to-date
system. It seems likely that a weather and road condition controlled system of variable speed
limits would be profitable in Finland. The benefit-cost ratio calculated with the most likely
starting values was 1.11.9. As a conclusion, it can be said that it is probably not advisable to
build variable speed limit systems very extensively but more of them should be built on highly
trafficked road segments.
planned in advance for efficient and effective winter road maintenance. To aid this process, a
salting route optimisation system which combines evolutionary algorithms with the neXt
generation Road Weather Information System (XRWIS) has been developed. The system can
cope with large-scale instances in the real world within reasonable computation times, to the
extent that daily dynamic salting route optimisation can be realised. However, the use of a
dynamic system, that each day adjusts salting routes in line with forecast road temperatures, may
increase complexity to the extent that user error will occur in the treatment regime. Therefore, a
robust (static) solution of salting route optimisation is also presented. Here, the emphasis is
placed on thermally ranking optimised routes so that the ‘warmer’ routes could be left untreated
on marginal nights.
to prevailing weather and road conditions. The model considers the variations in local climate
along the road stretches which is processed together with input from the RWIS. The speed
limits are set according to the severity level produced in four classes by the model.
Data from the Swedish RWIS system is used in order to classify type of slipperiness and the
severity of the situation in relation to the number of road accidents that take place during
each specific situation. The result shows a very clear picture that the number of accidents
increases with increasing severity of the situation. Most accidents occur during situation with
rain and a surface temperature around 0°C and during situation with snow.
TOPIC 4: ROAD WEATHER INFORMATION SYSTEMS: ACTUAL PROBLEMS
public. Meeting these demands posses unique problems for winter maintenance operations
personnel and the motoring public. Winter maintenance operations decision-making is a multidisciplinary process. It involves both earth and atmospheric sciences for maintenance
operations and effective communication of roadway and weather conditions into the motor
vehicle operator decision-making process. The global aspects of winter maintenance operations
call for an integrated systems approach combining the skills of the meteorologists, operations
personnel and the motorist.
This paper will outline the basic research which provided the foundational approach for
introducing advanced road weather information systems (RWIS) equipment and data into the
winter maintenance operations process and documents the successful implementation of the
technology. It will focus on the education and training processes used to change a workforce
culture from a reactive winter maintenance response to a proactive/systematic approach that has
raised level of service, reduced cost and lessened the negative impact to the environment.
Performance indicators will be illustrated and training investments quantified and explained.
Finnish Road Administration (Finnra, administrative authority), and the Finnish Road Enterprise (contractor).
The Finnish Road Weather Information System (RWIS) was started in the 1980s. Since then it has been
gradually developed as a winter maintenance tool.
As the winter maintenance was opened to competition, it also had an influence on RWIS. The road weather
information had to be made available to all winter maintenance contractors as easily and reliably as possible.
The new procedure also influenced the acquiring of the RWIS: according to Finnra’s strategies all acquisitions
have to be made through competitive bidding and contracts.
The purpose of this presentation is to describe the complex governing and administration of the Finnish RWIS,
which, as a very important system in Finnra, has an influence through the organization.
developed in co-operation between Czech and German national meteorological services during
winter season 20042005. This code is supposed for future cross-border exchange of road
weather data via Global Telecommunication System of the World Meteorological Organization
(WMO). The template and local descriptors were developed by Eva Cervena (CHMI) and
Sibylle Krebber (DWD) members of Expert Team for Data Representation and Codes (WMO)
in collaboration with authors of this paper. Local descriptors enable for example description of
road sensor position (fast/slow lane, between/in the wheel tracks), type of road construction, and
information about surrounding of the station. In RWIS it will make it possible to distinguish
stations according to different parameters, for example name of road sensor manufacturers if
necessary. In accord with national plans for migration from traditional meteorological codes to
BUFR, the new road weather BUFR template should be used operationally in the Central
Europe during winter season 2006/2007.
several servers. The user of these data wants to have full access to all information wherever
he / she may stay. The system should be extendable in order to provide for possible changes
in the future. This is the reason for developing a new complex system for presentation of
road weather data.
Weather Management Program and describes two projects sponsored by the program. The
first project developed siting guidelines for Road Weather Information System (RWIS)
Environmental Sensor Stations (ESS) to aid transportation agencies in the selection and
installation of ESS equipment, and to improve the usefulness of observation information
collected from ESS. The second project documented how weather information is being
integrated into transportation operations at 38 Transportation Management Centers (TMCs)
across the United States. Transportation agencies can benefit from the lessons learned from
TOPIC 5: FURTHER DEVELOPMENT OF ROAD WEATHER INFORMATION SYSTEMS
that provides information on the Sapporo area via website. The system has been in trial operation since
the winter of 2002/03, providing snow removal contractors of national highways with road icing and
snowfall information (current and forecast information) and with operation guidance on winter road
management based on that information. The trial MDSS provided current and forecast weather
information and operation support information, and incorporated an emergency reporting system.
In the winter of 2004/05, a new system that provides 1-km-mesh snowstorm information was developed
and added to the menu. A survey was given to snow removal contractors on the potential uses of
snowstorm information in the MDSS. Survey showed high potential for visibility information in winter
road management. They also showed that the accuracy of the visibility information that is currently
provided needs further improvement, and that this could bring increase the user satisfaction.
under construction. It was designed to combine modules related to geographic,
meteorological and snow cover information. To facilitate international application, it was
based on global topographic datasets such as products from the Shuttle Radar Topographic
Mission (SRTM, horizontal resolution < 100m) and on a meteorological forecast model data interface compatible with NetCDF and GrADS-readable formats. The downscaling system VERA (Vienna Enhanced Resolution Analysis), based on high-resolution terrain information, will be applied to add detail to meteorological model forecasts and to improve the analysis, using the readings of road weather stations. Hide Abstract
road condition models by establishing a novelty database containing detailed local
information from problematic road sections in Finland. Presently, a wealth of useful
information is available in various databases but not necessarily available to model
developers. A test set of some fifty most problematic locations have been selected by
studying how many of the accidents have occurred due to freezing, and basing on local
knowledge of the road maintenance people. More accurate information of these spots will be
implemented in the road condition models. The new system will be verified with pilot studies
during the next two winter periods by applying state-of-the-art verification methodologies.
Project ColdSpots is co-funded by the Ministry of Transport and Communications Finland,
Finnish Road Administration, and the consortium of three public and private partners: Foreca
Ltd, Finnish Road Enterprise and Finnish Meteorological Institute.
application of sophisticated telematic data systems to ensure fluent and safe transport flow.
Meteorological data acquired by numerous road weather stations installed along the
highways and motorways are essential source for creation of the specialized road weather
forecasts as well as for optimization of the road maintenance in an international scale.
Collaboration of meteorologists and road maintenance specialists from Czech Republic and
Germany resulted in establishing the unified road weather data code SH70 and development
of the BUFR template for road weather data.
The following contribution will present the data format and discuss the experience with the
mutual exchange, dissemination and presentation. Moreover introduce the basis of the future
Central European Road Weather Information System.
The flags of participating countries symbolize current status of international co-operation in
the region of the project.
is now widespread around the world. Road weather forecasts can be compared with road
surface temperatures and road condition on an hour by hour basis. However the road
weather forecasts are only normally made available for a limited number of road sensor
sites in a region. For example in Birmingham, in the UK, there is one forecast site for 26
salting routes. XRWIS is the NeXt Generation Road Weather Information System that
forecasts for every 20 metres around each salting route using a geographical information
system, sky-view factor analysis and mesoscale weather forecasts. Treatment requirements
for each salting route are then visualised in simple traffic-light style colours and a map is
displayed on the internet. In a recent trial in Devon in England, for 152 nights in the winter
of 2004/05, the XRWIS energy balance model IceMiser was run for more than 12,000
locations around 6 salting routes compared to just one forecast site that is normally used. If
the XRWIS system had been operational up to 78 salting runs on those six salting routes
could have been saved.