Role of Wind, Ground Surface, and Slope in Plastic Waste Movement on Terrestrial Environments

Trajectory of waste on land is a strategy for preventing and mitigating plastic pollution in the environment and a component of its management strategy. However, basic data related to the dynamics of the movement of plastic waste on land as the main data in the model design process has been limited. This research was conducted to meet this need by reviewing the influence of environmental factors such as wind, air runoff, soil surface, and slope on the movement of plastic waste on land. In this study, primary data collection of plastic movement was simulated in experimental scenarios for different categories of plastic based on wind-propelling factors and physical environmental factors represented by ground surface characteristic and slope variation. The results indicated that (1) the windspeed threshold through all the explanatory variables were significantly different, (2) wind speed shows a strong positive relation with the flux of plastics rate where plastic moves according to wind direction, (3) vegetated areas have become potential accumulation locations, and (4) the plastic rate through the slope variable were not significantly different. In constructing land-based plastic waste trajectory models, plastic type, ground surface, and slope variations can be regarded as the primary variables.


Introduction
Plastic waste is a global issue due to its pervasiveness in daily life and environmental impact.Plastic litter pollution hinders human activities, diminishes the aesthetic value of the environment, particularly in tourist areas [1,2], and is very expensive to manage which reach around $8,900 per ton of liter [3,4].In addition, plastic pollution is one of the leading causes of flora extinction [5] and is predicted to affect human health via food chain systems [6].Due to the high rate of plastic production (reaching 367 million tons per year) and inadequate management [7,8], the quantity of accumulated plastic litter in the environment is increasing.The plastic then migrates and accumulates at a specific location on the land surface.Drainage and water system is one location where plastic litter may accumulate [9][10][11].According to satellite images and previous studies [7,12], the plastic accumulation found in the Pacific Ocean resembles an island.
Although it passes through several stages, land-based plastic litter is not completely released and accumulates in the ocean.Seawater discharge along the coast [1], river currents [13], and animal behavior [5] can all contribute to the transport of plastic litter into the ocean.Therefore, there is a time and input latency between the total amount of plastic waste produced on land and the amount of plastic waste accumulating in the ocean.Based on the accumulation of plastic litter in the ocean, it was determined that plastic litter from land-based resources represents 80% of the waste, with sea-related activities accounting for the remaining 20% [7,14].The vast majority of human activities performed on the land, notably the method and intensity of use, significantly impact this [9,15].The main causes of the 4.8 to 12.7 million metric tons of plastic waste released annually are non-biodegradable polymers and single-use plastics.It may lead to more of it accumulating in the oceans.
Numerous studies have devised models for the transport of plastic litter, especially in rivers and on beaches [15][16][17].Recent research has begun to focus on the land to predict the probability of accessing aquatic bodies [18,19].The global probabilistic model concerns socioeconomic and social variables as input values [20,21].The actual method of transporting plastic litter from land to water bodies is not yet widely employed.Plastic pollution from land will eventually enter bodies of water and be deposited into the ocean.To enhance plastic pollution prevention, mitigation, and reduction strategies, it is necessary to comprehend the movement of plastic litter from the land.Experimental and theoretical methods can be used to determine the transport of litter waste on land.
Based on prior models, plastic mobilization identifies wind and discharge as the driving force and ground surface features as the resisting force.Moreover, depending on the incline's steepness and whether the wind is dragging the plastic upwards or downwards, the slope can function as either a driving force or a resisting force [15,18].In this study, we focused on wind speed as the driving force behind the identified plastic mobilization, with the physical environment serving as the experimental basis for the investigation.In addition, observing the various categories of plastic must be conducted under the supposition that each type of plastic has a unique movement mechanism and varying probabilities of reaching the river or the ocean.Based on this, the research will focus on three distinct plastics that are most prevalent in aquatic systems [19].

Ground Surface Characteristics
The area needs to be large enough, have a variety of slopes, and ground surface to simulate environmental conditions for this research adequately.During experiments, private spaces were also used to facilitate the control procedure.The optimal location for conducting research is the Jatinangor Campus of the Bandung Institute of Technology in the Sumedang Regency of West Java.The area, which measures 47 hectares from north to south, satisfies the following criteria: (1) It is not a public site because it has only a student study area; (2) It has a range of slopes (about 0% to 24%) between elevations of 737.50 and 742.50 meter above sea level.However, we only determined by two factors, flat and slope (around 18-20%); and (3) It has different ground surface characteristics, including urban routes as concrete asphalt roads, barren land with some gravel and sand particles (around 2 mm in size), natural grass with vegetation height greater than 50 cm, and vegetated areas with trimming maintenance or cut grass with vegetation height lower than 5 cm [18].The specific vegetation was determined as Perennial ryegrass.

Experimental design
The experimental study was carried out to gather the primary data.In this case, the fan was used as a wind speed simulator (Figure 1).To eliminate wind distractions outside the scope of the study, wind parameter procedures were carried out in a mobile wind tunnel with 1x1x6 m dimensions [22].The wind tunnel can reduce 53 to 80% on wind speeds that exceed 3 m/s, and 80 to 100% on wind speeds less than 3 m/s.The wind parameter will be defined in terms of two factors: the windspeed threshold or the minimum wind speed required to move the plastic and the flux, which compares the plastic rate to the wind speed.To measure the wind speed threshold, the plastic is first laid out 3 m from the fan and then slowly moved 10 cm in each direction.The process is complete when the plastic moves (by at least 5 cm), and the wind speed is recorded on that point at least three times.To determine the plastic rate, plastic was set sequentially at wind speeds of 2, 4, and 6 m/s, and the rate was then calculated using Video Tracker analysis.We used a point-to-fan distance basis for each predetermined wind speed, setting the plastic at a specific point corresponding to the wind speed.We used the following tools: power supply, fan, roll meter, action camera, anemometer, and calibration stick.

Plastic types
The plastic types used in this research is a large plastic (>0.5 cm) as a prototype of popular macroplastic waste found in the ocean [19].Various kinds of plastics were simulated to see the different mechanisms and get more accurate coefficient values as the reference primary data for the model simulation [18].Three types of plastic were selected based on their pollution dominance in the water system: plastic bags, plastic bottles, foam food containers.We realized that the dimensions and mass of plastic are estimated to affect the mechanism of plastic movement on land [23], so we limited the dimension with the minimum size to 7 cm in this study and has comparable dimensions to paper (Table 1).The plastic bottle we used in our study has a special condition that considers the volume of the bottle tube with air inside, so the density (0,02 g/cm 3 ) is lower than the density of water.

Data analysis
We applied Video Tracker and R Studio as our main software analysis.Video Tracker was used to analyse the vector, time, distance, and velocity based on the video graphic record, and linear fit analysis was used to calculate the plastic rate at each location.In our plastic rate analytical approach, we set the following software tools: (1) the calibration stick as the main comparison with the real size in the field, (2) rule to measure 2D dimension, (3) axis rule to determine the direction on x axis, (4) automated track control to follow the specified colour on plastic, and (5) linear analysis.Secondly, a data analysis of the windspeed threshold and the plastic rate in comparison analysis to the explanatory variable was done using R Studio Version 4.3.0.Multiple ANOVA without repeated was employed and followed additional tests were conducted using the Tuckey analysis to compare more specific mean over each group.Generalized Linear Model (GLM) also used to determine the correlation on different wind treatment on regression formula to fit the ratio of plastic rate and wind speed.Before applying the ANOVA and GLM, we utilized to standardize (Shapiro Wilk) and data transform (Bartlett test) for the best fitted data, and the interactions were only included if the value is significant.

Equations
The following equation is used as a conversion factor for wind speed in the scope of the study to wind speed at another altitude as an approach in comparing specific wind speeds [24].Where v2 is wind speed at specific altitude (m/s) was determined by the v1 wind speed reference (m/s) multiplies to the ratio of wind h2 height reference (m) and h1 determined height (m), and α is the Hellman coefficient (0.34 for surface neutral air).
Plastic movement can occur due to the force exerted by the wind.Based on the preceding principles the N denoted as wind force (F), was determined by multiplying the  air density 1.29 (kg/m 3 ), the v wind speed (m/s), surface A area (m 2 ) and the dimensionless drag coefficient (assumption = 1.0).

Data Exploration
Our measurement data was compiled in Figure 2, Figure 3, and Figure 4 to determine the windspeed threshold, the plastic motion, and the plastic rate.We have curious through the pattern of the data and analysed it based on the group.

Windspeed threshold.
Following the close examination based on Figure 2, each type of plastic requires a unique wind speed at varying ground surfaces.The pattern of plastic movement on flat and sloping barren land follows the same rank where plastic bags (0.9 and 0.8 m/s) and foam food containers (1.1 and 0.8 m/s) are quicker moved by the wind than plastic bottles (1.6 and 1.8 m/s).It was also observed that on sloping barren land plastic bags has the same windspeed threshold, besides foam food containers moved more easily, but plastic bottles need more wind speed.This is because plastic bottles prefer to roll on the ground as opposed to the other two varieties of plastic, which can wade through the wind, and can be stopped by building a tiny mound (0.2 cm high) of rock particles that move with the wind.Compared to the plastic's movement on concrete asphalt, where it moves in the following order: plastic bags (0.8 m/s), plastic bottles (1.3 m/s), and foam food containers (1.4 m/s) on flat, and plastic bottles (0 m/s), plastic bags (0.8 m/s), and foam food containers (1.1 m/s) on slopes.Because of its solid surface and round shape, plastic bottles can move without wind but are greatly impacted by gravity.On cut grass, the same pattern occurs on barren land where on flat and sloping surfaces for plastic bags are 1.6 and 1.1 m/s, foam food containers are 1.8 and 1.5 m/s, and plastic bottles at 2.9 and 3.5 m/s.even with the same sequence pattern, a larger windspeed threshold is required on this surface that was 3 cm high.However, there are necessary windspeed thresholds before plastic may travel unchecked over natural grass.The greater windspeed threshold requirements are needed, nevertheless, to move plastic over nature grass with more than 30 cm high.This demonstrates how vegetation has the capacity to both restrain the movement of plastic and serve as an accumulation path for it.

Plastic rate.
It is required to ascertain the plastic's rate in order to establish that it goes in the wind at a different pace.We utilised Video Tracked software to apply rate analysis and obtain accurate data.Principally, the analysis functioned by delaying and dividing the frame to simplify the analytical procedure.The calibration stick and quadrant line were required to establish the field's dimensions and orientation.In addition, the analysis only considers the x direction under the assumption that the wind moves in a single direction without considering the resultant force.To control for this, we limited our data collection to no more than 0.2 m of y-axis movement by aligning the quadrant lines (particularly the x-axis) at point 0 as the initial and plastic end points as the final position.The limit value is determined as the average side dimension of the plastic we have, taking into consideration the possibility of one rotation due to wind, and it is crucial to be limited because the tunnel we use has a dimension limit.In this study, only a few minor data travels along the y-axis and strike the tunnel wall.We eliminated and recollected any data that exceeded the y-axis limit, except for the plastic bottle.Figure 3 depicts the motion of each plastic when treated to wind at 4 and 6 m/s exerting 25.86 to 139.0 Newtons of force.This diagram illustrates the initial positioning of the plastic and the predominant field movement.Plastic bags and foam food containers will slide on the ground surface in the wind direction and begin rolling when one of their edges strikes the ground.The peculiar thing that was seen was that the plastic bag often had a gap between it and the ground following through the wind flow and allowing the plastic to escape all kinds of little obstructions.In the other hand, due to the unbalanced weight of the bottle cap, a plastic bottle will roll in the direction of the wind (x-axis) and begin rolling perpendicular (y-axis) even before the movement reaches 1 m.Due to the limited media and devices available, we discontinued the measurement when the object rolling perpendicularly more than 0.2 m, but we still documented as additional proof.However, at a wind speed of 2 m/s, the plastic bag and foam food container will slide directly through the surface and the bottles will continue to roll as the basic of its movement.
We specified analysing the plastic rate on the 6 m/s wind experiment.This was the maximum air flow velocity that our device could accommodate, and it will serve as representative data in our discussion since the results of the flux analysis fall within the same range.This part intrigued us that the plastic rate could move at the same rate as the wind.In this case, the flux value equals 1 regardless of the plastic moving rate relative to the wind speed.The results of our analysis demonstrate three facts.First, plastic bags have a wide range of flux values, between 0.15 and 0.31, with the highest flux occurring on flat and sloped barren land and the lowest on cut grass.Secondly, foam food containers have a flux of 0.24 to 0.31, the largest occurring on sloped barren land and the lowest on both flat and sloped concrete asphalt.Thirdly, plastic bottles have the lowest flux compared to other plastics, with a range of 0.1 to 0.24, with sloped barren land being the highest and sloped cut grass being the lowest.

ANOVA and Generalized Linear Model Results
ANOVA describes the differences between each wind speed threshold and plastic rate for various varieties of plastic, ground surfaces, and slope.Table 2 demonstrates that, with a 99% confidence level, the wind speed threshold parameter for each explanatory variable and interaction between groups was significantly different, and the 95% confidence level for the interaction between plastic type and slope.Additional analysis (post hoc Tukey) performed to identify differences between the means of pairs of groups.The 95% confidence level indicates that the difference between the means of plastic bag and plastic bottle is statistically significant (p-value = 0.025) based on the variable type of plastic.Cut grass differs considerably from concrete asphalt (p-value = 0.0001) and barren land (pvalue = 0.002), as the ground surface variable indicates.Table 3 describes the plastic rate parameter that reveals significant differences between plastic types, ground surfaces, and their interactions.However, the plastic rate at 6 m/s wind speed based on slope variation did not adequately explain the significant differences due to their similarity and inconsistency with the explanation that will be mentioned in the discussion section.The ground surface became significantly different due to the great influence of cut grass, where the plastic moved more slowly than the other.Thus, the vegetation became a literal barrier for the plastic.According to the results of the Tukey analysis, the following portions of each group had the greatest impact on the plastic rate analysis: First, plastic bottles had substantially different values than the other two types of plastic, while plastic bags and Styrofoam had identical moving rates (p-value = 0.98).In contrast, cut grass has significant differences in the ground surface group, which differ considerably from the other two ground surface categories, and the plastic rate on asphalt and barren land tends to have similar values (p-value = 0.32).The GLM describes the rate of plastic mobilized on land at wind speeds of 2, 4, and 6 meters per second.It indicates that with an increase in the wind speed range, the rate of plastics mobilized increased.The correlation coefficients on wind speed and the plastic range are highly correlated (Table 5).We have incorporated the linear equation to approximate the plastic rate from wind speed data.

Discussion and Suggestions
The results from this study clearly show that wind contributes to the remobilization of plastics on terrestrial environments.Specifically, the plastic types, the ground surface characteristic, and the slop of the ground are relevant in explaining the remobilization of plastics.Five main conclusions can be deduced from the analysis.First, our wind speed threshold results have significantly different values than the reference's extrapolation.In this case, the prior model simulated the plastic is nearly as dense as paper (1.2 g/cm 3 ) however we try to approach the interpretation using a dimensional scheme comparable to paper besides its dense.Secondly, all the explanatory variables have significant differences to explain the wind speed threshold and can be considered to categorize these parameters before processing the model.Thirdly, wind speed shows a strong positive relation with the flux of plastics rate.However, it's still potential to increase the wind speed for the next experiment.We keep presuming that it is possible for plastic to remobilize with the speed of the wind, just as it is possible for the wind to push larger and more massive objects, such as a house by the hurricane [18].Fourth, plastic rate can help to determine the possibility of plastics to reach the other terrestrial system or even the river.On this approach, plastics can be categorized based on its types to identify the mobilization, and ground surface can be categorized based on its categories to determine the potential obstacle.Plastic bag and foam food container can be as a group and different with the plastic bottle.These have the same case on the ground surface parameter that asphalt road and barren land can be as a group and different with the cut grass.Finally, based on our significant wind speed differences through the slope parameter, we could not find the consistent pattern to decide the value of reduction or addition winds.
But almost of the windspeed threshold reduces when it faces the decline surface.We would like to discuss more deeply about the wind threshold.The wind threshold on the prior reference was determined based on the wind scale by Francis Beaufort's interpretation.The initial measurement observed on seas wind waves and determined the wind into 1 to 12 levels based on the direct observation of how the wind interacted with its surroundings and the scale range came out to make it more accurate.In this analysis, we focused on where wind speed can blow the paper and dust on 5,6 m/s (conversion of Beaufort's fourth scale).It was approximately 1-2 m height.We attempted to convert the winds to the floor surface using an equation based on a height of approximately 0.1 m and obtained a Beaufort scale range of 2.2 to 2.7 m/s.This value is closer to the results of our research.
Due to the low proportion of plastic in relation to the slope variation, we have a novel hypothesis that plastic mobility on flat and sloped surfaces may differ, with plastics on sloped surfaces being more mobile and able to travel farther because the wind blows the plastic far above the ground.There is no statistically significant difference, and according to our analysis, the wind mobilized the plastics through its flow, which followed the surface irregularity of the ground.However, the wind flow will encounter turbulence when it interacts with an obstruction or disperses on the stable wind [24].Another indication shows that vegetation could impact the movement of plastic.In our explanation, the space between the vegetation and the edge of plastic plays the puzzle rule.When the plastic gets stuck between the plants, the plants will catch the extremities of the plastic.The taller the plant, the higher the probability of plastic retention.The finding on the accumulation of plastic in the vegetated region is not the first.According to prior research, vegetation along the river's border can act as a collection point for plastic.Predicted wind speeds of 22 m/s are required to transport the plastics over significant obstacles such as vegetation [20] and runoff that is highest than vegetation is required to eliminate the plastic [27].
The generalization of our results is limited by several factors.The measured wind speed was simulated in dry weight mass and on individual number, however almost of the mismanagement plastic waste is wet and in the collection on different plastic type.Our firsthand observation mostly demonstrates that the plastic bag may go farther than the other plastic type from the same beginning position, wind speed, and long intervention time.As a result, the probability approach to the bunch of plastic can lower its potential for mobilization on land.However, additional analysis of the extrapolation value is required.
There are still many opportunities to assess and examine plastics from other perspectives, such as variation in other types, size, shape, and density differences.Ground surface also plays a unique term in our study.Vegetation potentially holds the plastic between its space.So, the plastics needed more wind force to remobilize.There is no specific determination of friction in our plastic rate discussion, considering that gusts of wind can act as an additional layer flow between the plastic and the ground surface.There is a conclusive argument concerning plastic friction against the ground's surface.When the plastic reaches the surface, the rank of the ground surface must be concrete asphalt road as the smoothest one, followed by grass and cut grass based on simulation of Woven Polypropylene Plastic (bulk bag material) on different ground surface to hold the flood [28].Further research on observing actual events in the field relating to plastic transport on land by the wind is still required to determine the probability of plastic entering water bodies.

Figure 1
Figure 1 Wind's experimental design.(a) Wind tunnel design: the arrow indicates the trajectory of the plastic through the fan.Light green: the plastic was deposited 3 m away from the fan and replaced every 10 cm until it moved in the wind's direction.Dark green: the plastic was positioned at a predetermined point, (b) The default layout for the instruments.

Figure 2
Figure 2 Windspeed threshold required to move plastic over different types of flat and sloping terrain.Red-dotted line denotes the windspeed threshold based on the prior reference.

Figure 3
Figure 3 Motion path of three plastic types over a 3 m distance.The blue arrow denoted the wind flow, the dash grey arrow represents the direction of the motion path, the dashed-grey curved arrow represents the plastic's rolling motion, and the red point is denoted as the bump point.

Figure 4
Figure 4 Distribution of plastic rate range at 6 m/s winds

Table 1
Identification of three plastic type used in the experiment.

Table 2
ANOVA model output of windspeed threshold with explanatory variables, Sum square (Sq), degrees of freedom (df), F value, and the p-values based on a F test [Pr(>F)]

Table 3
ANOVA model output of plastic rate with explanatory variables, Sum square (Sq), degrees of freedom (df), F value, and the p-values based on a F test [Pr(>F)]

Table 4
Generalized linear model output with the rate of plastics mobilized on land a function of the plastic types on different wind speed (2, 4, and 6 m/s).y denotes the plastic rate and x denotes the wind speed.