PROCEEDINGS OF THE 12TH SIRWEC CONFERENCE, BINGEN, GERMANY (16-18TH JUNE 2004)
TOPIC 1: CLIMATOLOGY, WEATHER CONDITIONS
supersaturated stratus crystallize and fall out as industrial snow. Predictions of industrial snow are possible
if the air temperature and the cloud base and top, respectively, of low stratus can be predicted.
This study aims to assist road management by clarifying localized formation mechanisms of avalanches in Hokkaido and establishing a simplified method of determining the avalanche likelihood. This paper investigates the relationship between avalanche formation and meteorological conditions on National Highway 453 in the Lake Shikotsu area, toward establishing a method of determining the avalanche likelihood in a specific area. We report the results of an experiment using the method in road management.
Regional forecasting office of Czech Hydrometeorological Institute in Pilsen has been equiped with IceCast software and has access to data from Vaisala sensors installed on highway D5 between Prague and border crossing Rozvadov Waidhaus and on some other sites in Western Bohemia too. This part of Czech Republic has complex terrain with elevations between 300 and 900-1000 m above sea level (the highest point of highway D5 is 620 m asl), some road stretches are in forested areas and some are close to water reservoirs or lakes. Our border mountains are also natural frontiers for temperature inversions with frequent occurence of low cloudiness and fogs. Intensity of transit traffic in this region is very high in spite of the highway A6 to Amberg and Nurnberg on German territory is not still finished. The second very frequent border crossing is on the road E48 near the town Cheb where Vaisala sensors are installed too.
Accident statistics shows that besides of black ice and snow cover the hoar-frost is often the cause of crashes. The danger comes to different dimension on highways, due to higher speed. So we utilize the opportunity to study data from relatively dense measuring network along the highway D5 and we prepared some basic statistics of conditions of hoar-frost formation.
A primary goal of our study is to identify the effective and efficient traffic data gathering technique in order to develop winter road management measurement that might be implemented in the context of Sapporo. Throughout the course of this study, we have attempted to demonstrate how the cutting-edge technology has the potential to help solve the problems with existing traffic analysis methods and increase the quality of data to determine the unique traffic feature in winter.
The cutting-edge technology we are interested in is floating-car data collected through putting Global Positioning System (GPS) on taxis running through the City. Because of its advantage covering the great Sapporo area and running throughout the day, this would enable up-to-date traffic data to be supplied over a wide area. Besides, partnership with taxi-companies provides cost-effective data collection solutions.
Several studies have been conducted in recent years on height-dependent variation in visibility. Sato et al. (2002) obtained visibilities at different heights by mathematical calculation, and demonstrated that when the wind velocity at a height of 10 m is 12 to 14 m/s, the visibility at a height of 2.4 m is greater than 300 m but at a height of 1.2 it is less than 300 m. This, however, was visibility at a snowfield, not on a road. Tozuka et al. (2001) measured visibility on the road while driving a vehicle equipped with the Snow Particle Counters at different heights. They confirmed that visibility depended on height, with visibility improving with increase in height. However, no quantitative examination was made as to how the height-dependent difference in visibility varied with weather and road conditions, or which factor made the greatest contribution to such variation.
This paper uses observation of mass flux of snow to investigate the effects of weather conditions and snowbank height on the height-dependen
The wind speed in the wake decreased with distance from the rear end of the vehicle and increased with vehicle speed. The turbulence characteristics had a similar tendency. Applying the surface layer theory to the wind field in the wake, we found the turbulent momentum transfer in the wake to be enhanced. The heating of air in the wake due to the running vehicle extended up to about 20 m from the rear end. The temperature of the exhaust pipe depended not only on the vehicle speed but also on the inclination of the road, since the engine revolution changes accordingly. The heating of air, defined by the air temperature difference between front and end, depended on the vehicle speed, inclination of the road, and atmospheric stability on the road.
The study area is located in the mountainous region of central Japan (approximately 200 kilometers from Tokyo). Figure 1 illustrates the study area and highlights a short stretch of the national highway 17 which is the study segment for the prediction model. It has an elevation difference of 675 meters over a 25-kilometer stretch of the highway. There are several tunnels, snow sheds and snow shelters that protect the highway from natural hazards. Highly rugged, mountainous terrain is assumed to affect the input meteorological variables used for the prediction model.
This study looks into different meteorological input variables in terms of how mountainous terrain affects them, and then it investigates the varying effects that the mountainous terrain exerts on the road surface predicted model.
TOPIC 2: FORECAST METHODS AND ACCURACY
We investigated nocturnal cloudiness for a case from winter 2001/02 in northern Switzerland. There, an ultra-dense combination of two networks with 70 stations in total is operated, measuring air and surface temperature, wind and other parameters. With the aid of our equations, these measurements where converted into cloud maps, including also precipitation seen by radar. We identified a frontal precipitation area, postfrontal clearing, freezing, and the first clouds of a following frontal passage. These findings will improve the observation of clouds and cloud movement as well as the prediction of road surface state and the risk of slippery conditions
time monitoring and post-event verification.
Automatic forecasts of road weather conditions has become possible due to the rapiddevelopment of computers in recent years. The RWM forecast tool is a numerical computer model which determines the local energy and moisture fluxes at the road surface.
The final forecast parameters are water and icejsnow on the road surface. In practicethe evolution of road surface temperature, temperature and dew point at 2 m height,cloud cover and precipitation are also vital parameters in order to forecast whether icecan be formed or persist on the road surface. This is because these weather parameterscontrol the energy conditions and moisture evolution on the road. The developed modelhas been documented (Sass 1992; Sass 1997)
The weather parameters including precitation, cloud cover, wind, temperature and humidity in the atmosphere are not forecasted by the RWM which computes the local roadconditions. An atmospheric model at DMI, namely the operational numerical weatherprediction model (DMI-HIRLAM) is used to predict the time evolution of atmosphericweather parameters which are transferred as input to the RWM. In this way the quality of the road weather forecasts depends on both the DMI-HIRLAM forecast and the RWMprediction.
Over the years it has become clear tram operational experience (Kmit and Sass 1999;Sass and Petersen 2000) that the quality of the atmospheric weather input to the RWMis critically important. This feature is specifically addressed in the present report.
The rapid growth of commercial off the shelf geomatics technology including GIS and GPS, has enabled the development of a new ice prediction technique called IceMiser (Chapman and Thornes, 2003). The ESRI GIS program ArcView is used in this study to run the IceMiser model.
TOPIC 3: SENSORS AND EQUIPMENT
When measuring evaporation rates at a moist road surface the eddy correlation technique might provide an advantage. It implicitely integrates over a large surface area (Schmid, 2002) and thus over spatially inhomogeneous road conditions, i. e. dry patches. On the other hand the representativeness of such measurements is problematic if used in connection with areas of limited extent (Foken and Wichura, 1996). Wojcik and Fitzjarrald (2001), for example, used an eddy correlation technique in oder to obtain evaporation rates from a concrete bridge. The measurements turned out to be strongly in uenced by the bridge’s surrounding area.
This study adresses the use of a water lm sensor and an eddy correlation system for a better quantication of the drying-up process on paved surfaces. A field campaign is undertaken. The resulting data are assessed concerning the water lm sensor’s performance, and the representativeness of both, water lm thickness and evaporation rate measurements
western Sweden. The Eddy covariance technique seems to underestimate the ows, but the Bowen-ratio technique seemed to work under good conditions.
TOPIC 4: PRESENTATION AND INTERPRETATION OF ROAD WEATHER INFORMATION DATA
In this paper a description is given of the data flow, – that is, from the two systems sources mentioned before, out on to the Internet.
1. The inability to identify precipitation type (e.g., rain or snow)
2. The difficulty to detect winter precipitation, especially in mountainous areas, due to shadowing and clutter effects and due to the small height above ground of precipitating clouds.
The first point can be overcome with polarisation radars (Atlas, 1990). However, these nextgeneration radar systems are not yet available for operational use. We proposed a much simpler method to identify precipitation type as rain, melting snow or snow (Schmid et al., 2002). Profiles of air temperature and dew point temperature are generated from ground data at various altitudes. These data are converted into fractions of snow within total precipitation mass (Koistinen and Saltikoff, 1998). Specific thresholds of this fraction define the height and thickness of the melting layer. This method to identify precipitation type is referred to as the KSS-method (Koistinen/Saltikoff/Schmid) hereafter.
For operational applications, it is important to know the performance of methods measuring and nowcasting precipitation in winter. In Switzerland, radar image data and short-term radar image forecasts are widely used for road maintanance (Schmid, 2000). Products resulting from the KSS-method (e.g., a radar image showing the height of the melting layer, see Schmid and Mecklenburg, 2001) have become popular, and the customer responses are very positive. However, an objective and quantitave validation of the KSS-method is missing up to now, mainly due to missing direct measurements of the type of precipitation.
This situation can be overcome with the Vaisala road weather measuring network, operated by the canton of Lucerne (Mathis, 2000). The type of precipitation is registered by an optical sensor. Hence, the data from this sensor are suitable for validation of the radar information and the KSS-method. This is the main purpose of this study. For predicting precipitation and its type, we use the radar images from MeteoSwiss (covering Switzerland and the neighboring regions) and ground network data of temperature and humidity (the so-called ANETZ, also operated by MeteoSwiss). The Lucerne data are used for validation. In the next section, we describe the data and procedures. After that, the main results of the study are shown. We end up by summarizing our findings and by discussing their consequence for future studies and operations.
1. How exactly the data describes the road conditions (validity)?
2. How well the data can be generalized?
3. How comprehensively the data covers the road network?
4. Is the data available all the time?
Validity of data is further affected by several factors, including:
1. Properties of the road weather measuring device in use.
2. Assembly of the device.
3. Maintenance of the device.
The purpose of the project described in this paper is to ensure the quality of road weather information by developing a quality management system for the road weather information system in Finland. Road weather data is collected from about 320 road weather stations and about the same number of CCTV cameras. Weather forecasts and radar and satellite images as well are passed through the system from weather service providers to supplement data collected by FINNRA. The system was initially built up to provide data for the planning of winter maintenance operations, but nowadays data is also used to inform road users and to control variable message signs (for a more detailed description of the system see Toivonen and Kantonen, 2000).
The project was initiated by FINNRA and it started in autumn 2002. A project group was formed, members of which come from FINNRA, winter maintenance provider organizations, measuring equipment assembly and service providers, Finnish Meteorological Institute and the authors as quality experts. The group worked from autumn 2002 through spring 2003 to build a quality manual for the road weather information system.
TOPIC 5: WINTER MAINTENANCE: MANAGEMENT SYSTEMS, METHODS AND POLICY
Therefore, the Institute of Highway and Railroad Engineering (Head: Prof. Dr. Ralf Roos) of the University of Karlsruhe was commissioned in the summer of 2001 to carry out the research project Optimization of Winter Maintenance on Heavily Travelled Freeways. The project primarily pursues two aims: Firstly, checking the capacity of highway cross-sections under wintry road conditions through a traffic investigation of winter-related congestions. Secondly, several pilot projects with individual highway surveillance centres shall be investigated with regard to optimizing the winter maintenance. These centres shall further receive scientific support throughout practical tests. In spite of the high standards of winter maintenance observed in Germany, individual measures for dealing with problematic road sections are gaining in importance, especially when considering the loss of capacity under wintry road conditions. It are these measures that shall be further developed within the stated research project.
The Nagaoka National Highway Office that is responsible for the Yuzawa District that is one of these heavy snow zones has, as one measure to achieve this goal, studied Road Performance Management intended to provide management that satisfies the needs of regions through a partnership between the public and private sectors. This approach is implemented by, based on the characteristics of the region and the views of road users, setting management goals for winter road surface management that is the most important challenge facing the region, performing management based on these goals, and evaluating its effectiveness. And, according to circumstances, working cooperatively with road users to improve this service.
This research that has been undertaken in response to the above circumstances is a study of the winter road management level suitable for a heavy snow region by focusing on the management goals and analyzing characteristics of the region, snow removal costs, and the degree of satisfaction of road users. The conclusions presented in this report do not directly reflect the views of the Nagaoka National Highway Office
For rapid and target orientated interventions actual information on present and forecasted weather situation, on present weather and road conditions in the network as well as on actual winter maintenance activities are of major importance.
A major effort is underway in the US through the MDSS (Maintenance Decision Support System) to enhance the value of weather forecasts to winter maintenance agencies. While part of MDSS is focused on improving the accuracy of forecasts, an important part is also focused on providing real time operational advice to agencies deployed fighting the storm. Thus, for example, one aspect of the MDSS is to recommend to agencies when to apply chemicals on a given plow route.
Such advice clearly has enormous potential benefits, but it is also stretching the limits of the technology at present and it suggests that an intermediate step may have significant value. The intermediate step considered in this paper provides a strategic plan of action (as opposed to what might be termed a tactical plan focused on individual plow routes) based on the forecast. The plan of action is derived using a matrix similar to that presented in Appendix C the FHWA Manual of Practice for Anti-Icing. However, the Appendix C matrix did not provide a particularly flexible description of a winter storm and was missing a number of critical operational considerations. These are considered in greater detail in the paper.
This new matrix will be undergoing field testing during the winter of 2003-04. Three agencies will be chosen and will receive the strategic plan of action along with their tailored weather forecasts. The paper will discuss the challenges and opportunities presented by this field test and suggest directions for future development.
Most of traffic accidents in the snowy season happen on the road condition changed from non-snowy road. For example, such locations are curves, bridges, tunnel exits and so on. In other words, removing the snow on such location can be appropriate management for winter road. Warmer regions have generally higher thermal energy in natural. When the natural energy is used appropriately, the higher effect for preventing the traffic accident may be obtained with lower cost. Mountain tunnels are generally planned to enable to cross short over pass. Therefore some tunnel exits are located on the high attitude and steep slope, and bridges and sunshades are possibly located at tunnel exits. The traffic accidents often occur at such locations during snowy season in mountain road. On such locations, e.g. bridges near mountain tunnel exit, snow-removing facilities are needed in order to keep the safety traffic condition.
The present study mainly focused on the pipe heating system for applying to tunnel exit. This system has considerably economical advantage when warmer tunnel spring water is used and where sufficient spring water is available. This paper discusses on the cost performance of pipe heating system by natural energy in a tunnel, i.e. initial cost and running cost. Cost-benefit ratios of each facility are also evaluated with assumption of service interval 15 years.
Boschung pioneered the Fixed Automated Spray Technology with a first system installed back in 1979 on the bridge of Flamatt, Switzerland. The TMS 1000 and TMS 2000 generations have lead this unequalled technology to world-wide recognition. In its constant effort to improve and refine its products, Boschung has now set a new milestone in the