The Influence of Aeolian Sand on the Anti-Skid Characteristics of Asphalt Pavement

The influence of sand accumulation on the skid resistance of asphalt pavement was studied. Many scholars have researched the anti-skid performance of conventional asphalt pavements. However, there is a lack of research on the anti-skid performance of desert roads under the condition of sand accumulation. In this study, AC-13 and AC-16 asphalt mixtures were used. The British Pendulum Number (BPN) under different sand accumulations was measured with a pendulum friction coefficient meter, and the Ames engineering texture scanner was used to obtain different sand accumulations. The texture index of asphalt mixture was used to study the macro and micro texture of asphalt pavement under different amounts of sand accumulation, and the degree of influence of different particle sizes on BPN was obtained through gray correlation analysis. The test results show that the presence of aeolian sand has a significant impact on the macro and micro texture of the asphalt pavement and will cause the anti-skid performance to decrease. Moreover, there is an apparent positive linear correlation between the road surface texture index and BPN. The research results may provide reference and reference for the design and maintenance of desert highways.


Introduction
The anti-skidding of the pavement is of great significance to traffic safety [1]. The antiskid performance of the road surface depends on the macro-structure and micro-texture of the road surface, which provides sufficient adhesion for vehicle driving [2,3]. The lithology and angular characteristics of aggregate minerals affect the texture of the road surface [4][5][6]. Specifically, the lithology of the aggregate determines the resistance to abrasion of the road surface under traffic loads, and the angularity affects the micro/macro texture of the road surface. Different rocks have different resistance to abrasion.
The skid resistance of pavement surface is mainly controlled by microstructure and macrostructure, in which microstructure mainly affects the friction at low speed, while macro structure mainly affects the friction at high speed. Many scholars have done relevant research and discussion on the specific quantitative relationship between these two structures and road surface friction coefficient. Friel et al. [7] obtained the surface texture image of coarse aggregate through the scanning electron microscope, analyzed its digital image and fractal dimension by computer, and obtained the influence of micro-texture on pavement skid resistance. Serigost [8] collected many texture data and friction coefficient of Texas highway surface, compared and regressed them, obtained the relationship between pavement friction coefficient and microtexture, and realized the purpose of predicting BPN and friction coefficient with pavement microtexture. Based on the comprehensive analysis of surface characteristics such as friction, texture, and anti-sliding performance In this paper, two types of asphalt mixtures were prepared, AC-13 and AC-16. The aggregates are both conglomerates and have the same specific gravity. The passing rate of each sieve is shown in Figure 1. The optimal binder content of AC-13 is 4.7%, the air voids content 4.3%, the optimal binder content of AC- 16 4.2%, and the air voids content is 4.3%.
Based on the Test Methods of Aggregate for Highway Engineering [31], ties of aggregate were tested, and their results are given in Table 1. The aggreg this study is a conglomerate from Xinjiang of China. In this paper, two types of asphalt mixtures were prepared, AC-13 and aggregates are both conglomerates and have the same specific gravity. The p of each sieve is shown in Figure 1. The optimal binder content of AC-13 is 4.7%, t content 4.3%, the optimal binder content of AC- 16 4.2%, and the air voids content The sand sample comes from aeolian sand-covered on the road surface in area of Xinjiang. In order to explore the different particle size content and m morphology of aeolian sand, the particle size analysis of aeolian sand was carri the particle size distribution of the sample was tested with a laser particle siz (instrument model: Master-sizer 2000) (Malvern Instruments Ltd., Worcester The results are shown in Figure 2 by using the field emission scanning electron (Quanta FEG 250) (FEI Co., Hillsboro, OR, USA). The morphology of aeolian different magnifications is shown in Figure 3.  The sand sample comes from aeolian sand-covered on the road surface in the desert area of Xinjiang. In order to explore the different particle size content and microscopic morphology of aeolian sand, the particle size analysis of aeolian sand was carried out, and the particle size distribution of the sample was tested with a laser particle size analyzer (instrument model: Master-sizer 2000) (Malvern Instruments Ltd., Worcestershire, UK). The results are shown in Figure 2 by using the field emission scanning electron microscope (Quanta FEG 250) (FEI Co., Hillsboro, OR, USA). The morphology of aeolian sand under different magnifications is shown in Figure 3.
this study is a conglomerate from Xinjiang of China. In this paper, two types of asphalt mixtures were prepared, AC-13 and aggregates are both conglomerates and have the same specific gravity. The p of each sieve is shown in Figure 1. The optimal binder content of AC-13 is 4.7%, t content 4.3%, the optimal binder content of AC-16 4.2%, and the air voids content The sand sample comes from aeolian sand-covered on the road surface in area of Xinjiang. In order to explore the different particle size content and m morphology of aeolian sand, the particle size analysis of aeolian sand was carri the particle size distribution of the sample was tested with a laser particle si (instrument model: Master-sizer 2000) (Malvern Instruments Ltd., Worcester The results are shown in Figure 2 by using the field emission scanning electron (Quanta FEG 250) (FEI Co., Hillsboro, OR, USA). The morphology of aeolian different magnifications is shown in Figure 3.   According to the results obtained by the laser particle size analyzer, the m particle size of aeolian sand is 187 μm, and the minimum is 55 μm. Among th particle size distribution of aeolian sand between 94.6-121 μm accounts for the m nificant proportion, which is 52.8%. From Figure 3, it can be observed that the sand particles are relatively granular, which can provide a rolling medium for th friction between the wheel and the road surface and reduce the anti-skid perform the road surface when it exists on the road surface.

Test Method
The pendulum friction coefficient instrument was used to measure the Brit dulum Number (BPN) of the surface of the laboratory compacted slab (300 mm × × 50 mm) intended for wheel track rutting test with different sand accumulation In order to study the impact of sand accumulation on the anti-skid perform the pavement, a rut plate specimen was made indoors, and the sand on the road simulated through different amounts of sand, and the BPN value under differen tions was tested, as shown in Figure 4. Moreover, the different amounts of sand lation were 0 g, 10 g, 20 g, 30 g, 40 g, 50 g, 60 g, 70 g, and 80 g, respectively. Before clean the surface of the test piece with a brush and then spread the sand weighe vance evenly over the whole test piece. After the test, clean the sand on the surfa test piece with a brush and repeat this method many times until the test was com  According to the results obtained by the laser particle size analyzer, the maximum particle size of aeolian sand is 187 µm, and the minimum is 55 µm. Among them, the particle size distribution of aeolian sand between 94.6-121 µm accounts for the most significant proportion, which is 52.8%. From Figure 3, it can be observed that the aeolian sand particles are relatively granular, which can provide a rolling medium for the rolling friction between the wheel and the road surface and reduce the anti-skid performance of the road surface when it exists on the road surface.

Test Method
The pendulum friction coefficient instrument was used to measure the British Pendulum Number (BPN) of the surface of the laboratory compacted slab (300 mm × 300 mm × 50 mm) intended for wheel track rutting test with different sand accumulation [32].
In order to study the impact of sand accumulation on the anti-skid performance of the pavement, a rut plate specimen was made indoors, and the sand on the road area was simulated through different amounts of sand, and the BPN value under different conditions was tested, as shown in Figure 4. Moreover, the different amounts of sand accumulation were 0 g, 10 g, 20 g, 30 g, 40 g, 50 g, 60 g, 70 g, and 80 g, respectively. Before the test, clean the surface of the test piece with a brush and then spread the sand weighed in advance evenly over the whole test piece. After the test, clean the sand on the surface of the test piece with a brush and repeat this method many times until the test was completed.  According to the results obtained by the laser particle size analyzer, the maxim particle size of aeolian sand is 187 μm, and the minimum is 55 μm. Among them particle size distribution of aeolian sand between 94.6-121 μm accounts for the most nificant proportion, which is 52.8%. From Figure 3, it can be observed that the aeo sand particles are relatively granular, which can provide a rolling medium for the ro friction between the wheel and the road surface and reduce the anti-skid performan the road surface when it exists on the road surface.

Test Method
The pendulum friction coefficient instrument was used to measure the British dulum Number (BPN) of the surface of the laboratory compacted slab (300 mm × 300 × 50 mm) intended for wheel track rutting test with different sand accumulation [32] In order to study the impact of sand accumulation on the anti-skid performanc the pavement, a rut plate specimen was made indoors, and the sand on the road area simulated through different amounts of sand, and the BPN value under different co tions was tested, as shown in Figure 4. Moreover, the different amounts of sand accu lation were 0 g, 10 g, 20 g, 30 g, 40 g, 50 g, 60 g, 70 g, and 80 g, respectively. Before the clean the surface of the test piece with a brush and then spread the sand weighed in vance evenly over the whole test piece. After the test, clean the sand on the surface o test piece with a brush and repeat this method many times until the test was complet  As shown in Figures 5 and 6, we prepared aeolian sand samples with moisture tents of 5% and 10%, respectively, and used the pendulum friction coefficient met determine the BPN value under different sand accumulation in order to study the in ence of the moisture content of aeolian sand on the anti-sliding performance. As shown in Figures 5 and 6, we prepared aeolian sand samples with moisture contents of 5% and 10%, respectively, and used the pendulum friction coefficient meter to determine the BPN value under different sand accumulation in order to study the influence of the moisture content of aeolian sand on the anti-sliding performance. (a) (b) Figure 6. Specimens with a water content of 5% (a) and 10% (b) respectively after adding aeolian sand.
In order to study the influence of different particle diameters of aeolian sand on the anti-skid performance of two types of asphalt mixtures, the aeolian sand samples were first screened. The particle size distribution of aeolian sand was mainly divided into three ranges, namely less than 0.075 mm, 0.075-0.15 mm, and greater than 0.15 mm, as shown in Figure 7. The mass percentages are 12.5%, 83.6%, and 3.9%, respectively. Then, the BPN values of the three particle sizes of aeolian sand were tested separately on the road surface. The Ames engineering texture scanner (LTS 9400) (Ames Engineering Co., Ames, IA, USA) was used to detect the texture indicators of the specimens under different amounts (a) (b) Figure 6. Specimens with a water content of 5% (a) and 10% (b) respectively after adding aeolian sand.
In order to study the influence of different particle diameters of aeolian sand on the anti-skid performance of two types of asphalt mixtures, the aeolian sand samples were first screened. The particle size distribution of aeolian sand was mainly divided into three ranges, namely less than 0.075 mm, 0.075-0.15 mm, and greater than 0.15 mm, as shown in Figure 7. The mass percentages are 12.5%, 83.6%, and 3.9%, respectively. Then, the BPN values of the three particle sizes of aeolian sand were tested separately on the road surface. The Ames engineering texture scanner (LTS 9400) (Ames Engineering Co., Ames, IA, USA) was used to detect the texture indicators of the specimens under different amounts In order to study the influence of different particle diameters of aeolian sand on the anti-skid performance of two types of asphalt mixtures, the aeolian sand samples were first screened. The particle size distribution of aeolian sand was mainly divided into three ranges, namely less than 0.075 mm, 0.075-0.15 mm, and greater than 0.15 mm, as shown in Figure 7. The mass percentages are 12.5%, 83.6%, and 3.9%, respectively. Then, the BPN values of the three particle sizes of aeolian sand were tested separately on the road surface. (a) (b) Figure 6. Specimens with a water content of 5% (a) and 10% (b) respectively after adding aeolian sand.
In order to study the influence of different particle diameters of aeolian sand on the anti-skid performance of two types of asphalt mixtures, the aeolian sand samples were first screened. The particle size distribution of aeolian sand was mainly divided into three ranges, namely less than 0.075 mm, 0.075-0.15 mm, and greater than 0.15 mm, as shown in Figure 7. The mass percentages are 12.5%, 83.6%, and 3.9%, respectively. Then, the BPN values of the three particle sizes of aeolian sand were tested separately on the road surface. The Ames engineering texture scanner (LTS 9400) (Ames Engineering Co., Ames, IA, USA) was used to detect the texture indicators of the specimens under different amounts of sand. As shown in Figure 8, select 30 scan lines among them, and select the three texture indicators of MPD, Ra, and Rq. Among them, MPD is the mean profile depth of the The Ames engineering texture scanner (LTS 9400) (Ames Engineering Co., Ames, IA, USA) was used to detect the texture indicators of the specimens under different amounts of sand. As shown in Figure 8, select 30 scan lines among them, and select the three texture indicators of MPD, Ra, and Rq. Among them, MPD is the mean profile depth of the specimen surface, Ra is the average deviation of the contour arithmetic, and Rq is the root mean square deviation of the profile.

BPN Test
As shown in Figure 9, the BPN values of the two types of specimens change unde different amounts of sand. The amount of sand is inversely proportional to the BPN valu on the surface of the specimen. The BPN value decreases with the increase in the amoun of sand. The BPN values of the two asphalt mixture types show the same change law which can be divided into three stages. In increasing the initial sand volume from 0 to 1 g, the BPN value decreased slowly. The BPN value of AC-13 decreased by 4.1% and 3.8% and the BPN value of AC-16 decreased by 2.3% and 1.7%, respectively. Then, when th sediment volume increased from 10 g to 50 g, the decreasing range of BPN value of th two mixtures increased obviously, in which the BPN value of AC-13 decreased by 27.5% and 27.6%, and the BPN value of AC-16 decreased by 31% and 30.2%. In the later stage when the amount of sedimentation reached 60 g, the BPN value decreased slowly, an the final change curve gradually became flat. The BPN value of the two mixtures showe a slight rebound, of which AC-13 rebounded by 0.6% and 0.7%, and AC-16 rebounde 0.9% and 1.1%.

BPN Test
As shown in Figure 9, the BPN values of the two types of specimens change under different amounts of sand. The amount of sand is inversely proportional to the BPN value on the surface of the specimen. The BPN value decreases with the increase in the amount of sand. The BPN values of the two asphalt mixture types show the same change law, which can be divided into three stages. In increasing the initial sand volume from 0 to 10 g, the BPN value decreased slowly. The BPN value of AC-13 decreased by 4.1% and 3.8%, and the BPN value of AC-16 decreased by 2.3% and 1.7%, respectively. Then, when the sediment volume increased from 10 g to 50 g, the decreasing range of BPN value of the two mixtures increased obviously, in which the BPN value of AC-13 decreased by 27.5% and 27.6%, and the BPN value of AC-16 decreased by 31% and 30.2%. In the later stage, when the amount of sedimentation reached 60 g, the BPN value decreased slowly, and the final change curve gradually became flat. The BPN value of the two mixtures showed a slight rebound, of which AC-13 rebounded by 0.6% and 0.7%, and AC-16 rebounded 0.9% and 1.1%.
Materials 2021, 14, x FOR PEER REVIEW 6 specimen surface, Ra is the average deviation of the contour arithmetic, and Rq is the mean square deviation of the profile.

BPN Test
As shown in Figure 9, the BPN values of the two types of specimens change u different amounts of sand. The amount of sand is inversely proportional to the BPN v on the surface of the specimen. The BPN value decreases with the increase in the am of sand. The BPN values of the two asphalt mixture types show the same change which can be divided into three stages. In increasing the initial sand volume from 0 g, the BPN value decreased slowly. The BPN value of AC-13 decreased by 4.1% and 3 and the BPN value of AC-16 decreased by 2.3% and 1.7%, respectively. Then, when sediment volume increased from 10 g to 50 g, the decreasing range of BPN value o two mixtures increased obviously, in which the BPN value of AC-13 decreased by 2 and 27.6%, and the BPN value of AC-16 decreased by 31% and 30.2%. In the later s when the amount of sedimentation reached 60 g, the BPN value decreased slowly, the final change curve gradually became flat. The BPN value of the two mixtures sho a slight rebound, of which AC-13 rebounded by 0.6% and 0.7%, and AC-16 reboun 0.9% and 1.1%.  The analysis believes that due to the generally small particle size of aeolian sand, it enters the gaps of the asphalt mixture at the initial stage and blocks the gaps, which reduces the depth of the structure and decreases the anti-skid performance. Then, as the quality of the aeolian sand increases, the gap is gradually filled, and part of the aeolian sand covers the road surface. Due to its round appearance, the tire and the road surface form a micro-bearing system, which changes sliding friction into rolling friction, and the friction force drops sharply. The main factor causing the decrease of friction is that the aeolian sand is not adhered to the adhesive asphalt, resulting in rolling friction under the impact force of the pendulum. Finally, when the amount of accumulated sand increases to a certain level (60 g), the friction is controlled by the accumulated sand. The accumulated sand produces resistance to the rubber slider of the pendulum friction coefficient meter, causing its BPN value to rise. The reason for this phenomenon may be the limitation of the test equipment, which does not conform to the actual situation.
As shown in Figure 10a, when the amount of aeolian sand with the water content of 5% increases to 10 g, the BPN values of the two mixtures decrease by 4.9% and 4.5%, respectively. At this time, the AC-13 decreases. The amplitude is more significant than AC-16. The analysis believes that the air voids content and texture depth of AC-13 is less than AC-16. Under the interference of water and sand, the decrease of BPN value is slightly more significant than AC-16. With the increase of sand accumulation, the anti-sliding performance of the two mixtures decreases and finally tends to be stable. During the whole process, the BPN values of the two mixtures decrease by 28.5% and 35.7%, respectively. When the water content of aeolian sand is 10%, as shown in Figure 10b, the initial BPN value of the two mixtures decreases overall. The initial value of AC-13 is significantly lower than AC-16. When the amount of sand increased to 30 g, the BPN value droped to the lowest point. The BPN value of AC-16 reached the lowest point when the amount of sand increased to 50 g. There was a slight rebound. Among them, the BPN value of AC-13 rebounded by 3.1% and 4.4%, and the BPN value of AC-16 rebounded. The margins were 1.9% and 2.4%. The analysis shows that water becomes the lubricant between the sand, which makes the micro-bearing system formed by the wheel sand road more significant. The wet sand exists between the wheel and the road like a ball, and the anti-sliding ability drops sharply at this time.
sand increased to 50 g. There was a slight rebound. Among them, the BPN value of AC-13 rebounded by 3.1% and 4.4%, and the BPN value of AC-16 rebounded. The margins were 1.9% and 2.4%. The analysis shows that water becomes the lubricant between the sand, which makes the micro-bearing system formed by the wheel sand road more significant. The wet sand exists between the wheel and the road like a ball, and the anti-sliding ability drops sharply at this time. Compared with the dry aeolian sand acting on the road surface, the aeolian sand in the wet state makes the BPN value of the two mixtures generally lower. At this time, the pavement voids and macrostructures are not only filled with sand but filled with sand and water. This process accelerates the decrease in the depth of the pavement structure, allowing more sand and water to act on the road surface and hinder the tires and the pavement. The direct contact between the sand and the moisture exists in the sand surface and the gaps, which provides sufficient lubrication between the sand and between the sand and the wheel and the road surface, so the BPN value can quickly drop to the lowest point. Figure 11a shows the variation law of BPN values of three particle sizes of AC-16 under different sediment volumes. On the whole, the variation trend of three particle sizes is the same as that of conventional aeolian sand samples. However, the slope of the BPN value curve for aeolian sand less than 0.075 mm in the initial stage is the smallest, while the curve greater than 0.15 mm has the most significant variation range, and the curve decreases rapidly in the whole process. In the end, there was a significant recovery. In increasing the amount of sand from 0 to 10 g, the BPN corresponding to the three particle sizes decreased by 4.1%, 5.0%, and 6.5%, respectively. When the amount of sand increased from 10 g to 50 g, the three particle sizes, the corresponding BPN dropped by 29.6%, 31.6%, and 30.2%, respectively. The overall decline in Figure 11b is smaller than that in Figure 11a, indicating that AC-16 is more disturbed by wind and sand. However, the slope of the BPN curve at the initial stage was less than that of AC-13, and three particle sizes decreased by 1.2%, 2.3%, and 3.4%, respectively. The analysis shows that the macro structure of AC-16 is larger than AC-13, and it can contain aeolian sand particles. Relatively strong, the decrease in BPN value is slight in the initial stage. Then, when the macro structure was filled with aeolian sand particles, the BPN value dropped sharply, and the three particle sizes dropped by 34.7%, 34.2%, and 35.7%, respectively. When the BPN value of the two mixtures fell to the lowest point in the later period, because the surface texture of AC-16 was slightly larger than AC-13, the BPN value of AC-16 rose slightly.
of AC-16 is larger than AC-13, and it can contain aeolian sand particles. Relatively strong, the decrease in BPN value is slight in the initial stage. Then, when the macro structure was filled with aeolian sand particles, the BPN value dropped sharply, and the three particle sizes dropped by 34.7%, 34.2%, and 35.7%, respectively. When the BPN value of the two mixtures fell to the lowest point in the later period, because the surface texture of AC-16 was slightly larger than AC-13, the BPN value of AC-16 rose slightly.
(a) (b) Figure 11. BPN value of aeolian sand with three particle sizes existing alone on the road surface. (a) BPN value of aeolian sand with three particle sizes existing alone on the AC-16 pavement surface; (b) BPN value of aeolian sand with three particle sizes existing alone on the AC-13 pavement surface.
The decrease of the BPN value on the surface of the test piece is related to the particle size. The larger the particle size, the more significant the decrease of the BPN value on the surface of the test piece. The results indicated that the larger the particle size of the aeolian sand, the greater the ability to fill the gap and the surface structure of the specimen. It reduced the macrostructure of the specimen surface and its anti-sliding ability. The BPN The decrease of the BPN value on the surface of the test piece is related to the particle size. The larger the particle size, the more significant the decrease of the BPN value on the surface of the test piece. The results indicated that the larger the particle size of the aeolian sand, the greater the ability to fill the gap and the surface structure of the specimen. It reduced the macrostructure of the specimen surface and its anti-sliding ability. The BPN value decreases more rapidly. Therefore, under the same filling rate, the larger the particle size of the aeolian sand, the faster the anti-skid performance of the pavement decreases.
In order to determine the degree of influence of the three particle sizes on the BPN value, the test results were analyzed with a gray correlation. The BPN value under the three particle sizes was used as the comparison sequence. The BPN value under the conventional aeolian sand was used as the reference sequence for comparative analysis. The results are shown in Figure 12.
Materials 2021, 14, x FOR PEER REVIEW value decreases more rapidly. Therefore, under the same filling rate, the larger the size of the aeolian sand, the faster the anti-skid performance of the pavement dec In order to determine the degree of influence of the three particle sizes on value, the test results were analyzed with a gray correlation. The BPN value un three particle sizes was used as the comparison sequence. The BPN value under ventional aeolian sand was used as the reference sequence for comparative analy results are shown in Figure 12. It can be seen from Figure 12 that the gray correlation coefficient of particles l 0.075 mm in the two kinds of asphalt mixture is the smallest, which means that the size has the least influence on the BPN value. In the conventional aeolian sand, the size less than 0.075 mm accounts for a relatively small proportion. In addition, its size is small, which mainly fills the gaps in the road surface and contributes litt It can be seen from Figure 12 that the gray correlation coefficient of particles less than 0.075 mm in the two kinds of asphalt mixture is the smallest, which means that the particle size has the least influence on the BPN value. In the conventional aeolian sand, the particle size less than 0.075 mm accounts for a relatively small proportion. In addition, its particle size is small, which mainly fills the gaps in the road surface and contributes little to the rolling medium between the tire and the road surface. The 0.075-0.15 mm particles have an essential impact on the BPN value. They had a large mass proportion, and acted as a rolling medium for the rolling friction between the tire and the road surface. Although the particles larger than 0.15 mm had the smallest proportion, they played a significant supporting role between the wheel and the road surface, disturbing the BPN value. The gray correlation coefficient corresponding to the particle larger than 0.15 mm was in the middle level.

Texture Index
The most intuitive impact of aeolian sand on the pavement is to change its original macroscopic structure. Meanwhile, smaller particles enter the pavement voids to fill the macroscopic structure and reduce the structure depth, thereby reducing the friction coefficient. On the other hand, larger particles with diameters covering the road surface hinder the direct contact between the tire and the road surface and reduce the effective contact area between the tire and the road surface, thereby reducing the anti-skid performance.
In order to further understand the road texture under different sand accumulation, the Ames engineering texture scanner (LTS 9400) was used to scan the surface of specimens under different sand accumulation. Thirty scanning lines were selected for each specimen to extract three texture indexes, namely MPD, Ra, and Rq. Figure 13a MPD reflects the average section depth of the specimen surface. With the increasing amount of aeolian sand, the texture depth of the specimen surface is gradually reduced, and the surface texture changes passively. At this time, MPD decreases with the trend, and the BPN value greatly correlates with the surface texture of the specimen, which also decreases. The MPD of AC-13 in Figure 13a and AC-16 in Figure 13b decreased by 23% and 7.2%, respectively, when the initial sand accumulation increased from 0 g to 5 g. Then, with the gradual increase of the amount of aeolian sand on the specimen surface, the MPD decreased. During the whole process, the MPD of the two specimens decreased by 52% and 56.4%, respectively.
Ra is the average deviation of the contour arithmetic and is used to characterize the texture of the road surface in the scanning range. It reflects the dispersion degree of the change of the pavement structure profile concerning the reference line. The increase of MPD reflects the average section depth of the specimen surface. With the increasing amount of aeolian sand, the texture depth of the specimen surface is gradually reduced, and the surface texture changes passively. At this time, MPD decreases with the trend, and the BPN value greatly correlates with the surface texture of the specimen, which also decreases. The MPD of AC-13 in Figure 13a and AC-16 in Figure 13b decreased by 23% and 7.2%, respectively, when the initial sand accumulation increased from 0 g to 5 g. Then, with the gradual increase of the amount of aeolian sand on the specimen surface, the MPD decreased. During the whole process, the MPD of the two specimens decreased by 52% and 56.4%, respectively.
Ra is the average deviation of the contour arithmetic and is used to characterize the texture of the road surface in the scanning range. It reflects the dispersion degree of the change of the pavement structure profile concerning the reference line. The increase of aeolian sand makes the surface part of the specimen to be filled. As the depth decreases, the texture decreases. When aeolian sand increased from 0 g to 5 g, the Ra of the two mixtures decreased by 27.4% and 8.6%, respectively. During the whole process, the Ra of the two mixtures decreased by 62.2% and 74.6%, respectively. As shown in Figure 13, the Ra value of AC-16 is more significant than that of AC-13, and it sharply drops with the increase of sediment accumulation.
Rq is the root mean square deviation of the profile and is used to indicate the surface texture of the specimen. As the amount of aeolian sand gradually increases, the internal voids of the specimen are occupied by aeolian sand particles, the macrostructure is reduced, and the surface is rough. When the temperature decreases, the Rq index decreases, and then the BPN value of the specimen surface decreases. In the initial stage, Rq of AC-16 decreased slowly, compared with AC-13. The decrease of Rq index was 24.5% and 5.7%, respectively, and the decline of AC-16 in the late stage was significantly increased, and the overall decline was more than AC-13. During the whole process, the Rq of AC-13 and AC-16 decreased by 57.9% and 72.5%, respectively.
On the whole, when the number of sand increases from 0 g to 5 g, the three texture indicators of MPD, Ra, and Rq decrease by AC-13 larger than AC-16. Meanwhile, the voids and macro-structure of AC-13 are smaller than AC-16. In the presence of a small amount of aeolian sand, the texture index of the AC-13 surface is more sensitive. AC-16 can contain a small amount of sand and has little effect on its texture in the initial stage.
Aeolian sand covers the road surface, which directly affects the macroscopic structure of the road surface, which in turn makes the road surface texture change significantly. The result of the change in the road sign texture is the corresponding decrease in the BPN value. Therefore, it is believed that the relationship between the road surface texture index and anti-skid performance certainly exist.
The correlation analysis between the BPN value and three texture indicators of MPD, Ra, and Rq was performed. As shown in Figure 14, BPN and Ra values have an excellent linear relationship, indicating that the increase of aeolian sand make the contour of the pavement structure. The slope of the trend line for AC-16 is more significant than that for AC-13. aterials 2021, 14, x FOR PEER REVIEW The correlation analysis between the BPN value and three texture indic Ra, and Rq was performed. As shown in Figure 14, BPN and Ra values hav linear relationship, indicating that the increase of aeolian sand make the conto ment structure. The slope of the trend line for AC-16 is more significant than th The BPN value and Rq of the two mixtures were fitted, and the result Figure 15. There was also an excellent linear relationship between the BPN The goodness of fit for AC-13 is 0.9088, and for AC- 16  The BPN value and Rq of the two mixtures were fitted, and the results are shown in Figure 15. There was also an excellent linear relationship between the BPN value and Rq. The goodness of fit for AC-13 is 0.9088, and for AC-16, 0.8727. The increase of the contoured root means square deviation Rq will cause the increase of the BPN value. The two are in direct proportion, indicating that the more the amount of aeolian sand on the road surface, the stronger the filling capacity of the road surface. At this time, the road surface texture decreases, and the BPN value also increases. It decreases accordingly. The BPN value and Rq of the two mixtures were fitted, and the results are s Figure 15. There was also an excellent linear relationship between the BPN value The goodness of fit for AC-13 is 0.9088, and for AC-16, 0.8727. The increase of toured root means square deviation Rq will cause the increase of the BPN value. are in direct proportion, indicating that the more the amount of aeolian sand on surface, the stronger the filling capacity of the road surface. At this time, the road texture decreases, and the BPN value also increases. It decreases accordingly.

Conclusions
This paper studies the influence of aeolian sand on the anti-sliding performa texture index of asphalt mixture, which is a study based on the macro perspectiv anti-sliding performance changes caused by different amounts of sand and asph

Conclusions
This paper studies the influence of aeolian sand on the anti-sliding performance and texture index of asphalt mixture, which is a study based on the macro perspective of the anti-sliding performance changes caused by different amounts of sand and asphalt mixture types.

•
As the number of sand accumulation increases, the BPN values of AC-13 and AC-16 mixtures show a downward trend. By fitting the DoseResp model, the later changes can be predicted based on the changes in the initial BPN value. • By studying the anti-sliding performance of pavement surface whether aeolian sand contains water, we find that the sand with water is generally lower than that of dry sand, and the skid resistance decreases with the increase of water volume. • Through the grey correlation analysis, it is found that the aeolian sand particles between 0.075-0.15 mm have the most significant influence on the BPN value of the two mixtures because its mass percentage is the largest. • Ames engineering texture scanner was used to obtain texture indicators under different amounts of sand. It is found that all three texture indicators are correlated with the BPN value. With the increase of the sand deposition within 20 g, the effective structural depth of pavement macrotexture decreases, resulting in the decrease of BPN value of road surface, which means that a certain amount of aeolian sand has a significant impact on the anti-skid performance of the mixture at the macro level.