This study was conducted from May to June 2019. Considering the specification and location of the dump, three sample plots each were selected from the dump site of the Rujigou Open-pit Coal Mine (D1, D2, and D3) and the original land (G1, G2, and G3) with good vegetation growth around the dump. The dump site and the original land were under the same conditions with the same source of dump residue. The geographical locations and basic information of each site are shown in Fig. 1 and Table 1.

Excel 2013 was used to sort out, calculate and map the meteorological data and the experimental data of the physical and chemical properties of the residue. SPSS 22.0 was used to conduct One-way ANOVA for the physical and chemical properties of the soil, and Origin 2021 was used for mapping. Pix4D MAPper 2.0 was used to generate the digital surface model (DSM), digital elevation model (DEM), and digital orthophoto image (DOM), and ArcGIS 10.2 was used to process the above files to obtain the slope length, slope, and aspect of the dump slope.

Climatic factors mainly include precipitation, temperature, and wind, which are important abiotic factors affecting vegetation growth. As shown in Fig. 2, the average annual precipitation in the study area is less than 200 mm, but the potential evaporation can reach approximately 2000 mm, which is up to 10 times the precipitation. The intense evaporation aggravates the degree of drought in the study area. Moreover, the annual distribution of precipitation is uneven, mainly concentrated in May and July, with little or no precipitation in the remaining months. Therefore, the study area experiences a long dry period. Moreover, occasional heavy rainfall events occur, with precipitation reaching as high as 12.4 mm within 1 h, which is highly conducive to serious soil and water loss. Figure 3 shows that the average annual temperature in the study area is 8.45 ℃, and the average monthly temperature ranges from ‒5.65 to 21.04 ℃. Winter is long in the study area, even starting from the end of October till the end of March the next year. Therefore, the corresponding frost-free period is also short at only approximately 170 days. The annual effective accumulated temperature (temperature ≥10 ℃) is approximately 1249.09 ℃. The field investigation revealed that the plants in the study area generally emerged in April and generally stopped growing at the end of October. The short phenological period in the study area could be primarily attributed to the low temperature and less effective accumulated temperature ≥10 ℃. Figures 4 and 5 show that the average wind speed in the study area ranges from 0.39 m s^‒1 to 9.31 m s^‒1, with a relatively uniform annual distribution. The dominant wind directions are W, ENE, and SE. The maximum gust wind speed can reach 25.05 m s^‒1, and the number of days with gale exceeding level six (wind speed ≥10.8 m s^‒1) can reach more than 150 days per year. Strong winds increase drought in the study area by limiting vegetation recovery through mechanical damage to plants and reducing the survivability of seeds and seedlings. They also increase local evaporation.

The spatial distribution of vegetation and soil nutrients are closely related to topographic factors. Topographic factors primarily affect plants indirectly by changing light, heat, water, underlying surface, wind, and other conditions (Li et al., 2018). Among the terrain factors, slope, slope length, and slope aspect are commonly used to characterize the terrain. Figure 6 shows that the slope of the dump in the study area is primarily 30°-36°, reflecting a steep slope. According to the Code for Design of Soil and Water Conservation Engineering (GB 51018-2014), the slope of slope-type slag dumping fields should be lower than 25°. According to the Quality Control Standard of Land Reclamation (TD/T1036- 2013), the final slope of the dump should be 26°-28°. The larger the slope, the higher the instability, and the higher the susceptibility to low soil nutrients, soil erosion, and other problems (Yang et al., 2020b). The slope length of the dump in the study area ranges from 10 to 44 m, and the corresponding slope height ranges from 5 to 26 m. Excessively long slopes are prone to soil and water loss and more likely to incur instability issues (Gao et al., 2007). Previous studies have reported a positive correlation between slope length and annual erosion modulus, and that the slope length of sloping farmland should be controlled below 15 m (Wei and Zhu, 2002). According to the “Construction Slope Engineering Technical Code” (GB50330-2013), slopes with heights of 10-15 m correspond to safety grades one and two, for which the failure consequence is very serious and serious, respectively. Therefore, soil slopes with heights greater than 15 m are already relatively unstable. Under this standard, the high dump in the study area is not conducive to ecological restoration. As shown in Fig. 7, the slope direction of the dump in the study area is generally 15°-20° west by north, indicating shady slopes. Slope aspect is not a factor limiting ecological restoration, but has a strong impact on light, temperature, precipitation and wind, and it is the primary factor for consideration when taking vegetation measures (Yang et al., 2020a).

Soil particle size gradation refers to the mass proportion of soil particles with different particle sizes in the soil, and it is an index characterizing the soil texture. As shown in Figure 8, the gravel and clay components of the muck account for a large proportion, with gravel (73.40%±2.18%) accounting for more than 3 times the standard maximum. According to the International Classification of Soil Textures, the opencast coal mine muck of dump, and soil at G1, G2, G3 are heavy gravel (clay), light gravel (sandy clay), multi-gravel (clay), and light gravel (sandy clay), respectively. The clay and sand contents of the residue were significantly higher than those of the original geomorphic soil, while the silt content was significantly lower than that of the original geomorphic soil. In order to verify the accuracy of residuum texture, bulk density, and porosity, repeated experiments were carried out. As shown in Fig. 9, the bulk density of the residual soil was relatively large at approximately 1.55±0.01 g cm^-3, which is significantly larger than that of the original shrub land. The total porosity and capillary porosity of the residual soil were both small at approximately 30.39%± 0.83% and 21%±0.98%, respectively, which are significantly lower than those of the original shrub land by approximately 10%. According to the Land Reclamation Quality Control Standard (TD/T1036-2013), the bulk density of mine soil after reclamation should be less than 1.5 g cm^‒3, gravel content should be less than 50%, and the soil texture should be sandy or sandy loam. As shown in Table 2, no significant difference was observed between the pH of the residual soil (8.52±0.04) and the original soil (8.45±0.10), which belong to alkaline soil. Similarly, no significant difference was observed in organic matter content between the residual soil (44.87±0.74 g kg^‒1) and the original soil (47.00±0.50 g kg^‒1). Referring to the nutrient classification standard of the second National Soil Survey, both of them were far beyond the lower limit of the first level standard. The total nitrogen content of the residual soil (0.67±0.05 g kg^‒1) corresponds to the fifth grade and is significantly lower than that of the original geomorphic soil (1.47±0.21 g kg^‒1). The hydrolytic nitrogen content of residual soil (14.23±0.48 mg kg^‒1) corresponds to grade 6 and is significantly lower than that of the original geomorphic soil (40.67±7.99 mg kg^‒1). The content of available phosphorus (2.81±0.34 mg kg^‒1) in the residual soil corresponds to Grade 5 to 6, without any significant difference from that of the original soil (4.11±0.61 mg kg^‒1). The content of available potassium (118.89±5.58 mg kg^‒1) in the residual soil significantly differed from that in the original soil (222.67± 92.95 mg kg^‒1), but it was high according to the standard.

Water is an important environmental condition for plant survival. The study area is located in the arid area of northwest China, with scarce natural precipitation. Moreover, with the high altitude and large number of sunny days, this area experiences strong solar radiation, leading to intense evaporation. Even in arid areas, occasional heavy rains can increase the vulnerability of bare slopes to soil erosion and even landslides. In addition to water erosion, wind erosion also contributes to soil and water loss of slopes. Moderate winds facilitate seed and pollen dispersion and promote vegetation restoration. The mining area is affected by monsoon winds, which are often strong (≥10.8 m s^-1), and sandstorms, not only exerting damage to plants but also causing erosion of the soil surface. With these characteristics, plant seeds buried at shallow depths are likely to be blown away. In this study, the following climate limiting factors were identified: low amount and uneven distribution of precipitation, heavy rainfall, high evaporation, low temperature, less effective accumulated temperature, and large number of windy days. Among them, low precipitation and high evaporation mainly induce drought and limit the growth of vegetation, heavy rain aggravates the degree of soil and water loss, low temperature and low accumulated temperature mainly lead to short phenological period, and strong winds increase evaporation.

Climate conditions cannot be altered manually. To address the low number of water sources, the utilization rate of precipitation and water infiltration can be increased. In view of the damage to plant seedlings caused by severe evaporation and climate, direct contact between the slope and external elements can be reduced. Therefore, slope cover measures can be adopted to increase the roughness of the slope and reduce the direct contact between the slope and external factors, such that the amount of water infiltrating the soil can be increased, the amount of water evaporating from the soil can be decreased, the temperature of the surface soil on the slope can be increased, and soil erosion caused by heavy rain, large slope, and excessive slope length can be decreased (Lei et al., 2004; Myburgh and Philip, 2013). Commonly used covering materials include stone, wood, and geotechnical materials, among which vegetation blanket, as a new type of slope protection material, is widely used in ecological slope protection because of its low cost, short construction period, easy degradation, and environmentally-friendly and pollution-free characteristics (Gu et al., 2006). At the same time, vegetation measures can be implemented to reduce soil water evaporation, strengthen the stability of shallow slopes, reduce soil erosion, and improve soil physical and chemical properties. In the vegetation survey, the local natural plants were found to be primarily herbaceous, with a coverage rate of 30%-40%. Shrubs were few in number, relatively short in height, and mostly prostrate, and trees were sporadically distributed, most of which were elm and almond (Xiong et al., 2007). In addition, the slope direction of the dump in the study area is dominantly northwest. Therefore, native shrubs suitable for growing on shady slopes should be mainly selected. In conclusion, when taking vegetation measures in the study area, a combination of shrubs and grasses is recommended. Suitable plant species include Halogeton glomeratus, Setaria viridis, and Artemisia desertorum Spreng., and the target vegetation coverage rate should be set at approximately 40%. Previous studies have found that the vegetation coverage, soil moisture, total nitrogen, and available potassium contents of the Rujigou Mining Area significantly increased after setting a vegetation blanket for ecological restoration (Zhang et al., 2022).

Microclimate, light intensity, soil moisture, temperature, and evaporation widely vary under different terrain conditions (Bayat et al., 2017; Li et al., 2018), and they directly or indirectly affect abiotic and biotic factors, leading to spatial differences in vegetation and soil conditions (Huang et al., 2015; Méndez-Toribio et al., 2016; Ai et al., 2017; Li et al., 2018). Large slopes are generally prone to slope instability, soil and water loss, soil nutrient loss, texture deterioration, and soil erodibility (Yuan et al., 2017; Mao et al., 2020). Long slopes are prone to serious soil and water loss and slope instability. According to the Technical Specification for Protection and Restoration of Ecological Environment in Mines (Trial) (HJ 651-2013), slope cutting and grading should be carried out when the total height of the dump is greater than 10 m, and the height of each grade should not exceed 8 m, the width should be greater than 2 m, and the slope should be less than 35°. The height of the dump in the study area was 5-26 m, and the slope reached 35°. The topographic limiting factors of the dump slope mainly include excessive slope height and length.

The most direct and effective approach to solving slope steepness and slope length is slope cutting.According to the specification, some dumps in the study area require slope cutting and grading. However, due to the restrictions of road planning, sufficient space is not available for slope cutting and grading. Consequently, it is impossible to directly reduce slope and shorten slope length by cutting and grading. Shortening the slope length is also an effective measure to reduce soil and water loss (Wei and Zhu, 2002). The slope can be divided and upper runoff can be intercepted by setting slope level blocks, such that erosion can be weakened and infiltration increased. The main reason for the susceptibility of open-pit coal mine dumps to soil and water loss is their low consolidation rate (Bjorn et al., 1998). The consolidation rate can be increased via various approaches such as engineering measures, chemical consolidation measures, and vegetation measures. Among the physical methods, geogrid and shotcrete are commonly used. Among chemical consolidation materials, polymer water-absorbing resin cementing materials are widely used owing to their good water retention and consolidation effects (Liang et al., 2016). Consolidation through vegetation is mainly realized by plant roots in three ways: root network series, root soil bonding, and root biochemistry (Liu et al., 1996). At the same time, the Technical Code for Construction Slope Engineering (GB 50330-2013) stipulates that retaining walls can be set at the foot of slopes to increase slope stability. Previous studies have shown that restoration measures such as ecological bag blocking and honeycomb paving have significant effects on improving vegetation coverage, soil water content, total nitrogen and available potassium contents, and the models of vegetation blanket covering and ecological bag blocking present the highest effectiveness for ecological restoration (Zhang et al., 2022).

Soil provides structural support to plants, and its physical and chemical properties are of great significance for vegetation restoration. According to the common problems encountered in the residual soil of open-pit coal mine dumps, soil texture, bulk density, porosity, pH, organic matter, total nitrogen, hydrolytic nitrogen, available phosphorus, and available potassium were selected as measurement indicators (Pei et al., 2020). Texture is one of the basic properties of soil and the decisive factor of soil moisture, fertilizer retention capacity, temperature, and tillage. This study shows that the residual soil has some problems, such as high sand and clay contents, large bulk density, and small pores, which are different from the results of previous studies (Li, 2019). The poor quality of the residual soil in the dump may be attributable to the loss of finer particles during the process of mechanical loading and unloading, leaving behind gravel. In the later process of accumulation, the muck is subjected to high gravity impact, as well as heavy transport and repeated rolling, resulting in extremely compact soil. It is inferred that the muck contains large amounts of coal gangue. Owing to its excessive clay content and small pores, the muck is conducive to the formation of a water film after absorbing water, resulting in compaction, poor ventilation, and resistance to water infiltration, while promoting slope runoff and salinization (Liu et al., 2008). Moreover, the residual soil has poor capacity to retain water, fertilizers, and heat because of the high contents of sand and stone, making it more susceptible to the influence of external climate. pH is a basic property of soil and has an important impact on soil texture and nutrients. This study shows that both the residual and original soil are alkaline soil, and the optimal pH of the soil is between 6.5 and 7.5. If the pH is too high, the soil texture will be heavy and sticky, and the structure will be poor, with inappropriate nutrient contents for plant growth (Zhao, 2003; Zhao et al., 2009). The main factors affecting soil pH are soil parent material, climate, vegetation, and human factors (Biro et al., 2013). The parent material of soil formation in the Helan Mountain area is mainly gray-calcium soil, which is alkaline, and gray-calcium meadow soil itself is often accompanied by salinization (Jiang et al., 2013). In addition, evaporation is extremely high in the study area, which eventually leads to alkaline residue and high pH of the original soil. Therefore, this study concluded that pH is not a limiting factor for vegetation restoration in the dump site. The main soil nutrients absorbed by plants are nitrogen, phosphorus, and potassium. The total nitrogen content is an indicator of basic soil fertility, which exhibits little change in the short term. Hydrolytic nitrogen, available phosphorus, and available potassium contents mainly reflect the supply of nitrogen, phosphorus, and potassium in the soil in the near future, which fluctuates widely in the short term. This study showed that the content of available potassium was high, and the contents of total nitrogen, hydrolytic nitrogen, and available phosphorus in the residue were low, indicating the depletion of nitrogen and phosphorus in the residual soil. Therefore, hydrolytic nitrogen and available phosphorus were taken as limiting factors for vegetation restoration, which is consistent with the findings of other studies (Ma et al., 2008). Previous studies have shown that due to external disturbance, the substrate of abandoned mining sites often exhibit a lack of organic matter (Ma et al., 2008; Philip et al., 2010), but this study found that the residue of the waste dump has a high content of organic matter. According to the principle of the potassium dichromate volumetric method, which is one of the most commonly used methods for determining soil organic matter content at present, the residual soil has a high carbon content including not only organic carbon, but also coal gangue. However, the high carbon content does not represent a high level of soil fertility because not all carbon can be absorbed and used by plants. In terms of soil characteristics, the main limiting factors are the high contents of silt, sand and clay, and low contents of hydrolytic nitrogen and available phosphorus.

In view of the problem of large sand and stone content in the residuum, the soil screening method can be adopted to reduce the sand and stone content by sifting the residuum. The problem of excessive clay content can be solved by introducing foreign soil (Sun and Wang, 2013). Soil conditioners are widely used to adjust soil physical structure, adjust pH, and reduce heavy metals. Soil carbon, lignite, and weathered coal are rich in humic acid, organic matter, and NPK nutrients, which can effectively improve the soil structure, reduce pH, and increase soil fertility. It should be noted that the standard values in soil physical and chemical property specifications are often the optimal values in large databases, which are not necessarily applicable to every area, and large-scale soil improvement is often subject to many restrictions (Song et al., 2019). Therefore, as long as the physical and chemical properties of the modified residue are close to the standard value, ecological restoration should be successful.