Soil erosion results in land degradation and desertification in northern China. The Xilingol League of Inner Mongolia is an important part of the “Two Barriers and Three Belts”, and has been given the main function of “a windbreak and sand-fixing belt of northern China”. Accurate measuring of soil erosion moduli, analyzing the differences in soil erosion moduli across different periods and regions, are the basis for carrying out soil conservation and evaluating the effectiveness of ecological governance. Some radioisotopes are good environmental tracers because they are closely combined with the fine particles of the surface soil and are only affected by the mechanical movement of soil particles. In this paper, Taipusi Banner and Zhengxiangbai Banner, which are in the farming–pastoral ecotone in northern China, were selected as the study area. A regional reference inventory, that is, the activity of 137Cs and 210Pbex in the sample without any soil erosion, accumulation/deposition, or any kind of manual disturbances, as well as the soil erosion moduli, were determined by 137Cs and 210Pbex composite tracing technology and multiple lines of evidence. The results are as follows: (1) The regional 137Cs reference inventory was 1928 Bq∙m−2, and the regional 210Pbex reference inventory was 10,041 Bq∙m−2. (2) On a 50-year time scale, the soil erosion moduli in the study area ranged from 140 t∙km−2∙a−1 to 1030 t∙km−2∙a−1; on a 100-year scale, the soil erosion moduli in the study area ranged from 35 t∙km−2∙a−1 to 2637 t∙km−2∙a−1; the entire study area was in a lightly eroded state. (3) Compared with two periods before and after the 1970s, the southern parts (cultivated land and grassland) experienced an increasing trend in soil erosion moduli due to land reclamation, grassland grazing, and other activities. Due to weakening wind and increasing precipitation, soil erosion moduli in the northern parts (southern margin of the Hunshandake Sandy Land) slowed down. The study also discussed the uncertainty and application potential of isotope-tracing technology in sandy land of typical grasslands in northern China.

Grassland gross primary productivity (GPP) is an important part of global terrestrial carbon flux, and its accurate simulation and future prediction play an important role in understanding the ecosystem carbon cycle. Machine learning has potential in large-scale GPP prediction, but its application accuracy and impact factors still need further research. This paper takes the Mongolian Plateau as the research area. Six machine learning methods (multilayer perception, random forest, Adaboost, gradient boosting decision tree, XGBoost, LightGBM) were trained using remote sensing data (MODIS GPP) and 14 impact factor data and carried out the prediction of grassland GPP. Then, using flux observation data (positions of flux stations) and remote sensing data (positions of non-flux stations) as reference data, detailed accuracy evaluation and comprehensive trade-offs are carried out on the results, and key factors affecting prediction performance are further explored. The results show that: (1) The prediction results of the six methods are highly consistent with the change tendency of the reference data, demonstrating the applicability of machine learning in GPP prediction. (2) LightGBM has the best overall performance, with small absolute error (mean absolute error less than 1.3), low degree of deviation (root mean square error less than 3.2), strong model reliability (relative percentage difference more than 5.9), and a high degree of fit with reference data (regression determination coefficient more than 0.97), and the prediction results are closest to the reference data (mean bias is only −0.034). (3) Enhanced vegetation index, normalized difference vegetation index, precipitation, land use/land cover, maximum air temperature, potential evapotranspiration, and evapotranspiration are significantly higher than other factors as determining factors, and the total contribution ratio to the prediction accuracy exceeds 95%. They are the main factors influencing GPP prediction. This study can provide a reference for the application of machine learning in GPP prediction and also support the research of large-scale GPP prediction.

The transition zone between the Mu Us Sandy Land and the Loess Plateau is considered an ecologically fragile area. However, significant changes in land use have occurred in the past few decades due to changes in land policies and the implementation of major national ecological projects. Despite this, there is still a lack of clear investigation into the impact of these changes on the landscape structure and ecological health of the area. This study utilizes high-resolution annual land use data from China, along with multi-index models and algorithms, to comprehensively analyze regional land use changes, landscape patterns, and the ecological environment’s quality. Through a comprehensive analysis of various factors, including changes in quantity, transformation in land types, spatial dynamics, landscape structure, and ecological quality, we aim to provide a better understanding of the complex interactions between land use and ecological systems in this area. The research results indicate that: (1) Since 2000, 9057.4 km2 of land in the study area has undergone changes. The grassland area has the largest increase, the forest area has the fastest growth rate, while cropland and barren land have decreased to varying degrees, and impervious surface has slightly expanded. (2) The movement trajectory of the center of gravity for different land types is closely related to human activities such as land development and utilization, as well as ecological restoration. Land changes have resulted in an escalation of landscape fragmentation, a reduction in landscape diversity, and a decline in the uniform distribution of different types. (3) Ecological land is the key to improving the ecological environment. The increase in ecological land area in the study area has led to an improvement in the quality of the ecological environment. The net contribution rate of land change to ecological improvement reaches 1.99%. The analysis methods and perspectives used in this study can be applied to other similar studies. The study’s findings enhance the understanding of how land and vegetation changes affect the ecological environment in this crucial area. They are of great significance in guiding the development and utilization of land resources and the implementation of ecological environment projects.