Introduction

In agricultural production, drought-induced water stress is one of the primary adverse factors affecting crop growth, causing annual losses of up to 8 billion dollars globally due to water scarcity1. Against the backdrop of global climate change, agricultural drought and water scarcity have become global challenges that constrain agrarian development2. On the other hand, under intense rainfall conditions, inappropriate farming practices can exacerbate soil erosion, leading to water pollution, soil degradation, and reduced land productivity3. Soil erosion caused by agricultural production has emerged as a global challenge for environmental protection and a research hotspot in global climate change4. Overall, it is deemed necessary in agricultural production processes to comprehensively consider the importance of soil moisture conservation and soil erosion reduction, aiming to enhance the sustainability of agricultural production.

Red soil, a unique soil type in tropical and subtropical regions, is widely distributed in southern China’s mountainous and hilly areas, primarily found in provinces including Yunnan, Fujian, Jiangxi, and Hunan. The total area extends to 2.18 million km2, accounting for approximately 22.7% of the country’s total land area5. As a typical rain-fed agricultural area, sloped farmland constitutes the primary agricultural land composition in the red soil region. It serves as a crucial production base for economic and food crops in China. However, due to the concentrated distribution of rainfall, high erosive force of rainfall, low soil conservation effectiveness, and unique geographical features, the sloped farmland in the red soil region often experiences severe soil erosion during the rainy season6. Furthermore, due to uneven temporal and spatial distribution of rainfall and the impacts of climate change, this region is also prone to seasonal droughts in China, with seasonal droughts occurring frequently from March to June each year7. In addition, the uneven distribution of water resources and inadequate irrigation facilities in the region contribute to insufficient soil moisture supply, increasing the risk of crop water stress. Therefore, enhancing the soil’s water retention capacity and conservation effectiveness has become a vital issue for the sustainable utilization of red-sloped farmland, requiring further in-depth research and solutions8,9.

To address the challenges of frequent seasonal droughts and severe soil erosion in red soil sloped farmland utilization, current practices primarily involve two approaches: engineering measures and agronomic practices. These approaches aim to alleviate inadequate water supply for crop growth during seasonal droughts and enhance the soil’s erosion resistance on sloped surfaces. Engineering measures typically include water-saving irrigation techniques and slope runoff harvesting engineering measures10. On the other hand, agronomic practices involve cultivation practices, surface coverings, increased application of organic fertilizers, and the selection of drought-resistant crops11. Due to the complex terrain of red soil sloped farmland distribution areas, implementing engineering measures can be costly and disruptive during construction, while agronomic practices are considered an economically effective method for drought resistance and soil conservation on sloped red soil farmland12. As a critical component of agronomic strategies, tillage practices influence the boundary conditions of soil water and solute transport, thereby altering the water storage and soil conservation effects (WSE)13,14. Research has shown that no-tillage practices can increase surface coverage by incorporating crop residues, thereby reducing soil moisture evaporation and enhancing soil water retention capacity15. Plastic mulching (PM) can significantly decrease soil moisture evaporation, improve water use efficiency, and enhance soil water storage capacity16. Cross-slope ridge tillage (RT) can intercept surface runoff amount (SRA) through ridge formation, effectively preventing soil and nutrient loss17. However, conventional tillage (CT) can disrupt soil aggregate structure in heavy rainfall conditions, diminishing soil conservation effectiveness and exacerbating soil erosion processes18. Therefore, in recent years, the adoption of no-tillage, PM, and RT has become more prevalent in improving water retention or soil conservation effects in red soil slope farmland. However, there is still limited research on the comprehensive impacts of these tillage practices on WSE in red soil slope farmland19,20,21. The WSE of four typical tillage practices commonly used in red soil slope farmland in southern China—namely, CT, downslope ridge tillage (DT), RT, and PM—are poorly understood.

In this study, our work focuses on red soil slope farmland and utilizes a combination of experimental plot positioning, field observations, and artificial rainfall simulation tests. We aim to analyze the effects of plastic mulching (PM), cross-slope ridge tillage (RT), downslope ridge tillage (DT), and conventional tillage (CT) on WSEs. Our goal is to provide a scientific basis for optimizing agronomic practices and enhancing WSE in southern China’s red soil slope farmland. The specific objectives of this study are: (1) to quantify the differences in soil water retention and infiltration performance among different tillage practices in red soil slope farmland; (2) to analyze the runoff and sediment amount characteristics of red soil slope farmland under different tillage practices, and to (3) determine the most effective tillage practices for improving WSE in the region.

Results

Soil water retention capacity under different tillage practices

By using soil temperature, soil moisture content, bulk density, and total porosity as indicators, the study reflects the soil water retention performance under the four different tillage practices (Fig. 1). In this study, the effects of tillage practices and growth stages, along with their interactions, on various soil water retention performance indicators were all significant (p < 0.05). Under the same growth stage conditions, compared to the other three tillage practices, PM increased soil temperature, soil moisture content, and total porosity by 4.00% to 16.67%, 0.68% to 18.23%, and 1.30% to 13.47% (p < 0.05) respectively, while decreasing bulk density by 0.03% to 14.29% (p < 0.05). Additionally, RT, compared to DT and CT, increased soil temperature by 4.17% to 5.56% (p < 0.05), while the soil moisture content showed an increase of -0.87% to 9.01% (p > 0.05). The bulk density increased by 6.90% during the early growth stage (p < 0.05) and decreased by 4.30% to 8.25% during the maturity stage (p < 0.05). Furthermore, CT compared to DT showed increases in soil moisture content and total porosity of -1.75% to 9.63% (p > 0.05) and -7.35% to 9.14% (p < 0.05) respectively, while the soil temperature decreased by 1.40% to 4.35% (p < 0.05). There was no significant change in bulk density between the two tillage practices (p > 0.05). The results indicate that PM can increase soil porosity, reduce soil bulk density, decrease soil moisture evaporation, and enhance soil insulation effects, making it the most effective practice for water retention among the four tillage practices. Additionally, among the three tillage practices, RT, DT, and CT, RT also demonstrates stronger soil water retention.

Figure 1
figure 1

Changes of soil water retention parameters under different tillage practices. Secondly, different lowercase letters represent significant differences (p < 0.05) among various tillage practices within the same growth stage. In the analysis of variance, P: tillage practices, T: growth stages, and P × T: the interaction between tillage practices and growth stages. * and ** indicate that the effects of the variables reached significance levels of 0.05 and 0.01, respectively.

Full size image

Infiltration characteristics of soil water under different tillage practices

As the growth stage progresses, the time required to reach a stable infiltration rate becomes longer. Under the four tillage practices, changes in soil infiltration rate over time can be roughly divided into three stages: rapid decline, gradual decrease, and stable infiltration phase. Specifically, the rapid decline stages during the early, middle, and maturity growth stages occur at 0–40 min, 0–70 min, and 0–90 min, respectively. In the periods of 40–90 min, 70–100 min, and 90–110 min, the infiltration rates gradually level off, ultimately reaching a stable state after 90 min, 100 min, and 110 min (Fig. 2).

Figure 2
figure 2

Changes in soil infiltration rate under different tillage practices.

Full size image

The effects of tillage practices, growth stages, and their interactions on various soil infiltration parameters were all found to be highly significant (p < 0.01) (Table 1). In comparison to initial infiltration rates, the stable infiltration rates of PM and RT decreased by 69.50% to 92.80% and 84.92% to 94.80%, respectively, while DT and CT showed reductions of 29.08% to 82.32% and 28.68% to 70.92%. The results indicate that the variations in PM and RT are relatively substantial, while the variations in CT and DT are relatively small. In the early growth stage, the initial infiltration rate of the PM (101.86 mm/min) was the highest (p < 0.05), but its stable infiltration rate (7.33 mm/min) was the lowest, consistent with CT (p < 0.05). In contrast, the stable infiltration rate of DT (14.46 mm/min) was the highest (p < 0.05). In the middle growth stage, the initial infiltration rate of RT (74.74 mm/min) was the highest (p < 0.05), but its stable infiltration rate after stabilization (3.89 mm/min) was the lowest (p < 0.05). Conversely, CT had the highest stable infiltration rate (19.62 mm/min) (p < 0.05), while the initial infiltration rate (22.92 mm/min) and average infiltration rate (10.83 mm/min) of PM were the lowest (p < 0.05). In the maturity, the initial infiltration rate of the PM (73.34 mm/min) was the highest (p < 0.05), and the average infiltration rate of RT (14.62 mm/min) was also the highest (p < 0.05). The stable infiltration rate of CT (9.35 mm/min) was the highest (p < 0.05), while all infiltration parameters for DT were the lowest (p < 0.05).

Table 1 Characteristic parameters of soil infiltration under different tillage practices.
Full size table

The soil saturated hydraulic conductivity reflects the soil’s water infiltration capacity under stable conditions. Except for the early growth stage, the saturated hydraulic conductivity of CT (9.81 mm/min, 4.68 mm/min) increased by 38.46% to 181.09%, 19.08% to 405.67%, and 160.00% to 227.00% (p < 0.05) compared to PM (3.49 mm/min, 3.38 mm/min), RT (1.94 mm/min, 3.93 mm/min), and DT (3.00 mm/min, 1.80 mm/min), respectively.

Cumulative soil water infiltration amount characteristics under different tillage practices

During the initial stage of infiltration, the cumulative soil water infiltration amount (CIA) increases rapidly, but as time progresses, the incremental increase in CIA gradually decreases, eventually entering a phase of increasing stable infiltration rates (Fig. 3). In the early stage of crop growth, the CIA under RT is 42.53%, 17.11%, and 22.73% higher compared to PM, DT, and CT, respectively (Fig. 3a). In the middle stage of growth, for the first 40 min, the CIA under RT is 134.73% higher than that of PM and 7.26% and 9.82% higher compared to DT and CT, respectively. For 40 min to 120 min, CT shows an increase of 144.67%, 75.29%, and 46.13% compared to PM, RT, and DT, respectively (Fig. 3b). In maturity, the CIA under RT is 50.24%, 532.90%, and 105.28% higher than that under PM, DT, and CT, respectively (Fig. 3c). The comprehensive analysis suggests that RT is more conducive to increasing the rainfall infiltration in red soil slope farmland, thereby improving the soil water holding capacity of the slope farmland and ultimately enhancing drought resistance.

Figure 3
figure 3

Changes in cumulative soil water infiltration amount under different tillage practices.

Full size image

Surface runoff/interflow characteristics of different tillage practices under rainfall conditions.

There is a certain fluctuation in the changes of SRA and interflow amount (IA) during rainfall events. Initially, both SRA and IA are relatively small. However, as the rainfall continues, both SRA and IA gradually increase. After reaching a critical time point of rainfall duration, the SRA and IA stabilize (Fig. 4). The critical moments when SRA and IA reach stability under rainfall intensities of 0.94 mm/min, 1.15 mm/min, 2.35 mm/min, and 2.61 mm/min are 45 min, 30 min, 20 min, and 15 min respectively. This indicates that under the same rainfall duration conditions, the higher the rainfall intensity, the shorter the time needed for SRA and IA to reach a stable flow state. Furthermore, before reaching a stable flow state, the IA shows a relatively faster growth trend than SRA.

Figure 4
figure 4

Changes of surface runoff/interflow during rainfall. a, b, c and d refer to the change processes of surface runoff/ interflow under the conditions of rainfall intensity of 0.94 mm/min, 1.15 mm/min, 2.35 mm/min and 2.61 mm/min, respectively.

Full size image

With the increased rainfall intensity, both SRA and IA show an increasing trend. Furthermore, in terms of the total runoff produced in four rainfall events, the proportions of SRA generated under PM, RT, DT, and CT are 68.15%, 9.42%, 90.83%, and 16.69% of the total runoff, respectively (Table 2). Additionally, the SRAs generated from DT (47.64L, 56.21L) under moderate rainfall intensity conditions (experimental rainfall intensities of 0.94, 1.15 mm/min) were found to increase by 0.98–1.5 times, 18.52–37.24 times, and 13.98–37.5 times compared to PM (24.08L, 22.45L), RT (2.44L, 1.47L), and CT (3.18L, 1.46L), respectively (p < 0.05). Under heavy rainfall conditions (experimental rainfall intensities of 2.35, 2.61 mm/min), the increases were -0.06–0.49 times, 8.22–15.11 times, and 1.30–15.4 times (p < 0.05). Meanwhile, the IAs from RT (42.91L, 57.42L) were observed to increase by 1.18–1.27 times, 57.59–83.14 times, and -0.17–0.78 times in comparison to PM (19.65L, 25.34L), DT (0.51L, 0.98L), and CT (51.70L, 32.31L) under moderate rainfall conditions (p < 0.05). In heavy rainfall conditions (experimental rainfall intensities of 2.35, 2.61 mm/min), the increases were 0.10–2.59 times, 2.46–5.52 times, and -0.65–0.72 times (p < 0.05). The results indicate that surface runoff is the main form of runoff during rainfall events for PM and DT, while interflow is the dominant form of runoff for RT and CT. RT demonstrates the potential to reduce SRA effectively, promote IA formation, and enhance soil water retention capacity.

Table 2 Total surface runoff and interflow in the test plot under different rainfall intensity conditions.
Full size table

Soil erosion characteristics of different tillage practices under rainfall conditions

The variation in soil erosion amounts during rainfall events exhibits considerable fluctuation, and the magnitude of these variations becomes more pronounced with increasing rainfall intensity. Under moderate rainfall conditions (experimental rainfall intensity of 0.94 mm/min), the peak soil erosion value for DT (0.813 g) was 0.38 times, 19.33 times, and 3.28 times higher compared to PM (0.590 g), RT (0.040 g), and CT (0.190 g), respectively (Fig. 5a). The cumulative soil erosion amounts increased by 0.46 times, 29.02 times, and 6.84 times (Fig. 5b). Under heavy rainfall conditions (with an experimental rainfall intensity of 2.35 mm/min), the peak soil erosion value for PM (55.73 g) was 795.14 times, 0.81 times, and 21.65 times higher compared to RT (0.07 g), DT (30.80 g), and CT (2.46 g), respectively (Fig. 5c), with cumulative soil erosion amounts increasing by 346.27 times, 1.38 times, and 34.34 times (Fig. 5d). The results demonstrate that RT exhibits the best soil conservation effect, while PM and DT show less effective soil conservation effects.

Figure 5
figure 5

Changes of surface soil erosion during typical rainfall processes. a and b respectively represent changes of surface soil erosion and cumulative erosion under the condition of rainfall intensity of 0.94 mm/min (moderate rain); c and d represent the change process of surface soil erosion and cumulative erosion under the condition of rainfall intensity of 2.35 mm/min (heavy rain), respectively.

Full size image

Discussion

Effect of tillage practices on the soil water retention capacity of red soil slope farmland

Soil moisture serves as the medium for nutrient cycling and movement within agroecosystems, and crop growth is constrained by moisture conditions22,23. Soil pore size is a primary factor in measuring soil water retention capacity, where smaller soil bulk density and greater total porosity generally indicate increased soil water holding capacity24. Therefore, improving the soil pore structure of red soil slope farmland is crucial for enhancing soil water retention capacity, especially under seasonal drought conditions. Conservation tillage practices can reduce soil bulk density and promote the formation of soil pore structure, thereby improving soil physical properties and enhancing soil water storage capacity25. The experimental results indicate that PM can significantly increase soil porosity, reduce soil bulk density, and improve soil moisture content (Fig. 1), which is consistent with the findings of Bu et al.26. This is attributed to the physical barrier in the form of surface plastic film created by PM, which effectively reduces soil moisture evaporation, improves soil insulation, enhances soil microbial community structure, increases microbial activity, facilitates the conversion of soil residues into organic matter, promotes the formation of soil pore structure, and consequently improves soil water retention capacity27. Furthermore, compared to DT and CT, RT demonstrates a higher soil moisture content (Fig. 4). This result is related to the rainfall infiltration process on sloped farmland. Generally, RT is believed to be more effective in reducing SRA and increasing soil moisture infiltration, thus improving soil water supply capacity. Consequently, under similar soil moisture evaporation conditions, RT exhibits a higher soil moisture content. In conclusion, PM and RT are both effective conservation tillage practices for enhancing soil water retention and moisture preservation in red soil slope farmland during dry seasons.

Effect of tillage practices on the infiltration performance of red soil slope farmland

Under certain conditions of rainfall and irrigation water input, the variation in soil water storage available for crop use is closely associated with surface evaporation, crop transpiration, and soil infiltration processes. Among these, soil water infiltration is one of the most fundamental processes in hydrological cycles and agricultural irrigation28. The process of soil water infiltration involves the redistribution of rainfall resources in the soil, creating favorable conditions for crop water and nutrient absorption29. Soil permeability is an essential indicator for evaluating agricultural soils’ water regulation and water retention capacity. Studies have shown that soil permeability is mainly influenced by soil texture, structural characteristics, bulk density, porosity, and crop root characteristics30. Generally, soils with well-structured and higher porosity exhibit better infiltration performance under the same soil texture conditions, while the opposite holds true for poorly structured soils with lower porosity31. The experimental results demonstrate that the variation in soil water infiltration rate during the infiltration process is relatively larger for PM and RT, compared to DT and CT, which exhibit more gradual changes in soil water infiltration rate (Fig. 2). This difference is primarily due to the influence of changes in water potential caused by the effects of molecular forces and capillary forces as the main driving forces of soil moisture transport during the initial stages of infiltration32. In the cultivation process of red soil slope farmland, the increase in total porosity is likely mainly due to the increase in capillary porosity. This could lead to a greater capillary force in PM and RT compared to other tillage practices, ultimately resulting in a higher initial infiltration rate and a more significant infiltration rate variation for these two tillage practices. Additionally, the research results indicate that under the same duration, RT shows a higher CIA, which is beneficial for increasing rainfall infiltration in red soil slope farmland (Fig. 3). This finding is in dispute with the results of Yang et al.19. This discrepancy may be attributed to the larger soil porosity in RT, and the micro-topographic features formed by this tillage practice compared to other practices that also play a role in impeding surface runoff. As a result, it facilitates the subsurface movement of interflow, enhances the soil’s infiltration capacity on sloped farmland, and contributes to better water retention and soil preservation.

Effect of tillage practices on surface runoff/interflow generation in red soil slope farmland

Surface runoff and interflow in sloped farmland are important components of watershed runoff, directly influencing the hydrological processes of the entire watershed33,34. Under rainfall conditions, surface runoff and interflow are critical driving forces for soil erosion, nutrient loss, or leaching. They are also closely linked to soil water retention and crop water utilization in sloped farmland35. The experimental results indicate that both SRA and IA in sloped farmland are initially relatively small during rainfall events. However, as the rainfall continues, the SRA and IA gradually increase, with interflow responding more quickly (Fig. 4). This phenomenon is attributed to the relatively high initial infiltration rate of the soil in red soil slope farmland during rainfall-runoff processes. Initially, the rainfall primarily satisfies soil infiltration, with the infiltrated water filling the soil pores, resulting in relatively small SRA and IA during the early stages of rainfall. As the rainfall infiltration process continues, the soil pores are progressively saturated, leading to a rapid increase in IA. Furthermore, as the rainfall persists, the adsorption and swelling of clay particles in the soil fill the existing pores, further reducing the soil infiltration rate, and only then does SRA gradually increase. Additionally, the study results indicate that PM and DT primarily result in surface runoff during rainfall events. A possible explanation is that PM inhibits water percolation, and the ridge channels in DT provide natural pathways for the collection and flow of surface runoff. Both of these tillage practices enhance the erosive power of runoff, reduce the surface roughness of sloped farmland, and facilitate the generation of surface runoff2036. Therefore, although PM on sloped farmland helps improve soil water-holding capacity, it hinders the formation of soil water reservoirs through water infiltration under rainfall conditions. As a result, PM exhibits an evident duality in water management on sloped farmland.

Effect of tillage practices on sediment yield in red soil slope farmland

The erosion process of soil on sloped farmland leads to a decline in soil quality and a decrease in land productivity. It has become one of the primary sources of water pollution, ranking among the most severe non-point source pollutions3738. Soil erosion during rainfall events is influenced by soil conditions, topography, land use, vegetation, conservation measures, and rainfall characteristics39. Regarding soil conservation, practices including contour plowing and straw mulching have shown significant effects, with straw mulching considered an effective protective measure for sloped farmland. However, the suitability of soil conservation measures varies significantly with slope steepness40. This study indicates that RT provides the most substantial soil conservation effect, while PM and DT show relatively poor soil conservation effects (Fig. 5). This is because PM and DT result in higher SRA, leading to increased erosive forces from surface runoff and thus causing severe soil erosion on the hillslope. In contrast, the microtopography created by RT facilitates infiltration of surface runoff, significantly reducing surface erosion forces and thereby mitigating the extent of soil erosion on the hillslope. Additionally, the perpendicular orientation of RT to the slope allows for the interception of sediment-laden runoff, reducing the erosive power of the runoff and protecting organic matter and nutrients in the soil from being washed away. This orientation helps to resist erosion caused by rainfall and runoff by reducing the impact of scouring.

Evaluation of water storage and soil conservation effect under different tillage practices

Climate change has increased extreme rainfall events and seasonal droughts, causing significant soil erosion on red soil slope farmland in the south during the rainy season and severe water stress for crops due to insufficient soil moisture supply in the dry season6,7. Therefore, in the sustainable management of red soil slope farmland, it is essential to focus on both water retention and soil conservation. Consequently, a comprehensive evaluation of the WSE of different tillage practices on red soil slope farmland is necessary to identify the most effective practices for WSE, aiming to promote their application in agricultural production on red soil slope farmland in the south. his study established an evaluation system focusing on WSE, presenting the comprehensive impacts of WSE for four tillage practices in Table 3. RT received the highest score in the evaluation of WSE comprehensive effects (WSEI = 0.84), while PM and DT received relatively low scores in WSE comprehensive effects evaluation (WSEI = 0.38, WSEI = 0.33). CT scored moderately in evaluating WSE comprehensive effects (WSEI = 0.75). These evaluation results suggest that, concerning WSE, RT is the preferred tillage practice to be promoted on red soil slope farmland, followed by CT, while PM and DT should be avoided as much as possible. Despite its beneficial water retention and warming effects, PM can inhibit rainfall infiltration and increase SRA due to the coverage of plastic film on the soil surface, leading to severe erosion. Therefore, when implementing PM, improvements to traditional PM techniques are necessary. For example, a combination of RT and PM could be adopted to maximize rainfall infiltration after covering with plastic film, thereby reducing the erosive effects of surface runoff.

Table 3 Evaluation matrix of the comprehensive effect of water storage and soil conservation under different tillage practices.
Full size table

Conclusions

This study utilized a combination of experimental plot positioning observations and artificially simulated rainfall experiments to investigate the WSE of PM, RT, DT, and CT. The results indicate that the effects of tillage practices, growth stages, and their interactions on soil water retention and infiltration characteristics were all significant (p < 0.05). Under the same growth stage conditions, PM can improve soil pore structure, reduce soil bulk density, and have good water retention and warming effects. However, during rainfall-runoff processes, both PM and DT mainly resulted in surface runoff, which hindered soil water infiltration and increased erosion potential, leading to severe soil erosion on slopes. In contrast, the microtopographic characteristics created by RT can impede the generation of surface runoff, thereby reducing soil erosion on slopes and facilitating the infiltration of soil moisture during rainfall events. This results in higher CIA, demonstrating superior water retention capacity compared to DT and CT. In comparison, CT was rated at a medium level. Overall, RT exhibited the best WSE, followed by CT, while DT and PM performed poorly in this aspect. Therefore, in the cultivation of red soil slope farmland in Southern China, RT should be prioritized for promotion.

Materials and methods

Study site description

This experiment is carried out in the water-saving irrigation center of Yunnan Agricultural University in Kunming, Yunnan Province, China (25°07′56″N, 102°44′51″E). The location of the test site is shown in Fig. 6. The altitude of the test site is 1,930 m, with a low latitude plateau monsoon climate. The annual average temperature is 14.9 ℃, the annual evaporation is 1856.4 mm, and the annual sunshine duration is about 2327.5 h. Moreover, the average precipitation is 1000.5 mm, mainly between May and October. The study area is a constituent part of the Southern China Red Soil Region, where red soil is the predominant soil type. Due to topographical conditions, red soil slope farmland plays a significant role in the agricultural land use in this area, accounting for 69.79% of the total arable land area40. The soil sample for testing was collected from the 0–30 mm plow layer of red soil slope farmland at the teaching experimental farm at Yunnan Agricultural University. According to the Chinese classification standard, this soil is categorized as dry-land red soil. The soil texture is clay, with mass percentages of sand particles (20 ~ 2000 μm), silt particles (2 ~ 20 μm), and clay particles (< 2 μm) at 14.56%, 35.86%, and 49.58%, respectively. The experimental soil has a pH of 6.53, with organic matter content, total nitrogen content, available phosphorus content, and available potassium content of 54.69 g/kg, 2.66 g/kg, 47.42 mg/kg, and 673.00 mg/kg, respectively.

Figure 6
figure 6

Location of the experimental site in this study. The map was created using ArcGIS 10.5 software (https://www.esri.com/en-us/arcgis/about-arcgis/overview). The primary geographic information data comes from the resource and environmental science and data center (https://www.resdc.cn/).

Full size image

Experimental design

In this work, to simulate typical tillage practices in red soil slope farmland in southern China, four tillage treatments were established: conventional tillage (CT, planting crops on flat slope without ridges), downslope ridge tillage (DT, planting crops on ridges along the slope allowing for rapid surface runoff along the ridges), cross-slope ridge tillage (RT, planting crops on ridges following contour lines), and plastic mulching (PM, covering the sloping farmland surface with plastic film on top of conventional tillage). The experiment was set up using a completely randomized design, with each treatment replicated three times (Fig. 7). However, due to multiple influencing factors, precise field observations of soil water holding capacity, infiltration performance, and erosion processes related to surface and subsurface runoff in red soil slope farmland pose particular challenges. To address this, specific simulation test plots were established to mimic the natural management processes of red soil slope farmland, facilitating the observation of dynamic changes in soil water movement and infiltration performance at different growth stages of crops. The test plots were constructed using stainless steel, with dimensions of 160 cm (length) × 80 cm (width) × 30 cm (depth). Each test plot was equipped with devices for collecting surface runoff sediment and infiltration water, as detailed in Fig. 8. Prior to filling the test plot with soil, the soil was air-dried naturally and sieved through 10 mm diameter openings to remove plant roots and stones. During filling, the soil was compacted in layers to a dry bulk density of 1.0 g/cm3, with each layer 10 cm thick, totaling 30 cm in depth (the typical plow layer thickness in red soil slope farmland ranges from 25 to 35 cm. Hence the test plot soil thickness was set at 30 cm). Before planting crops in the test plots, irrigation was carried out to ensure that soil bulk density and structure were similar to the natural conditions of red soil slope farmland after tillage and natural settling. In addition, considering the slope distribution characteristics of red soil slope farmland in southern China, the slope gradient for all test plots was set at 10°40.

Figure 7
figure 7

Test plot treatment and artificial simulated rainfall test site. a Conventional tillage (CT); b Downslope tillage (DT); c Cross-slope ridge tillage (RT); d Plastic Mulching (PM); e Photos of artificial simulated rainfall test site.

Full size image
Figure 8
figure 8

Test plot design diagram.

Full size image

Corn was selected as the representative crop model for this study, with a planting date of May 21, 2021, and a harvest date of September 28, 2021. The corn growth cycle was divided into three stages: early growth stage (May 21 to June 21), middle growth stage (June 21 to August 10), and maturity (August 10 to September 28). The crop management in each test plot strictly followed the typical practices in red soil areas, with irrigation every 15 days at a rate of 225m3/hm2 (equivalent to 22.5 mm of water). Additionally, nitrogen fertilizer (urea) was applied on June 22 and July 26 at a rate of 75 kg/hm2 each time. Considering the erosive rainfall in southern red soil areas, which mainly occurs from July to mid-August during the mid-growth stage of corn, different rainfall intensities were simulated. Four artificial rainfall events were conducted between July 26 and July 30, with intensities of 0.94 mm/min, 1.15 mm/min, 2.35 mm/min, and 2.61 mm/min (where 0.94 mm/min and 1.15 mm/min represent moderate rainfall, and 2.35 mm/min and 2.61 mm/min represent heavy rainfall), each lasting 60 min. The artificial rainfall experiments investigated the red soil slope farmland’s runoff and sediment characteristics under different tillage practices.

Artificial rainfall simulation experiments

The experiment was carried out in the artificial simulated rainfall hall in the water-saving irrigation experiment center of Yunnan Agricultural University. The artificial simulated rainfall equipment employs the NLJY-10–03 indoor simulated rainfall system developed by Nanjing Forestry University. This rainfall system can adjust the rainfall intensity and duration according to the requirements of the experiment. Most raindrops can reach the terminal speed in the effective rainfall area, and the rainfall coefficient is above 85%. Before the artificial simulated rainfall, the soil should be wet to ensure the consistency of the water content of the tested soil at the early stage (The water content of soil mass remains at about 42%.). Each artificial simulated rainfall lasts for 60 min, and the SRA and IA are measured every 5 min. A total of 500 mL of surface runoff is collected. The oven-drying method measures the amount of eroded sediment, and the runoff sediment content is calculated accordingly41.

Soil sample collection and analysis

Soil samples were collected on June 10 (early growth stage), July 18 (mid-growth stage), and September 7 (maturity). Soil samples were collected using a soil auger (100 cm3) and aluminum boxes to measure soil bulk density, total porosity, and soil moisture content. Additionally, soil temperature was measured using a JM624 thermometer. Soil bulk density and total porosity were determined using the soil auger method42, while soil moisture content was determined using the oven-drying method43.

Determination of soil infiltration characteristic parameters

Soil infiltration was determined using the indoor double-ring method44. During the three crop growth stages, undisturbed soil samples were collected using a 100 cm3 double ring. A bottom plate with mesh and filter paper was placed at the bottom of the double ring containing the undisturbed soil column, and another double ring of the same size was placed on top. The interface was sealed with transparent tape to prevent water leakage. The prepared double-ring sample was placed on a funnel stand with a water-collecting beaker underneath. Water was slowly added from the upper end of the double ring into the inner ring, keeping the water level below 1–2 cm from the upper edge of the double ring. The amount of water infiltrating through the soil column within different time intervals was measured and recorded: every 2 min for the first 10 min, then every 5 min after 20 min, until the infiltrated water amount per unit time stabilized. The calculation methods for various infiltration characteristic parameters are as follows:

The initial infiltration rate equals the CIA within the first 2 min divided by 2 min. Likewise, the infiltration rate for the first 5 min equals the CIA for the first 5 min divided by 5 min. The infiltration rate for the first 15 min equals the CIA for the first 15 min divided by 15 min. The infiltration rate for the first 30 min equals the CIA for the first 30 min divided by 30 min. Moreover, the stable infiltration rate refers to the infiltration rate as the infiltration amount in unit time tends to be stable.

The saturated hydraulic conductivity (permeability coefficient) is calculated as follows45.

$$K_{t} = \frac{{10 \times Q_{n} \times L}}{{t_{n} \times S \times \left( {h + L} \right)}} = V\frac{L}{{\left( {h + L} \right)}}$$
(1)

Where Kt represents the saturated hydraulic conductivity at t ℃, mm/min. Qn is the water infiltration amount for the nth time, mL. tn is the time interval of each infiltration, min. S is the cross-sectional area of the infiltration cylinder, cm2. h is the thickness of the water layer, cm. L is the thickness of the soil layer, cm. V is the infiltration velocity, mm/min.

In order to make the Kt values measured at different temperatures easy to compare, they should be converted into saturated hydraulic conductivity at 10℃ and calculated as follows46.

$$K_{10} = \frac{{K_{t} }}{0.7 + 0.03t}$$
(2)

Where K10 is the saturated hydraulic conductivity at 10℃, mm/min. Kt is the saturated hydraulic conductivity at t ℃, mm/min. T is the water temperature measured in the infiltration experiment, ℃.

Method for evaluating the comprehensive water storage and soil conservation effect. In order to select the optimal tillage practice for water storage and soil conservation, it is necessary to assess the WSE of different tillage practices comprehensively. After a comprehensive analysis, the water storage effect of this study is evaluated by soil moisture content (mass moisture content, %), CIA (L) for 120 min, and IA (L) for 60 min. Furthermore, the soil conservation effect is evaluated by soil conservation effect (g) from the artificially simulated rainfall experiment in the experimental plot. The following formula calculates the water storage and soil conservation effect index (WSEI)47.

$$WSEI = \sum\nolimits_{i = 1}^{n} {K_{i} \times C_{i} }$$
(3)

Where Ki is the membership value of the ith evaluation index. Ci is the weight of the ith evaluation index. In order to simplify the calculation process of comprehensive effect evaluation, the weights of the four evaluation indices are all set to 0.25, considering that the four indices are equally important for the effect of WSEI in red soil slope farmland. The larger the values of soil moisture content, CIA, and soil IA are, the better the water holding of red soil slope farmland is. The membership values of the three indices are calculated using the S-type (upper limited type) membership function formula48.

$$K_{i} = \left\{ {\begin{array}{*{20}l} {1,} \hfill & {x_{i} \ge b} \hfill \\ {0.9\left( {x_{i} - a} \right)/\left( {b - a} \right) + 0.1,} \hfill & {a < x_{i} < b} \hfill \\ {0.1,} \hfill & {x_{i} \le a} \hfill \\ \end{array} } \right.$$
(4)

Where xi is the quantized value of the ith evaluation index, as well as a and b are the lower limit and upper limit of the critical value of each index, respectively.

The greater the amount of soil erosion is, the worse the soil conservation effect of red soil slope farmland is. Therefore, the membership value of soil erosion is calculated by the inverse S-type (lower limited type) membership function formula48.

$$K_{i} = \left\{ {\begin{array}{*{20}l} {1,} \hfill & {x_{i} \ge a} \hfill \\ {0.9\left( {x_{i} - b} \right)/\left( {a - b} \right) + 0.1,} \hfill & {a < x_{i} < b} \hfill \\ {0.1,} \hfill & {x_{i} \le b} \hfill \\ \end{array} } \right.$$
(5)

Data analyses

. The primary data processing in this study was conducted using Microsoft Excel 2013 software. Statistical analysis was performed using SPSS 19.0 software, and graphs were generated using Origin 2021. Before data analysis, tests were conducted to assess the homogeneity of variance and normality of all datasets to ensure the assumptions of statistical analysis were met. Two-way analysis of variance (ANOVA) was employed to quantify the effects of tillage practices, growth stages, and their interactions on soil water retention and infiltration performance. Additionally, one-way analysis was used to evaluate the impact of tillage practices on runoff volume during rainfall events. If the mean differences are found to be significant, a least significant difference test (LSD) with p < 0.05 will be applied.