Salting of Winter Roads: The Quantity of Salt on Road Surfaces after Application

Abstract

Chemical application forms an important part of winter maintenance activities with the aim of upholding a high level of accessibility, regularity and safety of roads during winter time. The quantity of chemical on the road surface is crucial for the road surface conditions and will determine whether ice formation or snow compaction occurs. For decision makers it is therefore essential to ensure that there is a sufficient quantity of salt on the road surface according to the prevailing road and weather conditions. At the same time adverse environmental effects are well documented. Therefore, there is a need to use as little chemical as possible while still ensuring safe driving conditions. An optimized chemical usage requires that decision-makers have sufficient knowledge of the quantity of chemicals on the road surface at any time.

The scope of this study is to acquire knowledge of how the quantity of salt on the road surface changes after application under different conditions and thereby to learn about the durability of salting actions. Further, it is to identify the physical processes that control the changes in the quantity of salt after application and important factors behind these processes.

To study the questions addressed, field observations have been conducted. The field observations have been carried out on an ordinary road, open for traffic. The method has been to document the salt quantity before and after application. During the observations the salt and water quantity on the road surface was measured along with weather parameters, traffic data and data from the maintenance trucks. The salt quantity has been measured with the Sobo 20 instrument.

Sobo 20 is a portable instrument that allows measurements on several locations and in different positions in the cross profile of the road. The unique feature of Sobo 20 is that the instrument itself adds measuring fluid onto the road surface during the measuring procedure and thereby is able to calculate the salt quantity on the road surface in terms of quantity per unit area. To document the accuracy and limitations of the instrument, some tests have been conducted. The conclusion is that Sobo 20 accurately measures the quantity of salt as brine, on both smooth surfaces and asphalt pavements. However, it only detects between 5 and 6% of dry salt particles. Re-crystallized salt, made of finer grains, is detected at 58 % on smooth surface and 49% on asphalt pavement. When using the Sobo 20 for the measurement of dry or pre-wetted salt, the displayed value must be interpreted only as the quantity of dissolved salt on the road surface, and not the total salt quantity.

The field observations clearly show that there are large spatial variations in salt quantity after application due to the effect of traffic. There is large variation in the cross-section profile of the road. Not surprisingly, higher quantities are measured at road edges, between wheel tracks and at the centre of the road compared to inside wheel tracks. This is explained by salt gathering because of the traffic effect and the fact that there are higher quantities of water in these areas that allow more salt to dissolve. For the further examination of the changes in salt quantity as a function of time or traffic, it has been chosen to focus mainly on the salt quantity in wheel tracks.

The results show significant differences in the quantity of salt after application between the various observations, and some of these differences can clearly be explained by the quantity of water on the road surface. The quantity of water on the road surface determines the quantity of salt after application. Wet road surfaces both dissolve and loose salt more rapidly than moist road surfaces. The data also show that there is a surprisingly rapid loss of salt, especially on wet road surfaces. After 200 to 400 passing vehicles, the quantity of salt equals that before application. Further, it is also clear that the measured salt quantity after application cannot be described by a simple linear or exponential decrease in salt quantity. Shortly after application there is an increase in the measured salt quantity, thereafter followed by a decrease in salt quantity.

Based on the results from the field observations, three different physical processes that control the changes in salt quantity after application are identified. They are initial loss, dissolution of salt, and loss of salt. The initial loss of salt occurs at the time of spreading. The dissolution process is the process whereby solid salt dissolves in the water or brine present on the road surface. The process is mainly time-dependent and is relevant when spreading dry or prewetted salt. The loss of salt after application is time and traffic dependent and three distinct mechanisms have been identified that remove salt from the road surface: blow-off, spray-off and run-off. Blow-off is described as solid salt that is blown off the road surface by traffic, spray-off is dissolved salt sprayed off the road surface by traffic, while run-off is drainage of dissolved salt from the road surface.

From the identification of the processes that control the salt quantity after application, a physically based model for the salt quantity is proposed. The model is certainly based on several simplifications and assumptions. The most important are that the dissolution and loss processes are independent, that there are linear relationship between time and traffic and that there is no run-off. The model is adapted to the empirical data from the field observations. The analyses show that the model produces a satisfactory fit with the data from the field observations, and it therefore seems reasonable to conclude that these processes and the model developed can explain the changes in salt quantity after application. Although the model seems to fit the data for salt measurements, it is realized that it makes some assumptions and simplifications that may be incorrect. To achieve a more precise model, further development is needed, for example the introduction of a function describing run-off and the incorporation of the effect of different vehicle types. In addition to a more complex model, data of higher quality are needed.

The main results of this study are in the identification of the physical processes and in the principle of building a physically based model for the salt quantity on road surfaces. The attempt to understand the physical processes is essential to achieve a more thorough understanding of the phenomena of salt quantity on road surfaces after application.