In traditional studies on grassland degradation drivers, researchers often lacked the flexibility to selectively consider driving factors and quantitatively depict their contributions. Interpretable machine learning offers a solution to these challenges. This study focuses on Inner Mongolia, China, incorporating four categories and sixteen specific driving factors, and employing four machine learning techniques (Logistic Regression, Random Forest, XGBoost, and LightGBM) to investigate regional grassland changes. Using the SHAP approach, contributions of driving factors were quantitatively analyzed. The findings reveal the following: (1) Between 2015 and 2020, Inner Mongolia experienced significant grassland degradation, with an affected area reaching 12.12 thousand square kilometers. (2) Among the machine learning models tested, the LightGBM model exhibited superior prediction accuracy (0.89), capability (0.9), and stability (0.76). (3) Key factors driving grassland changes in Inner Mongolia include variations in rural population, livestock numbers, average temperatures during the growth season, peak temperatures, and proximity to roads. (4) In eastern and western Inner Mongolia, changes in rural population (31.4%) are the primary degradation drivers; in the central region, livestock number changes (41.1%) dominate; and in the southeast, climate changes (19.3%) are paramount. This work exemplifies the robust utility of interpretable machine learning in predicting grassland degradation and offers insights for policymakers and similar ecological regions.

Wind erosion poses a major threat to ecosystem stability and land productivity in arid and semi-arid regions. Accurate identification of its spatiotemporal dynamics and underlying driving mechanisms is a critical prerequisite for effective risk forecasting and targeted erosion control. This study applied the Revised Wind Erosion Equation (RWEQ) model to assess the spatial distribution, interannual variation, and seasonal dynamics of the Soil Wind Erosion Modulus (SWEM) across Inner Mongolia from 1990 to 2022. The GeoDetector model was further employed to quantify dominant drivers, key interactions, and high-risk zones via factor, interaction, and risk detection. The results showed that the average SWEM across the study period was 35.65 t·ha⁻¹·yr⁻¹ and showed a decreasing trend over time. However, localised increases were observed in the Horqin and Hulun Buir sandy lands and central grasslands. Wind erosion was most intense in spring (17.64 t·ha⁻¹·yr⁻¹) and weakest in summer (5.57 t·ha⁻¹·yr⁻¹). Gale days, NDVI, precipitation, and wind speed were identified as dominant drivers. Interaction detection revealed non-linear synergies between gale days and temperature (q = 0.40) and wind speed and temperature (q = 0.36), alongside a two-factor interaction between NDVI and precipitation (q = 0.19). Risk detection indicated that areas with gale days > 58, wind speed > 3.01 m/s, NDVI < 0.2, precipitation of 30.17–135.59 mm, and temperatures of 3.01–4.23 °C are highly erosion-prone. Management should prioritise these sensitive and intensifying areas by implementing site-specific strategies to enhance ecosystem resilience.

Identifying and ranking the key drivers of grassland conversion at the county level is crucial for developing targeted policies and improving protection efficiency. However, this process faces methodological challenges because of spatial and temporal variability. Partial order theory offers a robust framework for addressing these complexities. This study applies partial order theory (POT) combined with the Hasse diagram technique (HDT) to analyze grassland conversion in the Hohhot–Baotou–Ordos region during two time periods (2000–2010 and 2010–2020). First, patterns of grassland transformation are quantified, and the dominant driving factors of grassland conversion out (GCO) are identified and ranked, highlighting regional differences and temporal shifts. By integrating POT and HDT, this study offers a novel approach to handling complex, nonlinear, and hierarchical relationships among multiple drivers. The results provide scientific insight and policy recommendations for region-specific grassland management and sustainable land-use planning. The results show that (1) transitions between grasslands and other land-use types became more frequent across the two periods. Specifically, the rates of grassland conversion out and conversion increased from 2.1% and 3.5% during the period 2000–2010 to 4.7% and 4.8% during the period 2010–2020, respectively. (2) Urbanization was the primary driver of grassland conversion in 11 and 10 of the 18 counties during the first and second periods, respectively, followed by factors related to weather variables. (3) In the future, the eastern region of the study area needs to prioritize mitigating the impacts of urban development, while the western region should focus on enhancing ecological construction projects. This study recommends adopting region-specific ecological protection and economic strategies for balanced outcomes in conservation and development.

Wind erosion is a multidimensional, dynamic process driven by natural and anthropogenic factors, but existing statistical methods struggle to capture its complex nonlinear relationships, resulting in incomplete quantification of drivers and their spatial variability. To address this, we integrate the Revised Wind Erosion Equation (RWEQ)model with explainable artificial intelligence to disentangle the spatiotemporal positive and negative effects of dominant drivers and their synergistic interactions in Inner Mongolia. Results show that, from 2000–2022, wind erosion has been decreasing on average by 1.1 t·ha⁻¹·yr⁻¹, mainly in the western deserts and locally in Hulunbuir sandy land. Severe erosion is mostly due to nature (78.7%) rather than anthropogenic (21.3%). Normalized difference vegetation index (NDVI), clay content (CL), windy days (WD), precipitation (PRE), temperature (TEM), and sand content (SA) were found to be the most important drivers of wind erosion. Critical threshold conditions for severe wind erosion are NDVI < 0.14, CL < 12%, GD > 26, PRE < 73.15 mm, and SA > 66%. When there is a certain combination of variables, wind erosion risk is greatly increased, which mainly happens in the western part of Alxa, Bayannur, and the area near the desert edge. Wind erosion control should shift toward region-specific precision management, including engineering protection, optimized grazing management, and vegetation restoration.

Spatial imagery (SI) provides an important framework for understanding how rural spatial transformation affects residents’ spatial cognition and supports sustainable rural planning. However, existing studies have mainly examined static spatial sections, with limited attention to longitudinal changes in residents’ SI, especially in micro-community spaces. This study takes Naobao village as a case and uses cognitive maps (CMs) and interviews to compare residents’ CM drawing processes, SI elements, and SI structures between the old and new villages. The results show that: (1) The old village follows a point-and-line-to-area CM drawing pattern, while the new village shows an area-oriented pattern from local areas to the whole. (2) In terms of SI elements, districts and paths dominate in the old village, whereas districts and landmarks are more prominent in the new village. (3) Regarding SI structure, the old village is mainly characterised by sequential CMs with relatively higher-level structures, while the new village is dominated by spatial CMs with relatively lower-level structures. This study developed a three-dimensional SI framework within the context of rural spatial transformation, which expands the theoretical framework and applicability of the SI model and provides guidance for spatial planning, community renewal, and sustainable development in similar rural tourism destinations.

The Hulunbuir Grassland is located in the semi-arid region of northern China and is highly sensitive to environmental changes, with sandy inlays within the grassland and heavy desertification problems. However, there has been a lack of systematic research on the long-term distribution pattern of desertification and influencing factors. This study adopted the multi-indicator fusion desertification level extraction method to explore the temporal distribution characteristics, spatial distribution pattern, and desertification center of the gravity migration pattern of the Hulunbuir sands and applied the parameter optimal geographic detector to generate the optimal discretized factors from the continuous meteorological and topographical factors to detect the main driving factors and mechanism of desertification of Hulunbuir Grassland over its entire range. The results showed that the explanatory power of meteorological factors was significantly stronger than that of topographic factors in the entire area of Hulunbuir Grassland. Among meteorological factors, annual precipitation had the strongest explanatory power for desertification; meteorological and topographic factors were nonlinearly enhanced, and the interaction between precipitation and slope had the strongest explanatory power, indicating that the interaction between precipitation and slope influenced vegetation and soil humidity and thus indirectly influenced the desertification evolution process.