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.