3.1 General
The role of a good gravel road wearing course is to increase the functional and structural condition of the road. It improves the functional condition by providing a smooth and comfortable drive with minimal dust problems. At the same time the wearing course should be dense enough to ensure that rain, or water from melting snow, does not infiltrate into the road structure and weaken it, and instead flow to the road shoulders and further to the ditch. Where this is not the case, the life cycle cost of a poor quality wearing course material will be that much higher as it will require many more maintenance actions to keep the road serviceable.
The performance of the gravel road wearing course will also depend on the traffic volume on the road. On roads with high traffic volume the material specification should be higher, while in low traffic volume forest roads the material quality can be slightly less. That is why on forest roads the wearing course can also be called the “binder course”.
3.2 Wearing course material
3.2.1 Material specifications
Gravel road wearing course material should have a) enough friction to bear the load caused by vehicles driving on the road, and b) enough suction/cohesion to cause tensile stress in the wearing course to keep the surface firm and prevent dust arising. The problem however is that materials with high suction properties turn plastic easily when their water content gets high. Additionally materials should not have large pebbles/stones that could cause unevenness to the road surface and break car windows. With this background the recommended grainsize distribution, or ‘grading’, of the gravel road wearing course in Nordic countries is based, according to the authors, on the tests made in Sweden in 1920’s and in 1930´s. In this research, wearing course gradings were analysed from material that 1) did not have plasticity problems when they were wet, and 2) did not create severe dust problems in dry summer months. That part of the grading curve that met these two requirements was chosen to be the optimal grading curve envelope.
Based on this research the grading curves in Finland have been classified into three categories where the recommended grading envelope is also dependent on the material the wearing course is made of (envelopes 1 and 2). Gravel and forest roads with very low traffic volumes have a slightly wider grading envelope (3).
The work above however is based only on experimental data and does not explain the theory behind the performance. For that, thermodynamic soil suction theory explains how gravel road wearing course materials perform. Soil suction describes the energy level, also known as Gibb´s free energy, at which water is bonded to a particle surface, and that which is needed to release bonded water molecules back to free water. The most important suction components affecting the mechanical performance of unbound road structures and subgrade soils are 1) matric suction, 2) osmotic suction and, in cold climate areas 3) cryosuction. The sum of matric suction and osmotic suction can also be called “total suction”. Matric suction is mainly affected by void ratio, voids’ size and the amount of fines in the material, while the amount of ionic compounds mainly affects the level of osmotic suction. Cryosuction becomes effective when the temperature in the soil or road materials drops below 0 oC. Cryosuction explains the forces leading water to move towards the freezing front and the formation of ice lenses in road structures and subgrade soil.
In unfrozen soils with low moisture contents, suction generates tension in the pore water between soil particles and increases the stiffness of the material, and leads to high modulus values. But when moisture content increases suction decreases, and further at high moisture content positive pore water pressure under increased loads can reduce the material’s resistance to permanent deformation and the material can become plastic.
Simplifying the theory described above, suction in a gravel road wearing course can be controlled by the amount of fines content in the material (matric suction) or by increasing dust binding chlorides in the wearing course (osmotic suction). In addition it is important to remember that when matric suction is moisture dependent, osmotic suction is not.
Gravel road wearing course source materials can vary. In most cases the material is made of crushed natural gravel or crushed hard rock aggregate. If these materials have low fines, the fines content can be improved by adding fine grained soil, such as clay. For instance, in the “old days” a Finnish gravel road wearing course was called “savisora”, i.e. “clay gravel”. In addition very good quality gravel road wearing course material has been made from crushed coarse moraine (glacial ablation till), where the fines are mainly crushed particles and the plasticity of the fines are low.
The mineralogy of the wearing course aggregate also has an effect on the material performance. For instance, the fines of crushed quartzite are like glass grains and provide high friction, but not so much suction. These aggregates are good for moist road structures and for roads with deformation problems, but their binding properties are not good. On the other hand crushed dolomite has good binding properties, but may become plastic and slippery after heavy rain and repeated loading.
In addition to grading, the course aggregate of the wearing course material can also have some specifications for the shape of the aggregate. They cannot be too flaky as “knife-like” flakes can have a tendency to punch car tyres. Specifications can also state how hard the aggregates should be such as using the Los Angeles test or BallMill test. Finally in some countries have some mineralogical specifications for the aggregates, especially the avoidance of use of aggregates that have a high mica content.
3.2.2 Material testing methods
Gravel road material testing methods are essentially the same as testing methods for unbound base course materials. Of these, the grain size distribution analysis using the wet sieve test method is the most important. This should be always done when analysing the quality of the current wearing course and the quality of the new wearing course aggregate. For instance, low fines content (< 0.074 or < 0.063 mm) means a potential dust problem and, on the other hand, high fines content indicates a high risk for plastic deformation during the spring thaw or after heavy rain. However it is not just the amount of fines, but also the quality of the fines that matters.
A general rule of thumb is that if the fines content after wet sieving is higher than ~10-15 %, the particle size distribution of the fines should also be determined. This can be done by using several methods, which can be based on sedimentation or laser diffraction. The most common method, based on sedimentation, is the Areometer method. This method is cheap, but slow, and the test normally takes a few days. Once obtained, the results from the fines particle size analysis should be integrated with results from the wet sieving test. When analysing these results, special attention should be given to the clay content (content of particles under size 0.002 mm). In some countries, like Iceland, the amount of clay particles in fines should be 10-30%.
Due to the fact that in gravel road widening the wearing course material from the road shoulders is often graded back to road surface it can contain some organic material. This can increase the water adsorption properties and moisture susceptibility of the wearing course. For this reason many laboratories when making particle size distribution analyses also determine the organic material content as well. Organic material content can be defined, for example, by loss on ignition
ROADEX recommends the Tube Suction test for the measurement of water suction properties of wearing course material especially if it has moisture susceptibility problems. This test measures if the aggregate adsorbs moisture when it is available in its surroundings. The amount of adsorbed water is monitored by measuring the dielectric value of the material at specified time intervals. The dielectric value is mainly a function of volumetric water content in the material. It is also recommended that the electrical conductivity of the material is measured during the Tube Suction test as this indicates the amount of osmotic suction in the sample, and how much of the aggregate has chlorides or organic material.
The Tube Suction test is carried out in the following way. Before starting the test, samples are compacted into plastic tubes, 200 mm long and 150 mm diameter, and dried at +40-45°C for at least 3-4 days before being left to stand at room temperature for at least 2 days. The bottoms of the dried samples are then placed in distilled water. Dielectric values and electrical conductivity are then measured with a measuring device from the top of the sample at specified intervals (30 minutes, 1, 2, 4, 6, 8, 24, 32 hours and thereafter once a day from 2 days up to a minimum of 10 days until values have become steady). The magnitude and growth rate of the dielectric value reveals how much and how fast water rises to the top of the sample by capillary forces.
Unbound materials can be classified according to dielectric value. If the dielectric value is less than 8, the material may have too little suction for a wearing course and could have dust problems. If the dielectric value is 8-12, the material has optimal total suction and moisture. If the dielectric value is 12-16 the material has optimal suction for dust binding purposes, but the amount of the fines should not be increased. If the dielectric value is higher than 16, the material will adsorb too much water and will be susceptible to permanent deformation, as well as being slippery during rain. If the dielectric value is higher than 20, the material is inappropriate for use as a wearing course material. If the electrical conductivity of the sample is high this means that the sample may contain high amounts of chlorides, or contain harmful weathering products from the aggregate minerals.
If the Tube Suction test results show too high dielectric values and the electrical conductivity is low, the solution is normally to coarsen the aggregate, ie decrease the amount of fines. If the dielectric value is too low the fines content can be increased or dust suppressant used.
If the quality of fines is of specific interest, the method that could be used is the water adsorption test. The water adsorption index indicates the potential for moisture to adsorb to the surface of the fine particles at 100% relative air humidity. It is also an indicator of how active the interaction is between the material and water. If the water adsorption index is modest (< 1 %), it is very unlikely that the analysed material will be moisture susceptible, and that moisture susceptibility is the reason for road damages. Adsorption values greater than 3% indicate some kind of problem if there are no chlorides in the sample. High water adsorption values should be always compared with electrical conductivity values from the Tube Suction test.
When interpreting test results, it should always be kept in mind that the samples from road structures may have traces of contaminants, e.g. dust suppressant salts. If the material contains chlorides, these can have an influence on test results, for example the water adsorption index and electrical conductivity as measured in the Tube Suction test. Chloride content can be defined by a number of means, e.g. titration.
3.2.3 Wearing course thickness
Most countries recommend gravel road wearing course thickness to be 50-100 mm. Thinner wearing courses wear easily exposing large stones from the base course requiring many maintenance operations to correct. The problem with thicker wearing courses is that, because their fines content is high, they can become water and frost susceptible. In these cases, and if they lose their strength, the resulting thick and plastic wearing course can make the road difficult or impossible to drive with normal passenger cars.
Unlike a bituminous pavement the problem with a gravel wearing course is often that its thickness varies substantially both in longitudinal and transverse directions. This thickness variation causes problems in designing any maintenance operations.
The reason why thickness changes can be the result of deformation, as described above, but it can also be due to the action of maintenance machines and how they are used. They too can have an effect on the thickness of the wearing course
If this movement of material is not understood, or ignored by the road maintenance crew, it can lead in the long term to deformation problems in the wearing course. This effect can be seen many times during spring-thaw or rainy period
3.2.4. Dust binding additives
Dust suppressants are used to prevent the emission of dust (fines) into the environment and this is especially harmful for households living close to gravel roads. Gravel roads with high dust issues can create hazards for traffic safety. In addition loose gravel road wearing course material can increase the risk of broken windscreens in vehicles. The use of dust suppressants can reduced the overall costs of gravel road maintenance as, for instance, they can increase the lifetime of the wearing course and reduce the number of blading operations necessary during the summer.
There have been hundreds of different dust binding products, also called dust stabilizer or dust suppressants, for gravel road maintenance on the market. They have been classified in the following way, the final three (in bold) being the most commonly used:
- Organic non-bituminous binders, including different industrial waste products and for instance vegetable oils
- Electrochemical products (enzymes)
- Microbiological binders, microbes
- Polymers that cement fine particles
- Ionic stabilizers that flocculate clay particles
- Surfactants that reduce the surface tension of water
- Hygroscopic salts (chlorides)
- Lignosulphates
- Bitumen emulsions
Calcium chlorides (CaCl2) and magnesium chlorides (MgCl2) are globally the most common used products. They can be used either in flake or liquid forms. The third chloride product is sodium chloride, also called road salt, but this is not so effective as calcium and magnesium chlorides in dust binding. They are generally termed hygroscopic products as they adsorb moisture (osmotic suction) even from the air and keep materials moist. Other benefits are that they reduce the thickness of the adsorption water film around fine particles and further make compaction of the material easier.
Calcium chloride has been the most popular dust binder in Sweden and Finland, but magnesium chlorides have also been used both in flaky and liquid forms. The recommended amount of chlorides to be used in gravel roads in Finland can be seen in the following table:
If liquid chlorides are used, the calcium chloride content of the liquid can vary roughly between 30-40%. The amount of dust binders can be higher in open and sunny areas and especially around households. On the other hand, in shady road sections chlorides are used mainly to help in compaction of the wearing course. If road structures are thin, and there is a capillary connection from road surface to ground water, they should not be used as the chlorides can infiltrate into the groundwater very quickly through osmotic forces. Evidence of this has been seen in the Percostation frost monitoring stations in Finland. High contents of chlorides may also road surfaces to turn slippery during heavy rain.
In Nordic countries dust binding is normally used during the spring maintenance operations, and later during the summer only if road suffers from dust problems.
Lignosulphites, also known as lignin, are a major constituent of wood. Lignossulphites act as a natural glue, binding the dust particles so that they also act as binders. Their main problem has been that they are water soluble and need to be respread after heavy rains. However lignosulphites do not cause recurring risks to the environment as they can be degraded by micro-organisms. Lignosulphites are more sensitive to variations in fines content than chlorides and bitumen emulsions.
Bitumen emulsion has also been used as a dust suppressant in gravel road wearing courses. Bitumen emulsifiers can be cationic, of positive charge, or neutral. Cationic emulsifiers have proved superior due to their ability to coat and adhere to aggregates. These emulsions work best where the fines amount is small.
3.3. Wearing course maintenance in different seasons
This section will focus on wearing course maintenance during different seasons based mainly on Nordic practices and their recommendations for gravel roads. Other seasonal maintenance operations will be discussed in later sections.
3.3.1. Spring maintenance
Following the end of winter proper, spring maintenance operations become very critical when thinking about the long term performance of a gravel road over the following summer. If spring maintenance operations are properly made, the need for grading or blading in the summertime will be substantially reduced. During the spring maintenance operations it is also very important to reshape the road surface crossfall to 3-5% as it will have most likely flattened during the summer.
Spring maintenance operations should be normally started after the spring-thaw weakening period, after the road surface weakening is over and the road surface structures are dry enough. Operating heavy maintenance machines too early on the road may turn wet materials to become plastic and cause more damage. On the other hand, spring maintenance operations should not be delayed too long as then the road surface, and especially the wearing course, may be too dry and stiff to be processed. There will always be a few road sections during spring road maintenance where the frost has not thawed yet. At the same time those sections with spring thaw weakening problems should be treated by adding extra aggregate.
During the spring gravel road wearing course maintenance, especially on roads with higher traffic volumes, it is recommended to add some salt into the wearing course. There are two main reasons for this. First of all salt reduces the thickness of the adsorption water film around the aggregate particles, especially the fines, which leads to the material getting a much better compaction level even without separate compaction equipment. The other reason is that the salt also prevents dust issues in areas with dry road bodies
The most important issue in spring maintenance is that the wearing course is processed deep enough and at least deeper than the bottom of any potholes. If this is not done the potholes will appear soon afterwards in the same places than they were before. The other important issue in spring grading/blading of the wearing course is that the speed of grader/blader is not too high as this can easily lead to a wavy road surface and later problems with washboarding.
Finally, compaction of the gravel road wearing course during the spring maintenance is always recommended after grading or blading. This will reduce the number of maintenance measures required during the summer time. The compaction operation however adds to the maintenance cost and there are now several systems where compaction wheels can be added to the grader/blader to have an inclusive operation. If this is not possible another way to cheaply compact the wearing course is to drive the maintenance truck along the different lines of the wearing course cross section during operations.
On forest roads spring maintenance operations are mainly focused on fixing the most severe problems that have appeared on the road during the preceding winter or spring period. Because road structures are mainly weak at this time heavy maintenance machines can do more harm than heal the problems. For this reason maintenance operations should be carried out in the summer when the crust is dry. If needed, some light blading can be made during the spring.
3.3.2. Summer maintenance
The ultimate goal with a gravel road wearing course is that there should not be any need for the heavier maintenance measures during the summer. This means that the wearing course should stay firm and without dust problems for the whole summer. This will not happen however in most cases and some maintenance measures will be required during the summer such as dust binding, patching potholes and levelling sections with washboarding problems.
In summer the amount of dust binders should be lower compared to spring, and binders should be used only on sections with dust problems. Both flaky and liquid chlorides can be used, but if flaky chlorides are used then water should be spread on the road surface before spreading the flakes or they should be spread immediately after rain. The treated road surface should not be graded or bladed.
Potholes should also be fixed locally using a grader or blade, or they can be filled manually. If the road has washboarding (corrugation), these sections should be fixed with a grader or blade after rain at a very slow speed. The grading depth should be deeper than the bottom of the potholes and the bottom of the lowest parts of any corrugations.
On forest roads early summer maintenance operations should mainly focus on shaping and making the road surface even and returning the material from the road sides back to road centre. Gravelling can be done locally on spots where rut depths are deep and new material is needed. The main gravelling operation should be made during the fall.
3.3.3. Fall maintenance
The main task for fall maintenance should be to prepare the gravel road surface for the winter. Special interest should be paid to bringing the road back to proper crossfalls. Crossfalls often get too flat during the summer time as material is moved to the road shoulder. If this is the case, water will not be able to flow away from the road surface during autumn rain and this can lead to severe surface thaw weakening problems after the winter. Conversely, if the crossfall is too steep (gt; 5%), vehicles may slide away off the road during the early winter days when the road surface gets icy.
If the wearing course thickness has proven to be too thin during the summer, or if the grain size distribution is not adequate, gravelling should also be made during the fall maintenance. For this, new material is spread on to the road from the truck and then shaped with a grader, dragged blade, or underblade below a truck. The added material should then be compacted.
On forest roads the main maintenance measures should be made during the fall when the road has had sufficient moisture from rainfall. At that time the aggregate is easily bound to the existing wearing course which reduces the risk for permanent deformations during the continuous rains in the fall. Further gravelling in the fall can help with friction when the road is frozen during the winter.
3.3.4. Winter maintenance
In Nordic countries gravel roads wearing courses, and forest road surfaces, are normally covered with a thin layer of compacted snow during the winter time. If winters are cold enough the compacted snow layer should be stable, and not subject to periodic thawing every now and then. This compacted snow layer can protect the wearing course and keep it even over winter time. It is also easier to blade compacted snow than a frozen wearing course. Compacted snow cover can be created in the early winter by not ploughing the snow away from the road surface and letting the traffic compact the snow into a layer. During the winter time special attention should be paid that the compacted snow layer does not grow to be too thick. In the spring this layer should be removed before the weather warms and makes the layer soft.
In countries where a compacted snow layer is not possible the wearing course has to be maintained in the normal fashion.
Friction control is also part of winter maintenance when the road surface turns icy. This is done mainly by spreading sand on to the road surface.
3.4. Monitoring wearing course condition
The techniques and methods for monitoring the condition of gravel and forest road wearing courses are mainly the same as used for monitoring the service level of paved roads. The only exception is monitoring the thickness of the wearing course. What can and should be monitored is discussed in the following.
3.4.1. Monitoring wearing course thickness
The easiest, and possibly the most commonly used, method to monitor gravel road or forest road wearing course thickness is by visual inspection. This method however only identifies those locations where the wearing course is missing and bigger stones from the base course are showing through, i.e. when a gravelling operation is necessary. The downside with this method is that it does not point out sections where the wearing course is too thick. Furthermore, as discussed earlier, the thickness of the wearing course can vary substantially across the road cross section, as results from the Finnish tests made with GPR and test pits have shown. That is why the thickness of the wearing course should be checked with GPR or by test pits especially if there are problems with the wearing course performance. At the same time samples can be taken from the wearing course.
Experiences from Finland have also shown that wearing course thicknesses may vary substantially from one maintenance area to another based on how the decision on regravelling has been made.
3.4.2. Monitoring gravel and forest roughness and rutting
The main performance indicators for gravel and forest roads are longitudinal roughness and rutting, but visual firmness evaluation parameters can also be used. In general roughness problems can be summarized by three main types from the longest to the shortest wavelengths: 1) bumps, 2) potholes 3) corrugation or washboarding. Rutting can also be termed transverse roughness. Bumps are not actually a wearing course issue but they are described in this part of the learning. A special case of transverse problems are weak road shoulders.
ROADEX has also tested the use of a laser scanner in pothole detection and the results were promising. The only problem was that, if cheaper laser scanners were used, the driving speed of the survey vehicle could not be very high. Also, if the road had a dust problem, the laser scanner had to be in front of the vehicle.
For longitudinal roughness surveys, the only technique that has been used for a long time has been the bump integrator that measures the changes in the distance between a car axle and car body. The bump integrator can also be built as a special single wheel trailer. In recent years however a range of different kinds of accelerometer roughness surveys have entered the market. These are applications that use accelerometers in cell phones or specially designed 3D accelerometers that are installed either on the back axle of the vehicle or on the car floor under the driver’s seat, which simulates driver comfort the best. Finally some tests have also been carried out using laserscanner scanning longitudinally along the survey vehicle.
Over the recent years smartphones have become popular methods in roughness surveys. In these survey systems a smartphone is connected to the front windscreen of the vehicle and a cell phone app collects its accelerometer data and takes photos of the road during the driving. Positioning is based on the cell phone GPS. The data collected is uploaded to the cloud. These systems are cheap but their repeatability has had some issues. They can however provide a great view where the problem sections are located on gravel or forest roads.
For gravel road surveys there are also industrial 3D accelerometer base systems where accelerometers are installed on the back axle of a vehicle or on the vehicle floor under the driver’s seat. These accelerometers can be calibrated for different speeds and different vehicles. The greatest benefits of these new systems are that they collect and store the data automatically and can be installed in any vehicle driving regularly on the gravel roads.
Laser scanners can also used in measuring the longitudinal profile of the road surface before and after the vehicle. If the road has dust problems the calculations can only be done from the data from the front of the vehicle. The benefit of this technique is that it can measure other features that other systems cannot do, e.g. the height of a frost bump or settlement.
Gravel road roughness problems can also be classified based on the wavelength of each problem: short, medium and long. Short wavelength problems with wavelength lt;1-2 m can be mainly related to potholes or washboarding. Medium wavelength problems with wavelength, 2-10 m, are generally bumps due to frost heave, or sharp settlements that can be related to culverts or utilities. Long wavelength problems can be mainly related to geotechnical issues such as settlements.
Because of the nature of gravel road wearing courses, rut depth monitoring and specifications have not had much attention in the ROADEX countries. In most cases gravel roads have had other more urgent problems needing maintenance measures before rut depth becomes an issue and that is why there are only a few recommendations for the rut depths. In the USA gravel rut depths have been classified in the following way: low severity rutting 0-2.5 cm, medium severity rutting 2.5-7.5 cm and high severity rutting >7.5 cm. However on narrow forest roads high rut depths are often a more severe issue. In these roads maintenance or rehabilitation measures are recommended if the rut depth exceeds 10 cm.
3.4.3. Wearing course moisture / plasticity risk monitoring
Gravel road wearing course quality has not yet been monitored with non-destructive methods. The only known test by the ROADEX countries has been the Ground Penetrating Radar (GPR) test in Finland where gravel road wearing course surface quality was measured with an air coupled GPR horn antenna using the surface reflection technique. The results obtained were then compared to the performance data of each gravel road section, as well as the wearing course grain size distribution and wearing course chloride content of the samples taken. The results of these tests were quite promising and, based on the dielectric value of the wearing course, a general recommendation for wearing course actions based on dielectric values were made for Finland.
3.4.4. Monitoring dust
Dust is an environmental problem in many ways. Dust problems appear mainly on gravel and forest roads with “dry road structures”, but many times dust binders are spread evenly over the road section. In these cases dust binding chlorides can quickly move through capillary connections to groundwater, which can be hazardous to the environment. That is why an objective monitoring system for dust is needed.
So far dust in ROADEX countries has been monitored mainly based on the different visual emission observations using 3-5 class classification systems. In many countries maintenance crews react also to dust problems based on feedback from people living in houses adjacent to the roads.
Stationary dust monitoring systems measure dust concentrations objectively using advanced forward light scattering principle. These monitors run for extended periods of time to determine the maximum, minimum and average dust concentrations at a location. Motion cameras can also be placed by the side of the gravel road to collect data over time. These can provide photographic evidence of any dust plumes generated by traffic.
One potential test method that has been tested in Finland and Sweden is the laser scanner based method. In this method a laser scanner is mounted on the survey vehicle to survey the gravel road behind the vehicle. While driving, the laser scanner scans the road and calculates all the laser points hitting dust particles before hitting the road surface. The number of these “air hits” correlate directly to the severity of dust emissions from the road and tell exactly the locations where appropriate dust suppressants should be used.
3.4.5. Monitoring crossfall and its changes
Correct crossfall is one of the most important parameter indicators in the condition of a gravel road. Too flat crossfall leads often to potholes and rutting, and high crossfall can be a traffic safety issue especially in winter. Crossfall can be roughly measured with 3D accelerometers and very accurately with a laser scanner and inertial measurement unit (IMU) combination.
Rapid changes in crossfall can also be a safety issue especially to longer trucks because of the high risk of warping. ROADEX has introduced the Swedish technique of ‘Rut Bottom Cross Slope Variance’ (RBCSV) to calculate warping risk. If this risk is higher than 1% the section should be corrected, or the allowed driving speed lowered.
