Effect of Destruction and Residual Geomembrane on Soil Organic Matter, Evaporation Cracking, and Aggregates Under Dry–Wet Cycles

Abstract

Geomembranes are widely used in water storage ponds and irrigation systems in arid and semi-arid regions. The residual damage of geomembranes will lead to soil pore blockage, water infiltration obstruction, and soil degradation. This study investigated the effects of geomembrane residues at different depths on soil under drought and rainfall conditions. Using a ZHS multifunctional climate chamber to simulate dry–wet cycles, the effects of residual geomembranes on soil evaporation and cracking processes were investigated, along with their impacts on soil aggregates and aggregate-associated organic carbon. Results showed that with increasing residual geomembrane content, the Mean Weight Diameter (MWD) of aggregates decreased by 3.95%, 28.25%, and 51.41%, and the Geometric Mean Diameter (GMD) decreased by 10.59%, 42.35%, and 61.18%, respectively, compared to the residual geomembrane-free treatment. Organic carbon content in all aggregate size fractions consistently declined. Under the same dry–wet cycling conditions and soil thickness, residual moisture content decreased with higher residual geomembrane addition. Meanwhile, residual geomembrane enrichment promoted soil cracking, as evidenced by increased crack ratio and fractal dimension, and the initial evaporation rate increased with increasing residual geomembrane content. Compared to the control group, residual geomembrane significantly reduced residual moisture content, reducing it by 4.48–29.37%, 7.14–21.92%, and 4.19–35.95%, respectively. The final crack ratio increased by 5.33–58.89%, 0.97–63.39%, and 0.87–72.46%, respectively. Meanwhile, the final fractal dimensions increased by 0.50–15.26%, 2.92–15.96%, and 3.22–13.33%, respectively. Mechanistically, residual geomembrane fragments occupy soil pores and reduce interparticle cohesion, thereby promoting the expansion of crack ratios and disrupting aggregate stability, which accelerates organic carbon decomposition. This study provides scientific insights into how residual geomembranes affect soil ecosystems and supports agricultural soil conservation and management.

1. Introduction

As an important geosynthetic material, geomembrane is mainly made of high-molecular polymer materials, with components including high-density polyethylene, linear low-density polyethylene, and polyvinyl chloride [1,2]. It is widely used in geotechnical engineering, environmental engineering and agricultural fields to play roles of isolation, protection and reinforcement. Geomembrane plays a key role in regulating soil water transport, and has a significant impact on soil physical and chemical properties, hydrological processes and ecological environment [3]. In recent years, with the widespread application of geomembrane in industry and agriculture, a large number of geomembrane residues have been improperly disposed of into the environment. The persistent presence of residual geomembrane in soil poses a significant threat to soil health and triggers a series of ecological and environmental risks [4].

Although the geomembrane buried in the soil is protected from light, it is still affected by ground temperature. High temperature facilitates the release of strain energy stored in macromolecular chains and induces chain fracture, which destroys the mechanical properties of materials. The continuous stress generated by farming, mechanical rolling or soil expansion and contraction will reduce the stability of polymer molecular bonds and accelerate the breaking of geomembrane through physical and chemical mechanisms. Polyethylene geomembrane is extremely difficult to degrade under natural conditions, and complete decomposition of geomembrane has been estimated to take hundreds of years [5]. Eventually, the geomembrane gradually breaks into millimeter- or micron-size plastic particles, resulting in soil microplastic pollution [6]. Residual geomembrane fragments cut off the continuity of soil pores, hinder water infiltration and air exchange, lead to increased soil bulk density and decreased porosity, and caused soil hardening and weakened drought resistance. Figure 1 shows the schematic diagram of soil degradation process caused by damaged geomembrane residue.

Residual geomembranes significantly impact the formation of soil aggregates and the storage of organic carbon. The formation of soil aggregates is the result of the combined effects of microorganisms, plant roots, and other complex physical and chemical processes [7]. The content and size of aggregates are core indicators for measuring soil quality and are closely related to soil fertility and biological activity. Aggregate stability significantly affects soil porosity, erodibility, permeability, and the regulation capacity or function of soil water, fertilizer, air, and heat [8,9]. The content and stability of aggregates largely depend on the complex interactions among soil organic carbon content, particle composition, and microbial communities. As residual geomembrane accumulate in the soil, some of them become embedded within aggregates, altering the quantity of soil aggregates [10]. Different types of residual geomembrane form different aggregates when combined with soil: shredded residual geomembrane can form loose aggregates, while plastic fibers form dense aggregates [11]. Additionally, as a non-degradable organic pollutant, residual geomembrane exert toxic effects on soil microorganisms, reducing microbial activity [12], and affecting the soil organic carbon cycle and altering soil organic carbon content [13].

Evaporation is a key process in hydrology and agriculture, and in industrial drying [14]. In arid and semi-arid regions, the loss of soil water to the atmosphere in the form of water vapor, which is an important component of the terrestrial water cycle [15]. Soil water evaporation is mainly influenced by soil structure, soil texture, meteorological conditions, and initial moisture content. Soil water evaporation determines the effective utilization of soil water resources, thereby affecting plant water use efficiency. The formation of desiccation cracks in soil serves as the main pathway for water movement, and the evaporation rate also affects the degree of soil shrinkage and cracking. The retention of residual geomembrane in soil alters the water cycle, affecting soil water infiltration capacity [16], which in turn influences soil water evaporation and pollutant migration, and percolation into deeper soil layers, exacerbating soil water scarcity in polluted areas [17].

After the geomembrane is aged and broken, the components such as HDPE and PVC, which are difficult to degrade, will lead to microplastic pollution in the soil [18]. In recent years, microplastic pollution has attracted great attention in the field of environmental science [19]. Microplastic pollution not only significantly affects the physical and chemical properties of soil [20], but also alters soil structure, bulk density, water retention capacity, contact angle, and nutrient availability [21]. Due to their large surface-area-to-volume ratio, microplastics can enhance the absorption, release, and transport of chemical substances, making them more likely to cause environmental problems [22]. This characteristic allows microplastics to more easily adsorb pollutants such as heavy metals, polychlorinated biphenyls, polycyclic aromatic hydrocarbons, and organochlorine pesticides, significantly increasing the toxic risks to soil organisms [23]. In soil systems, microplastic pollution inhibits the growth of plant roots and stems, significantly reduces the total root length of seedlings, and thereby affects crop growth and yield. This impact is not only reflected in the cumulative effects on the plants themselves but also interferes with their normal antioxidant stress responses [24]. Furthermore, the migration behavior of microplastic particles in soil further exacerbates the scope of pollution. Under the influence of animal activities and human disturbances, microplastics can migrate vertically downward, extending pollution to deeper soil layers [25]. Even more concerning is that through the ingestion of groundwater, agricultural products, and animals contaminated with microplastics, these pollutants pose a direct potential threat to human health [26].

Although existing studies have revealed the multifaceted impacts of residual geomembrane on soil systems, the specific mechanisms and extent of their effects on soil aggregates and organic carbon stability remain insufficiently understood, and the impact of residual geomembrane on soil evaporation and cracking is still unclear. Therefore, the focus of this study is to investigate the impact of microplastic pollution on soil aggregates and organic carbon in arid regions. Based on laboratory experiments, the desiccation cracking characteristics of residual geomembrane soil and their underlying mechanisms are examined in detail.

2. Materials and Methods

2.1. Materials

Soil samples were collected from Zhengzhou City, Henan Province, China. This region exhibits a typical northern temperate continental monsoon climate with distinct seasonal characteristics: dry springs with little rainfall, hot and rainy summers, clear and sunny autumns with long daylight h, and cold winters with minimal snowfall. The annual average precipitation reaches approximately 640.9 mm. Under extreme weather conditions, the maximum temperature can climb to 43 °C, while the minimum temperature can drop to −8 °C. The specific physical parameters of the test soil are detailed in

Table 1. Basic physical properties of test soil.

Moisture Content (%)	Density (g/cm3)	Specific Gravity	Liquid Limit (%)	Plastic Limit (%)
23.7	2.36	2.62	33.8	19.5

.

2.2. Broken Residual Geomembrane

High-density polyethylene geomembranes were selected as the residual geomembrane additive for this experiment. The performance parameters of waterproof and barrier geomembranes made of high-density polyethylene are shown in

Table 2. Basic physical properties of high-density polyethylene geomembrane.

Density (g/cm3)	Thickness (mm)	Puncture Resistance (N)	Resistance to Environmental Stress Cracking (h)	Oxidation Induction Time (min)
0.96	0.5	300	300	400

. In order to simulate the impact of broken geomembrane on soil, the high-density polyethylene geomembrane was broken, and the residual geomembrane with a particle size less than 1 mm was selected for testing.

2.3. Testing Process

To replicate real-world farmland conditions, a plot in Zhengzhou, Henan Province, was selected and divided into multiple rectangular sections of equal size (2 m × 2 m). The experimental field was leveled and partitioned to ensure consistent conditions across all four test areas. Residual geomembrane particles were added to the soil of each rectangular section at concentrations of 0%, 1%, 3%, and 5%. Following the experimental design, the calculated residual geomembrane particles were uniformly spread over the soil surface of the corresponding test areas and thoroughly mixed with the soil through tillage. The field-configured test blocks were then left in their natural state for one year. The field plot treatment was exclusively applied to the aggregate structure and soil organic carbon determination tests, which aimed to investigate the long-term effects of geomembrane residual particles on in situ soil physicochemical properties under natural field conditions.

Soil aggregates were determined using the wet sieving method. Fifty grams of soil sample were weighed onto sieves with different apertures: 2 mm, 1 mm, 0.5 mm, 0.25 mm and 0.053 mm. Each sieve was immersed in water and shaken vigorously to obtain soil aggregates of particle sizes > 2 mm, 1–2 mm, 0.25–1 mm, 0.053–0.25 mm, and 0–0.053 mm, respectively. The content of soil aggregates with particle size > 0.25 mm (R > 0.25) is defined as the mass ratio of wet-sieved aggregates > 0.25 mm to the total mass. The mass ratio calculation formulas for aggregates of different particle sizes are as follows:

$${MP}_i={\displaystyle \frac{M_i}{M_{SS}}}\times 100\%$$

(1)

In the formula: MP_i denotes the mass percentage of soil aggregates in a specific particle size fraction, expressed as a percentage; M_i represents the dry weight of soil aggregates in a specific particle size fraction, in grams; and M_SS indicates the total weight of the soil sample, in grams.

To investigate the impact of residual geomembrane particles on soil aggregate stability, the Mean Weight Diameter (MWD) and Geometric Mean Diameter (GMD) of aggregates will be employed to characterize soil aggregate stability. The specific calculation formulas are as follows:

$$MWD={\sum }_{i=1}^n{\overline{x}}_i{MP}_i$$

(2)

$$GMD=\mathrm{e}\mathrm{x}\mathrm{p}({\displaystyle \frac{{\displaystyle {\sum }_{i=1}^n}{MP}_i\times {ln\overline{x}}_i}{{\displaystyle {\sum }_{i=1}^n}{MP}_i}})$$

(3)

In the formula: MWD denotes the mean mass diameter of soil aggregates, mm; GMD denotes the Geometric Mean Diameter of soil aggregates, mm; ${\overline{x}}_i$ denotes the mean diameter of aggregates at a specific size, mm; n denotes the number of size groups for aggregates.

Soil organic carbon plays a crucial role in promoting soil aggregate formation, enhancing aggregate stability, and improving the chemical properties of soil aggregates. In this study, the organic carbon content in soil aggregates of different size fractions was determined using the potassium dichromate oxidation–external heating method.

The workflow of evaporation cracking test and crack image processing is shown in Figure 2. In this study, the collected undisturbed original soil without on-site particle modification was air-dried and crushed. Subsequently, a standard sieve with a mesh size of 2 mm was used for screening to effectively remove coarse impurities such as gravel. In this experiment, uniformly sized 0.15 mm residual geomembrane particles were selected as the additive material (Figure 2a). Different masses of residual geomembrane particles were uniformly mixed with the same soil sample to prepare soil samples with residual geomembrane contents of 0%, 1%, 3%, and 5%, respectively. Distilled water was then added to the prepared soil samples and stirred thoroughly, controlling the moisture content to 100% to achieve a supersaturated state. A transparent acrylic square box measuring 18 cm in length and 5 cm in height was selected as the test container. Vaseline was applied inside the experimental container to prevent sample-container friction from affecting results. Prepared samples were poured into the treated containers to create specimens with soil layer thickness of 1 cm, 2 cm, and 4 cm. By combining different levels of residual geomembrane concentration with different levels of surface soil thickness, 12 different experiments were set up, and a set of parallel experiments was established.

The evaporation cracking test was conducted in the ZHS multifunctional climate chamber (Figure 2b). The equipment was sourced from manufacturer Hangzhou Jiuhuan Environmental Testing Equipment Co., Ltd., located in Hangzhou, China. This chamber offers powerful and flexible capabilities to simulate diverse atmospheric evaporation environments, ensuring stable and precise experimental conditions. The sample was precisely placed on a balance scale before being transferred into the chamber. To comprehensively capture sample changes, a high-resolution camera was installed above the sample to ensure complete recording of its surface. The camera automatically recorded cracking morphology every 2 h, continuously tracking crack development until no significant weight change occurred, signifying the end of the evaporation process.

2.4. Crack Image Processing and Parameter Calculation

Image processing technology not only provides an effective means for monitoring crack formation and propagation induced by soil evaporation at the macro level, but also significantly advances the precise quantitative study of specimen cracks. This study employs high-resolution image acquisition methods, utilizing a Sony A7M4 high-definition camera for real-time capture. The equipment was sourced from manufacturer Sony Group Corporation, located in Tokyo, Japan. With a maximum resolution of 33 million pixels, the camera ensures the fineness and information content of the acquired images. Appropriate preprocessing of raw images is particularly crucial, aiming to protect images from damage while fully preserving their clarity and authenticity. This process primarily involves three core stages (Figure 2c):

Stage I: First, convert the original color image into a grayscale image. Given that color images contain rich chromatic information, this could interfere with subsequent crack feature parameter calculations. Therefore, the primary step is to convert the color image into a grayscale image containing only luminance information. The pixel values of the grayscale image are subdivided into 256 levels, ranging from 0 to 255, where 0 and 255 represent black and white, respectively. Then, convert the grayscale image into a binary image. Finally, apply noise reduction to the binary image to obtain a clear grid pattern. During conversion, each pixel in the color image is originally defined by values from three color channels: red (R), green (G), and blue (B). To derive the grayscale image, the values from these three color channels are combined using Formula (4) to calculate the corresponding grayscale value.

$$\mathrm{Intensity}=0.4R+0.4G+0.2B$$

(4)

Stage II: The grayscale image obtained after preprocessing can be clearly distinguished into target and background regions with differing grayscale levels. Therefore, selecting an appropriate threshold to achieve precise image segmentation becomes a critical task. Specifically, we first label the original image as m(a, b), then convert it into a binary image n(a, b) through image processing techniques, during which the threshold is set to Q.

$$n(a,b)=\left\{\begin{array}{l}0,m(a,b)<Q\hfill \\ 1,m(a,b)\ge Q\hfill \end{array}\right.$$

(5)

Stage III: Disturbing noise often appears in binarized images. To effectively address this issue, we employ an 8-neighborhood denoising algorithm for noise processing. The algorithm works as follows: if a pixel’s binarization result is 0, it is treated as noise and displayed as black; background pixels are set to 255 and displayed as white. Specifically, isolated noise pixels are typically surrounded by a white background. Therefore, we identify noise based on the following criterion: a pixel is black if all eight adjacent pixels are white. Additionally, for different gaps appearing in the binarized image, we only need to set an appropriate threshold for denoising. Through this step, we ultimately obtain a clearer and more accurate denoised image.

Then, based on the processed images, quantitative analysis was conducted to examine changes in crack characteristics during the evaporation cracking process by calculating the crack ratio and fractal dimension. The crack ratio of the specimens was calculated using the following formula.

$$P={\displaystyle \frac{{\displaystyle \sum Z_c}}{Z}}$$

(6)

In the formula: P denotes the specimen crack ratio, ∑Z_c represents the total area of surface cracks on the specimen (cm²), and Z is the surface area of the specimen (cm²).

Among the various methods for calculating fractal dimensions, techniques such as the scale method, island method, and box-counting method have been applied. Given the irregular characteristics exhibited by crack morphology during specimen fracture, fractal dimension serves as an effective tool for measuring such irregularity. Specifically, the box-counting method, a commonly used technique for measuring fractal dimension, was adopted in this study. This method involves selecting small boxes with side length x and defining E as a finite set. Here, Y(x) denotes the minimum number of boxes with side length x required to cover the set E. Based on this, the fractal dimension dE can be precisely calculated using the following formula:

$$dE=-\underset{x\to 0}{\lim }{\displaystyle \frac{\log Y(x)}{\log x}}$$

(7)

In the formula, dE is the fractal dimension, and Y(x) represents the minimum number of boxes required to cover the set E with x as the side length.

In practical image processing, continuous infinitesimal scale x cannot be achieved. Therefore, discrete box sizes of x = 2, 4, 8, 16, 32, and 64 pixels (powers of two) were selected. Linear regression was performed between logY(x) and logx within the valid scaling range, and the absolute value of the fitted slope was taken as the box-counting fractal dimension of crack patterns.

The dynamic changes in sample moisture content can be indirectly reflected by monitoring the platform scale. To ensure data continuity and accuracy, the test design employs a 2 h interval. The formula for calculating sample moisture content is as follows:

$$w={\displaystyle \frac{m_0-\Delta m}{m_d}}\times 100\%$$

(8)

In the formula: w denotes the moisture content of the sample, m₀ is the initial weight of the water in the sample, Δm denotes the difference in the balance scale reading before and after a 2 h interval for the sample, and m_d is the weight of the dry soil.

The evaporation rate is determined by measuring the reduction in moisture content of the sample per unit time. The evaporation rate is calculated using the following formula:

$$E_{\mathrm{v}}={\displaystyle \frac{\Delta m}{t}}$$

(9)

In the formula: E_v represents the evaporation rate, and t indicates the time interval.

3. Results

3.1. Effect of Residual Geomembrane on Soil Aggregate Particle Size Distribution

We used samples from test blocks that had been left for one year under natural state conditions to ensure representativeness and reflect the actual condition of soil residual geomembrane in the experimental area. Figure 3a–c illustrates the relationship between soil aggregate particle size distribution and residual geomembrane content, the relationship between residual geomembrane content and MWD in soil and the relationship between residual geomembrane content in soil and GMD, respectively.

Table 3. The changes in particle size, MWD, and GMD of soil aggregates with different residual geomembrane content.

Indicator	Residual Geomembrane Content in Soil Samples
	0%	1%	3%	5%
d < 0.053 mm	7.35	8.27 (12.52% ↑)	11.66 (58.64% ↑)	14.24 (93.74% ↑)
0.053 mm ≤ d < 0.25 mm	16.87	18.02 (6.82% ↑)	25.35 (50.27% ↑)	32.57 (93.06% ↑)
0.25 mm ≤ d < 1 mm	23.27	25.29 (8.68% ↑)	27.82 (19.55% ↑)	31.42 (35.02% ↑)
1 mm ≤ d < 2 mm	9.88	9.17 (7.19% ↓)	8.36 (15.39% ↓)	6.85 (30.67% ↓)
d > 2 mm	42.63	39.25 (7.93% ↓)	26.81 (37.11% ↓)	14.92 (65.00% ↓)
R < 0.25 mm	24.22	26.29 (8.55% ↑)	37.01 (52.81% ↑)	46.81 (93.27% ↑)
R ≥ 0.25 mm	75.78	73.71 (2.73% ↓)	62.99 (16.88% ↓)	53.19 (29.81% ↓)
MWD (mm)	1.77	1.70 (3.95% ↓)	1.27 (28.25% ↓)	0.86 (51.41% ↓)
GMD (mm)	0.85	0.76 (10.59% ↓)	0.49 (42.35% ↓)	0.33 (61.18% ↓)

Note: “↑”, “↓” indicating an increase or decrease in soil samples compared to those without residual geomembrane.

shows the changes in particle size, MWD, and GMD of soil aggregates with different residual geomembrane content. The representative diameters of soil aggregates with particle sizes >2 mm, 1–2 mm, 0.25–1 mm, 0.053–0.25 mm, and 0–0.053 mm are 2.5 mm, 1.5 mm, 0.60 mm, 0.15 mm, and 0.026 mm, respectively. After calculation, compared with the sample without residual geomembrane, the addition of 1%, 3%, and 5% residual geomembrane increased soil aggregates with R < 0.25 mm by 8.55%, 52.81%, and 93.27%, respectively, while soil aggregates with R ≥ 0.25 mm decreased by 2.73%, 16.88%, and 29.81%, respectively.

3.2. Effects of Residual Geomembrane on Soil Organic Carbon Content

Figure 4 shows the relationship between soil aggregate size fraction and organic carbon content under residual geomembrane influence. The figure indicates that at equivalent residual geomembrane concentrations, soil aggregates with particle sizes between 0.25 mm and 1 mm exhibit the highest organic carbon content, while aggregates larger than 2 mm show the lowest content. As residual geomembrane particle concentration increases, organic carbon content decreases across all soil aggregate size fractions.

3.3. Effects of Different Residual Geomembrane Concentrations on Soil Evaporation

Figure 5 illustrates the evaporation process classification for soil samples. Taking a 2 cm thick soil sample with 5% residual geomembrane content under the first dry–wet cycle as an example, the evaporation process is divided into three stages: rapid evaporation stage, deceleration evaporation stage, and residual moisture evaporation stage.

Figure 6 shows the temporal evolution of evaporation rates for soil samples with varying residual geomembrane concentrations and thicknesses under dry–wet cycles. Specifically, under identical dry–wet cycling parameters, thicker soil samples exhibited lower initial evaporation rates. For samples of equal thickness, the initial evaporation rate (recorded at 2 h) increased progressively with increasing residual geomembrane content. For a 1 cm thick soil sample, the rapid evaporation phase lasted 24 h, 18 h, 16 h, and 16 h during the first dry–wet cycle for samples containing 0%, 1%, 3%, and 5% residual geomembrane, respectively. Under the same dry–wet cycle and residual geomembrane concentration conditions, the thicker the soil sample, the lower the initial evaporation rate and the longer the duration of the rapid evaporation stage.

Figure 7 shows the moisture content variation characteristics of soil samples with different residual geomembrane contents and thicknesses under dry–wet cycles. The figure indicates that the moisture content of soil samples undergoes a rapid decline phase, a slow decline phase, and a relatively stable phase. The duration of the rapid water loss stage decreased with decreasing soil thickness, with the 1 cm thick specimen having the shortest duration, followed by the 2 cm and then the 4 cm thick specimens.

Figure 8 shows the residual moisture content of soil samples with varying residual geomembrane concentrations and thicknesses under dry–wet cycles. The figure indicates that for the same dry–wet cycle and soil sample thickness, residual moisture decreases as residual geomembrane concentration increases. Conversely, for the same residual geomembrane concentration under the same dry–wet cycle, residual moisture increases with greater soil sample thickness. Under three dry–wet cycles, residual geomembrane significantly reduced residual moisture content, with reductions of 4.48–29.37%, 7.14–21.92%, and 4.19–35.95%, respectively.

3.4. Effects of Residual Geomembrane on Soil Crack Formation

Figure 9 shows the variation in crack ratio over time for soil samples of different thicknesses and residual geomembrane content under dry–wet cycles. The figure reveals that during sample cracking, the crack ratio curve generally exhibits two distinct phases: a rapid cracking stage and a slow cracking stage that eventually stabilizes. For samples of identical thickness, the initial cracking time after the second and third dry–wet cycles occurred earlier than that after the first cycle. Under identical dry–wet cycling conditions and soil sample thicknesses, a higher residual geomembrane content resulted in a higher crack ratio at the end of cracking. Taking a 1 cm thick soil sample during the first dry–wet cycle as an example: after cracking completion, the crack ratio for the residual geomembrane-free sample was 6.39%, while samples containing 1%, 3%, and 5% residual geomembrane exhibited crack ratios of 6.74%, 7.03%, and 7.21%, respectively. Compared to the residual geomembrane-free sample, the crack ratio at the end of cracking increased by 5.48%, 10.02%, and 12.83% for samples containing 1%, 3%, and 5% residual geomembrane, respectively, indicating that adding residual geomembrane effectively promotes soil cracking. Under the first dry–wet cycle, taking the 5% residual geomembrane sample as an example, the crack ratios at the end of cracking for soil samples with thicknesses of 1 cm, 2 cm, and 4 cm were 7.21%, 6.88%, and 6.11%, respectively, indicating that a thicker soil sample resulted in a lower final crack ratio.

Figure 10 shows the final crack ratio of soil samples with varying residual geomembrane content and thickness after cracking under dry–wet cycles. Under the same dry–wet cycle conditions, while the soil sample thickness is the same, the final crack ratio of the soil sample increases with the increase in residual geomembrane content. While the residual geomembrane content in the soil remains unchanged, the final crack ratio decreases with the increase in soil thickness. With identical soil sample thickness and residual geomembrane content, the final crack ratio of soil samples increases with the number of dry–wet cycles. Under three dry–wet cycles, residual geomembrane significantly increased the crack ratio, with increases of 5.33–58.89%, 0.97–63.39%, and 0.87–72.46%, respectively.

Soil shrinkage cracks have self-similarity characteristics, and fractal dimension can quantitatively characterize the complexity of crack networks. According to Formula (7), the fractal dimension of the sample after cracking can be calculated. Figure 11 shows the fitting curve of the initial fractal dimension under the first dry–wet cycle with SST = 1 cm and RGC = 3%. The absolute value of the fitted slope is the fractal dimension under the box-counting calculation method.

Figure 12 shows the relationship between the fractal dimension of soil samples with varying residual geomembrane content and thickness over time under dry–wet cycles. The figure reveals that the fractal dimension of the samples generally exhibits two distinct phases: a rapid increase phase followed by a slow increase phase. Taking the first dry–wet cycle with a 4 cm thick soil sample as an example, the fractal dimension at the end of cracking was 1.40657 for the residual geomembrane-free sample. Samples containing 1%, 3%, and 5% residual geomembrane exhibited fractal dimensions of 1.50524, 1.58795, and 1.62118, respectively, at the end of cracking. Compared to the residual geomembrane-free sample, the fractal dimensions of samples containing 1%, 3%, and 5% residual geomembrane increased by 7.01%, 12.90%, and 15.26%, respectively, at the end of cracking. During the second dry–wet cycle, using the 5% residual geomembrane sample as an example, the fractal dimensions at the end of cracking for soil samples with thicknesses of 1 cm, 2 cm, and 4 cm were 1.84845, 1.71471, and 1.65147, respectively.

Figure 13 shows the final fractal dimension of soil samples with varying residual geomembrane content and thickness after cracking under dry–wet cycles. Under the same dry–wet cycling conditions and with equal soil sample thickness, the final fractal dimension of soil samples increases with rising residual geomembrane content. Under the same dry–wet cycling conditions and with identical residual geomembrane content, the final fractal dimension of soil samples decreases as soil sample thickness increases. With consistent soil sample thickness and identical residual geomembrane content, the final fractal dimension of soil samples increases with the number of dry–wet cycles performed. Under three dry–wet cycles, residual geomembrane significantly increased the fractal dimension of the sample cracks, increasing by 0.50–15.26%, 2.92–15.96%, and 3.22–13.33%, respectively.

4. Discussion

Under the influence of residual geomembrane, water evaporation and drying cracking change synchronously and interact with each other. Soil water retention and evaporation are governed by both soil intrinsic properties and external environmental conditions. Accurately quantifying the mass transfer during soil moisture evaporation is crucial for improving understanding of the hydrological cycle and for agricultural purposes [27]. In this study, without altering the internal soil properties, an increase in soil thickness resulted in a decreased soil water evaporation rate (Figure 6) and a significantly higher residual moisture content (Figure 8). This phenomenon is attributed to the shorter water diffusion path in thinner soil layers, which facilitates water loss. Conversely, thicker soil samples increase the resistance and length of the water diffusion path, thereby retarding water loss. Residual geomembrane significantly affects soil behavior, including fracture rate and water retention capacity [28]. Our results demonstrate that the presence of residual geomembrane in soil influences the evaporation process, with a more pronounced effect at higher concentrations. Soil cracking is a phenomenon characterized by surface failure due to localized stress concentration induced by internal water loss and the formation of weak points. The extent of stress concentration is influenced by moisture content and soil structural characteristics. As shown in Figure 14a, the incorporation of residual geomembranes, which fill the internal soil pores, reduces soil density and overall cohesion. Nevertheless, residual geomembranes do not act solely as inert fillers; they significantly affect soil structural stability and water transport behavior during drying–wetting cycles. Under cyclic drying–wetting conditions, the mechanism by which residual geomembrane influence soil evaporation and cracking can be summarized as follows: the prepared loose soil specimens contained initial interparticle fissures. After the addition of residual geomembrane particles, a portion of these initial pores were filled. Following simulated rainfall that saturated the specimens, a temporary equilibrium was established in the soil–water–residual geomembrane system. As evaporation progressed, the specimen surface entered the initial cracking stage, dominated by primary cracks, while the hydrological evaporation remained relatively low. With rapid moisture evaporation, the width, depth, and length of primary cracks increased, and secondary cracks began to develop. residual geomembrane particles embedded between soil particles disrupted the original cohesion among soil particles, thereby enhancing crack propagation. Simultaneously, the cracks provided new pathway for water evaporation [29], which strongly promote water evaporation [30]. Previous studies have found that the ratio of the diffusion coefficient of cracked soil to that of intact soil ranges from several times to tens of times [31]. Upon the completion of cracking, the crack ratio reached its maximum, with only residual moisture remaining in the soil specimen. When evaporation and cracking concluded, simulated rainfall was reapplied. The cracks served as rapid infiltration channels for water, leading to swift downward percolation and the gradual closure of most cracks; however, some primary cracks did not fully close, impairing the restoration of surface continuity.

Furthermore, Figure 14b illustrates the mechanism diagram of residual geomembrane particles in soil evaporation and cracking process under dry–wet cycles. Broken geomembrane residues are distributed randomly within soil matrix. They occupy soil pore space, cut off the contact between soil particles, and reduce the interparticle cohesion and friction. Under alternating drying and wetting conditions, the weak structural surface formed by geomembrane fragments becomes the preferential expansion path of tensile stress, which promotes the initiation and expansion of soil cracks from a mechanical perspective. During evaporation and cracking, residual geomembrane become embedded within soil aggregates through physical abrasion and other interactions, destabilizing the aggregates and leading to a generally looser soil structure with larger cracks. Following the disruption of soil aggregates, the previously encapsulated organic carbon became exposed to the environment, significantly increasing its contact area with microorganisms. This accelerated microbial decomposition, promoting the mineralization of organic carbon into CO₂ and directly leading to a decline in soil organic carbon content.

This study, through controlled laboratory experiments, systematically elucidates the key mechanisms by which residual geomembrane affect soil water evaporation and cracking processes. It innovatively clarifies the interrelationships among residual geomembrane, soil, water, and aggregates, providing a theoretical basis for understanding the physical and ecological effects of residual geomembrane in soil systems. However, discrepancies exist between the experimental conditions and actual field environments, necessitating future validation through in situ field observations. On a practical level, this study indicates that residual geomembrane pollution in farmland may exacerbate soil drought vulnerability and carbon loss risks, highlighting implications for agricultural water management and soil carbon sequestration practices. Future research should focus on the interfacial processes between residual geomembrane and soil components at the microscopic scale and enhance long-term studies on the effects under coupled environmental factors to support comprehensive soil health management.

5. Conclusions

The effects of different residual geomembrane concentrations on soil aggregates, organic matter, evaporation, and cracking were investigated, and the following conclusion was reached.

Compared to residual geomembrane-free samples, soil aggregates with R < 0.25 mm increased by 8.55%, 52.81%, and 93.27% for 1%, 3%, and 5% residual geomembrane additions, soil aggregates with R > 0.25 mm decreased by 2.73%, 16.88%, and 29.81%, respectively. Meanwhile, residual geomembrane accelerates the disintegration of soil aggregates, thereby weakening their protective effect on organic carbon and leading to a decrease in organic carbon content.

During each dry–wet cycle, compared to the control group, residual geomembrane significantly reduced residual moisture content, reducing it by 4.48–29.37%, 7.14–21.92%, and 4.19–35.95%, respectively. The final crack ratio increased by 5.33–58.89%, 0.97–63.39%, and 0.87–72.46%, respectively. The final fractal dimensions increased by 0.50–15.26%, 2.92–15.96%, and 3.22–13.33%, respectively.

With increasing dry–wet cycle iterations, the duration of the rapid evaporation phase generally shortened. Multiple dry–wet cycles accelerate soil moisture loss by altering pore structure and water migration pathways, promoting the expansion and complexity of crack networks, leading to sustained deterioration of soil structural stability.