1 Introduction

Unsealed roads (also called unpaved or gravel roads) are roads covered with gravel as a wearing course. Australia has an unsealed road network with a length of 574,660 km, which is over 60% of the entire road network in the country [1]. Maintenance and construction of most of these unsealed roads fall into the care of Local Government bodies such as Regional Councils. While traffic volumes are generally low, being less than 25 vehicles per day [2], the need to maintain the gravel roads to an acceptable level of service and provide all-weather access still exists.

The behavior of unsealed roads is dependent on several factors, including material specification, drainage capacity, traffic loading, precipitation, climate, and maintenance practices. The main problems associated with unsealed roads are dust, corrugation, shape loss, insufficient drainage, gravel loss, potholes, ruts, loose aggregate, and frost damage [3, 4]. Gravel is a finite resource, and as preferred borrow pits reach the end of their useful life, the ability to find other pits with suitable gravel material for road construction is becoming more difficult. According to “Austroads: guide to pavement technology”, the three critical characteristics of an unsealed wearing course are particle-size distribution (PSD), plasticity, and aggregate hardness [5].

Particle-size distribution plays a vital role in the gravel strength through the particle interlock and the maximum density principle. PSD also controls the material’s permeability, focusing on the percentage of material finer than 0.5 mm.

Strength is also achieved through plasticity, whereby the fine material contributes to the aggregate’s density through reduced interlock when wet and the provision of a cohesive strength to hold the aggregate in place when dry.

In addition, hardness is an essential characteristic in that aggregates are expected to be of sufficient hardness to resist significant breakdown under compaction and trafficking and be durable enough not to break down due to exposure [5].

Jones and Green [6] outlined new performance-related specifications (i.e., maximum particle size, shrinkage product, grading coefficient, treton impact value, and California bearing ratio value or CBR) for wearing course materials on unpaved roads (Table 1). These specifications have been included in the Austroads Guide to Pavement Technology—Part 6.

Table 1 Recommended material specification for unpaved roads [5]
Full size table

Generally, pit gravel will rarely meet these specifications, and some form of stabilization is usually required to improve the gravel properties to conform to these specifications.

It is well reported that there have been varying levels of success in modifying gravels through different methods to prolong the life of these pavements and therefore extend the life of gravel pits and reduce the impact on the environment. This also, in turn, reduces costs associated with maintaining the unsealed road network and improves rideability for road users due to the slower development of defects due to gravel loss and shape loss.

Current stabilization procedures can involve using lime cement or a cement blend as an additive, chemical polymers, or blending various gravel to achieve this improvement.

The use of lime is usually incorporated when the material has a Plasticity Index greater than 12, signifying a higher clay content. In this method, the lime reacts with the silica and alumina in the clay [7].

Cement performs a similar role to lime, only requiring water to take effect. The use of blends incorporating blast furnace slag is also a common practice as it assists by slowing the curing process, providing additional working time, and reducing shrinkage as it reduces cracking [7].

Chemical stabilization is another process incorporating polymer for stabilizing materials, such as clay, silt, and soil particles. This creates a soil matrix resistant to water’s detrimental effects by reducing permeability. This method can be beneficial in gravels and performs well in dry weather; however, it might result in quicker deterioration of materials when exposed to water [7].

While these methods have proven successful, there is an additional cost to purchase and incorporate the additives, which may be cost-prohibitive. In light of this, another technique called granular stabilization could help achieve the required properties of gravel materials. Granular stabilized materials are those to which another granular material is added to correct a deficiency in the intrinsic material properties of the parent material [8].

Blending gravels to improve performance is not a new area of study. In Queensland alone, several attempts have been made on blending and granular stabilization methods.

Hanson [9] conducted a study in the Toowoomba district based on road trials of various gravels already used throughout the district for unsealed wearing courses. A couple of these road trials were constructed using blended gravels. The tests conducted during this project were Particle-Size Distribution and Atterberg tests. Although CBR is a critical factor in determining the strength of pavement material, the CBR test was not considered in this research due to time and facility limits. Nevertheless, this project showed the best chance of optimal performance for the blended gravel.

Yin [10] conducted another study within the Toowoomba region in Southeast QLD (Queensland) to evaluate the specifications of local gravel sources and stabilized gravel materials for sealed and unsealed roads. Tests conducted in this project included Particle-Size Distribution, CBR, and Atterberg tests. In this research, the materials were sampled directly from the pit. Therefore, they did not appear to be a representative sample of the actual materials, usually prepared for use on the road surface through different types of machinery such as rollers or crushers.

In another study conducted by Smith [11] in the Maranoa Regional Council area of Southwest QLD, 31 gravel pits and 17 blends derived from these pits were analyzed using a mix design tool to meet the same criteria developed by Paige-Green [12]. The tests conducted during this project consisted of Particle-Size Distribution, CBR, and Atterberg tests. The results of this research confirmed the benefits of blending gravels to meet target specifications for wearing course performance based on desktop analysis. However, this project did not include the lab testing of blended materials to validate the desktop analysis.

Philip [13] also studied surface deterioration and sustainability modeling for unsealed roads. This study concluded that surface materials with a plasticity index above six have a 20% slower rate of deterioration. This finding supports the previous research studies for target specifications and the Central Goldfields study [14], in which the best-performing trial site was composed of materials with a 5% clay addition for increasing plasticity.

Information and knowledge gained through this literature review have generally complemented each other regarding the specifications to follow when modifying gravel for use on an unsealed wearing course. Hanson [9], Smith [11], Philip [13], and Yin [10] all made direct reference to work carried out by Paige-Green [12] and the performance-related specification developed by Jones and Green [6]. While some studies did not show the final testing data, the results recommended the addition of clay to increase plasticity and concluded the importance of particle size distribution.

A testing method that was not evident while reviewing previous studies was testing the material for the capillary rise. The capillary rise test assesses the potential for moisture absorption of pavement materials from external water sources [15]. Since moisture ingress is a crucial contributor to the breakdown and failure of wearing courses, water absorption potential could help provide insight into why certain materials behave differently and aid in material selection.

This paper provides a concise overview of various stabilization methods, emphasizing the significant role of gravel material specifications in the performance of wearing courses. The research focuses on evaluating the characteristics of ten gravel pits and three corresponding blends to determine their suitability as wearing course materials for unsealed roads in the Goondiwindi Regional Council area in Southwest Queensland. The outcomes of laboratory investigations are utilized to develop a material analysis tool, which recommends optimal mixing ratios to enhance the performance of wearing course materials according to target specifications. It is anticipated that this analysis tool will support the justification of field trials to further assess the feasibility of the granular stabilization method.

Importantly, the issues associated with unsealed wearing course materials are not unique to this particular region, state, or country. Geotechnical properties of gravel-wearing courses have been a subject of interest among researchers worldwide since the early 1900s.

1.1 Gravel Material Specifications

The whole purpose of improving the quality of materials on unsealed roads is to reduce the defects that arise from marginal materials. Performance and deterioration of gravel-wearing courses are influenced by interrelated factors, including material properties, road geometry, pavement design drainage, environmental factors, and traffic conditions. Common defects are usually identified by potholes, a slippery surface, excessive dust, raveling, corrugating, and rutting [16].

Of the above factors, traffic conditions and environmental factors are out of our control; however, we can design the pavements and construct the roads and drainage to help deal with these factors and reduce the impact of the associated defects. Efforts should focus on optimizing the quality of materials used for road construction; otherwise, any additional effort through these other methods seems futile.

The wearing course is the first layer of contact between the pavement and the traffic loading. Therefore, wearing course materials for unsealed roads needs to be well-graded materials with higher plasticity and fines percentage than sealed roads. Using high-quality materials for wearing course reduces the frequency of re-gravelling on unsealed roads. To anticipate the performance of gravel-wearing course materials, Austroads [5] uses a chart developed by Jones and Green [6] illustrating the relationship between Shrinkage Product (SP) and Grading Coefficient (GC) and expected performance and failure modes of materials outside the recommended specification (Fig. 1).

Fig. 1
figure 1

Relationship between shrinkage product, grading coefficient, and performance of wearing course gravels [5] based on South Africa Department of Transport, 2009

Full size image

Figure 1 shows the predicted performance for a material based on the grading coefficient (GC) and the shrinkage product (SP), which results from lab-tested material and following equations.

$${S}_{P}={L}_{S}\cdot {P}_{0.425}$$
(1)
$${G}_{C}=\frac{\left({P}_{26.5}-{P}_{2.0}\right){P}_{4.75}}{100},$$
(2)

where \({L}_{S}\) is linear shrinkage, and \({P}_{26.5}\), \({P}_{4.75}\), \({P}_{2.0}\), and \({P}_{0.425}\) are percentages passing 26.5, 4.75, 2.0, and 0.425 mm sieves, respectively.

While the grading coefficient in this specification focuses on the percentage of materials passing the 26.5 mm, 4.75 mm, and 2.0 mm sieves, Austroads provides a grading envelope incorporating the full range of sieve sizes from 100 mm to 0.01 mm, as illustrated in Fig. 2.

Fig. 2
figure 2

Grading envelope for the unsealed wearing course [5]

Full size image

According to Austroads, the particle-size distribution for gravel-wearing course shall comply with the requirements in Table 2 and Fig. 2.

Table 2 Requirement for particle-size distribution, gravel-wearing course [5]
Full size table

Adequate grading and sufficient plasticity of materials ensure an impermeable, stable, and robust wearing course to appropriately withstand traffic loads and detrimental environmental conditions. Typical properties of wearing course materials in unsealed roads, as specified in Austroads [5], are summarized in Table 3.

Table 3 Typical properties for unsealed road wearing course [5]
Full size table

These specifications provide an optimum grading envelope and plasticity index for material, and this research focus will be to achieve those optimum specifications through blending gravels.

As discussed earlier, optimizing the PSD and plasticity properties will improve the load-bearing capacity performance through increased mechanical interlock, increased inter-particle friction, and increased resistance to weakening by water ingress.

The strength of a granular material is derived solely from the mechanical interlock of particles throughout the particle-size spectrum. While maximum frictional strength does not necessarily coincide with maximum density, achieving high density will generally provide high friction. This is often referred to as the ‘maximum density PSD principle’ in which successively smaller sized particles fit into the remaining void space, thereby reducing the void space to a minimum (maximizing density). The figure below provides broad guidance on appropriate and inappropriate uses of granular pavement materials with various PSDs under different traffic conditions [8].

Figure 3 emphasizes the importance of particle size in ensuring pavement strength, as different traffic loads require different PSDs.

Fig. 3
figure 3

Relationship between the performance of gravel materials and traffic conditions [5, 17]

Full size image

In addition to particle-size distribution and plasticity, the California Bearing Ratio (CBR) plays a vital role in the performance of wearing course materials. The CBR value is a measure of a material’s strength. The CBR test is designed to test cohesive materials less than 19 mm (PCTE 2020). This test can be carried out on a soaked or unsoaked specimen, depending on the requirements of the guideline. When reviewing the requirements of the California Bearing Ratio on wearing courses, the minimum soaked value varied from an upper limit of 40 [5] to a lower limit of 15 (South Africa 2009 as cited in [5]. It should be noted that a lower value of 15 is recommended for rural roads, while a minimum value of 40 is recommended for haul roads.

1.2 Study Area

Goondiwindi Regional Council is a local government area in Southwest Queensland, with an approximate population of 10,700 over 19,294 km2. The unsealed road network within the Goondiwindi Regional Council area consists of almost 2000 km of gravel roads. These roads require regular maintenance to accommodate the traffic volumes of residential users and the commercial industry in all weather conditions. This maintenance must be carried out within the constraints of allocated budgets and to a standard that allows users to travel at designated speeds safely.

The general construction of these gravel roads consists of two layers, with the subgrade layer being the natural soil existing at that location and the base-course/wearing surface being an imported gravel layer with a compacted thickness of 150 mm to 200 mm.

Day-to-day maintenance of these roads is carried out by conducting a light formation grade, which removes minor surface defects to reduce roughness from wear due to trafficking and weather.

Over time, the weather, trafficking, and grading result in a reduction of this wearing course thickness to a point where a re-sheet of gravel is required to prevent exposure of the subgrade material and a road surface that does not meet the requirements for road users under all weather conditions.

Current practices of re-sheeting at the council involve gravel selection from a nominated borrow pit which is primarily governed by the proximity of this pit to the road segment undergoing re-sheeting. Material is then ripped and pushed by a dozer and carted onto the roadway with trucks, where it is prepared for laying. This process involves creating a layer of gravel section by section using a grader and making numerous passes with a suitable roller to break the materials up to the desired level. During this process, water is incorporated to ensure good compaction at the laying stage.

When road conditions deteriorate sooner than expected and maintenance intervals exceed what is practical, it becomes more difficult to justify using marginal gravels due to their proximity to the site. An increase in construction cost is usually the justification for these problems, with the long carts associated with sourcing better gravels being cost-prohibitive. A far better rationale for the materials would be underpinned through the research, testing, analysis of available materials, and conducting on-site trials to prove or disprove the study. If it can be proven that there are processes available that would result in reductions in maintenance costs and an increase in pavement life, then provided the costs associated with these processes are offset by the decrease in maintenance and improved pavement, it would be a complex case to argue.

Goondiwindi Regional Council currently uses several gravel pits throughout the shire to maintain their unsealed road network. While the current practices attempt to make the best use of this material through proper construction methods, there is still an acceptance that performance will suffer due to the nature of marginal materials being placed onto roads. This will emphasize the necessity of exploring an effective method of gravel materials improvement and developing an analysis tool to optimize the available gravel materials properties.

It should be noted that the issues with gravel sources and unsealed wearing courses discussed previously are not restricted to Goondiwindi Regional Council alone, and the development of this analysis tool could be applied to any organization worldwide with similar issues wishing to improve the quality of material available to them.

2 Materials

In this research, gravel materials were sourced from identified gravel pits (Table 4) following Australian methods for sampling and testing aggregates (AS1141.3.1).

Table 4 Evaluated gravel pits
Full size table

The location of the evaluated gravel pits is shown in Fig. 4.

Fig. 4
figure 4

Location of studied gravel pits

Full size image

A series of tests were conducted on the materials from the identified gravel pits to analyze the gravel materials. Based on these test results, a desktop analysis tool was developed to analyze the data from individual pits and propose the most appropriate blends of gravel pits that follow the Australian specifications for unsealed pavements. The blends were tested through laboratory investigation to validate the analysis tool.

3 Results and Discussion

To determine the suitability of studied gravel pits as an unsealed wearing course material using the specifications previously mentioned during the literature review by Jones and Green [6] and Austroads [5], the essential properties of gravel samples from different pits were assessed. These samples were collected from windrows after crushing before the final compaction. This was done to provide the best representation of the material. Several tests were conducted to assess the material against target values for PSD, CBR, Atterberg Limits, and Gravel Performance Indication derived from Austroads [5] ‘Guide to Pavement Technology Part 6—Unsealed Roads’ and Paige-Green [18] ‘New Perspectives of Unsealed Roads in South Africa’. Capillary rise testing was also conducted to provide an indication of the materials’ ability to draw up water through capillary action when free water is available.

3.1 Particle-Size Distribution

All pits’ particle-size distribution was assessed against the Australian grading envelope for unbound wearing course materials. As discussed earlier, PSD indicates the strength, so the proper PSD provides strength through particle interlock and increased density. The results of the PSD test as per Q103A (QLD specifications) for all pits are presented in Fig. 5.

Fig. 5
figure 5

Particle-size distribution test results of studied gravel pits

Full size image

As can be observed in Fig. 5, most pits except Pit 6 and Pit 10 do not meet the PSD requirements for the unsealed wearing course. However, for some pits, such as Pit 3, the particle-size distribution shows a grading meeting all requirements of the Austroads envelope except for the fines passing the 0.075 mm sieve. Therefore, if assessed on PSD alone, these pits may be accepted with the condition that it has sufficient plasticity for an unsealed wearing course.

The particle-size distribution of other gravel pits, such as Pit 5 and Pit 7, shows a grading just under the Austroads lower limit for sieve sizes above a specific size. Further works through increasing the quantities passing some upper sieve sizes may place these gravel pits in the accepted category. In adjusting the gradation curve for these pits, caution would need to be taken, so that the additional work does not result in putting the sieve sizes of 2.36 mm and below beyond their upper limits.

For Pit 4, the grading sits just under the lower limits for sieve sizes above 2.36 mm. Increasing the materials passing the upper sieve sizes may make this gravel pit acceptable for PSD requirements; however, because Pit 4 gravel is river gravel, the aggregates are very hard and difficult to break.

Among these gravel pits, the particle-size distribution of Pit 2 shows a grading very similar in shape to that suggested by Austroads, with it uniformly falling just under the lower limit. Therefore, this gravel pit may be accepted with the condition that it is further worked to increase the quantities passing each sieve and eliminate the small percentage retained on the 37.5 mm sieve.

Due to these challenges, the blends were prepared, and the different blends’ PSD was evaluated, which will be discussed in the following sections.

3.2 CBR

Samples were tested for both soaked and unsoaked CBR values as per Q113A (QLD specifications). This test method was used to obtain soaked CBR data for comparison with target values of 15% (min) as specified in Austroads [5]. The results of the CBR test are presented in Table 5.

Table 5 CBR test results
Full size table

As shown in Table 5, Pit 1 CBR values are expected of a sandy material and easily exceed the literature-specified minimum value of 15. However, the CBR values for Pit 2, Pit 5, and Pit 6 are low and do not meet the Austroads standard. Based on this, they should be avoided as an unsealed wearing course.

CBR values of Pit 3, Pit 8, and Pit 10 exceed those required by Austroads and would be well suited for an unsealed wearing course.

The CBR values of Pit 4 and Pit 7 just pass the required CBR of 15%, and while it could still be accepted in this form, it would be deemed marginal and not preferred.

3.3 Atterberg Limits

QLD Main Roads testing specifications of Q104, Q105, and Q106 were used to evaluate the Atterberg limits of gravel materials from different pits. These tests were then used to calculate the plasticity index and weighted plasticity index, as presented in Table 6.

Table 6 Atterberg limit test results
Full size table

As discussed earlier, average annual rainfall needs to be considered when assessing the plasticity index or weighted plasticity index of material, because the maximum values of these parameters vary for high and low rainfall areas (Table 3). The average annual rainfall for Goondiwindi is 624 mm [19]. This value exceeds the 500 mm criteria for low rainfall. For reference, the average annual rainfall for Brisbane on Queensland’s south coast is 1200 mm [19], so while Goondiwindi’s rainfall exceeds 500 mm, this will be considered if tested values only marginally exceed this criterion.

Due to the sandy nature of the material of Pit 1, PI and Linear shrinkage, as expected, were both low. For Pit 2, Plasticity Index and Weighted Plasticity Index are close to the limits for regions with high rainfall; however, as the Goondiwindi region does not exceed the 500 mm criteria by a much, these figures of PI and WPI should be considered on the low side.

For Pit 3, Pit 8, and Pit 10, the Plasticity Index is low. Low PI combined with a low percentage of fines results in an even lower Weighted Plasticity Index. With this additional Atterberg limit information, the use of this material may only be deemed suitable with the addition of clayey fines.

For Pit 4, Plasticity Index and Weighted Plasticity Index are beyond the higher limits. Therefore, they would require the addition of a sandy fine to bring the Atterberg limits into specification.

For Pit 5, Plasticity Index and Weighted Plasticity Index are within limits for an area with less than 500 mm annual rainfall. However, PI only just exceeds the limits for areas exceeding 500 mm annual rainfall, while WPI is at the higher end of this threshold.

For Pit 6, Pit 7, and Pit 9, Plasticity Index and Weighted Plasticity Index are both within acceptable limits for an area that slightly exceeds the 500 mm annual rainfall threshold. However, Pit 9 WPI is above the permitted limit.

3.4 Gravel Performance Indication

The results of Atterberg limit tests were then further used to evaluate the performance of gravel materials as per the Performance Indication Chart, as shown in Fig. 6.

Fig. 6
figure 6

Performance indication of studied gravel pits

Full size image

The linear shrinkage values were used to calculate the shrinkage product and then evaluate the performance of gravel materials. Based on the performance indication chart in Fig. 6, the materials from Pit 1, Pit 2, Pit 3, Pit 8, and Pit 10 pits are unsuitable for use as an unsealed wearing course.

The lack of clayey material and poor grading would lead to corrugation and raveling under traffic. However, the acceptable CBR values of Pit 1 and Pit 3 may make them good options for blending with clayey gravel that has a low CBR.

The performance indication chart shows that Pit 4 materials are suitable for use as an unsealed wearing course. However, the only concern would be the marginal CBR results of these materials.

Pit 5, Pit 6, Pit 7, and Pit 9 all sit in the E Section of the performance indication chart, indicating that these materials are suitable for use as an unsealed wearing course. However, all these materials have unsatisfactory CBR values, while CBR is an important parameter when investigating material suitability. For Pit 7, further work and the addition of a clayey material are required to improve the shrinkage product.

3.5 Capillary Rise

Capillary rise testing was also conducted to provide an indication of the materials’ ability to draw up water through capillary action when free water is available. The results of the capillary rise test are presented in Table 7. This research also investigated the relationship between the capillary rise and PSD or CBR values.

Table 7 Capillary rise test results
Full size table

Capillary rise for Pit 2 material was very high, with a rise through 100% of the sample in just 4 h. Therefore, using this material would not be recommended in areas prone to hold water. For Pit 3 material, the capillary rise was relatively high, with a rise of 83% in 72.5 h. However, this material exhibited high strength-soaked CBR results and would not affect this moisture ingress.

Capillary rise for Pit 4 material was low, with a rise of 29% in 72.5 h. This was expected with the higher value shrinkage product providing a level of protection from moisture.

Capillary rise for Pit 5 material and Pit 6 material had a rise of 49% and 41% in 72.5 h, respectively. In addition, the capillary rise for Pit 7 material rose 50% in 72.5 h. It should be noted that the shrinkage product was relatively high for all these three pits.

The capillary rise results were used to investigate any correlation with the other target specifications. Results show no noticeable relationship with CBR and grading coefficient; however, there was an apparent relationship with shrinkage product. This relationship can give confidence that if the material has a higher shrinkage product, it will likely not be as susceptible to capillary rise. This information is essential if a material has an acceptable unsoaked CBR value and an unacceptable soaked CBR value. The materials with a high capillary rise level will have a reduced CBR when soaked.

3.6 Comparison of Test Results and Desktop Analysis Tool Outcome

Based on the test results and data from the available guidelines, this research has developed an analysis tool for gravel materials’ evaluation, which aims to predict the best blend of gravel materials considering various essential parameters, such as PSD, CBR values, and Atterberg limits.

Since verifying any technical tool is essential in making engineering predictions with quantified confidence, this study has conducted verification to validate desktop analysis tool results. To conduct the verification process, three gravel blends were designed based on the results of tests on the individual gravel samples and Australian requirements to predict the blend performance incorporating this desktop analysis tool.

Table 8 summarizes the results of initial tests on individual pits. It should be noted that the proximity of pit locations was not considered when choosing materials to blend. While this would be a focal point of consideration in the real world, for this paper, the intent was to test the accuracy of the desktop analysis tool. This section discusses the results of blend evaluations.

Table 8 Summary of test results for individual pits
Full size table

The test results confirmed that the materials from Pit 1, Pit 3, Pit 4, and Pit 6 could be used for road construction. Therefore, upon initial testing and incorporating results into the analysis tool, three blends were chosen to be lab-tested for further assessment against Austroads target specifications, as follows:

  • Blend 1- Pit 3 (50%) and Pit 4 (50%)

  • Blend 2- Pit 1 (50%) and Pit 4 (50%)

  • Blend 2- Pit 1 (25%) and Pit 6 (75%).

The blends were prepared and evaluated for PSD, CBR, Atterberg limit, and performance indicators. Figures 7, 8 and 9 show the particle distribution curve of these blends compared to Australian limits and the original pits gradation curves.

Fig. 7
figure 7

Gradation curve for blend 1 (50% Pit 3–50% Pit 4)

Full size image
Fig. 8
figure 8

Gradation curve for blend 2 (50% Pit 1–50% Pit 4)

Full size image
Fig. 9
figure 9

Gradation curve for blend 3 (25% Pit 1–75% Pit 6)

Full size image

As shown in Fig. 7, although Pit 4 gradation was not suitable for road construction, its blend with Pit 3 at this certain ratio can provide a good blend in particle distributions. The same observation was noticed for blends 2 (50% Pit 1 and 50% Pit 4) and Blend 3 (25% Pit 1 and 75% Pit 6).

In addition, as can be seen in Figs. 7, 8 and 9, the PSD test results on blends confirmed the desktop analysis tool for all three blends.

A comparison of CBR values and Atterberg limits for these blends are also presented in Table 9.

Table 9 Test results of blends
Full size table

From Fig. 8 and Table 9 and using the analysis tool, it was shown that the blend of materials from Pit 3 (50%) and Pit 4 (50%) conforms to Australian specifications for unsealed pavements, which offers an agreement with laboratory testing of Blend 1. However, blends 2 and 3 did not meet the Australian requirements for unsealed pavements for soaked CBR values.

This was unexpected and further investigation as to why this may have occurred exposed potential contamination of material that may have produced this result. The initial material samples from Pit 1 were taken from a windrow during construction and could have contained an unknown quantity of the existing wearing course material. The material used during the blending process was sampled from the stockpile, leaving the possibility for error with the two samples not being representative of one and other.

Using the Atterberg limits, the performance of the blends was also evaluated. The performance indication chart is presented in Fig. 10.

Fig. 10
figure 10

Performance indication of blends

Full size image

From the performance indication chart, it can also be seen that the low linear shrinkage of blends 2 and 3 has affected the performance of these blends.

4 Conclusion

Granular stabilization emerges as an economically favorable method compared to chemical and mechanical stabilization in many cases. This study specifically focused on investigating granular stabilization by examining samples from ten gravel pits in collaboration with the Goondiwindi Regional Council during their annual re-sheeting program. These samples were subjected to testing to meet the target specifications of Grading Coefficient, Shrinkage Product, and CBR based on Queensland Transport and Main Roads testing methods.

Furthermore, this research successfully developed a desktop blending tool utilizing Excel and incorporating a set of target specifications derived from the Austroads Guide to Pavement Technology—Unsealed Pavements [5]. The tool aims to assist decision-makers in evaluating the suitability of gravel materials for use as unsealed wearing courses, thereby reducing the need for additional testing while facilitating the creation of optimal gravel blends that meet road requirements.

To verify the accuracy of the desktop tool, three blend samples were selected for granular stabilization and subsequent laboratory testing. The results indicated a close resemblance between the grading coefficients obtained from the laboratory testing and those generated by the desktop blend in all three cases. However, the results for CBR values and shrinkage products were inconclusive, potentially due to the misrepresentation of the Pit 1 sample, which requires further testing and verification. Based on these findings, if users are content with the shrinkage products and CBR values of two samples and only intend to use the tool to assess the potential for blending to achieve target grading coefficients, it can be deemed suitable.

Additionally, comparisons between soaked and unsoaked CBR values emphasized the significance of considering adequate drainage when selecting pavement materials. While Austroads employs soaked CBR values for its target specification, the use of unsoaked CBR values is reasonable in the absence of water ponding and with sufficient drainage.

It is essential to note that the gravel source and unsealed wearing course challenges discussed in this paper extend beyond the Goondiwindi Regional Council, and the developed analysis tool can be applied to any organization worldwide facing similar issues and seeking to enhance the quality of available materials.