Spatiotemporal Change of Crop Yield and Its Response to Planting Structural Shifts in Northeast China from 2001 to 2021

As a pivotal region for safeguarding China’s food security, Northeast China requires a quantitative evaluation of crop yield dynamics, planting structure shifts, and their interdependent mechanisms. Leveraging MODIS NPP data and remote sensing-derived crop classification data from 2001 to 2021, this study established a crop yield estimation model. By integrating the Theil–Sen median slope estimator and Mann–Kendall trend analysis, we systematically investigated the spatiotemporal characteristics of maize, rice, and soybean yields. Phased attribution analysis was further employed to quantify the effects of crop type conversions on total regional yield. The results revealed: (1) strong consistency between estimated yields and statistical yearbook data, with validation R2 values of 0.76 (maize), 0.69 (rice), and 0.81 (soybean), confirming high model accuracy; (2) significant yield growth areas that spatially coincided with the core black soil zone, underscoring the productivity-enhancing role of conservation tillage practices; (3) all three crops exhibited upward yield trends, with annual growth rates of 1.33% (maize), 1.20% (rice), and 1.68% (soybean). Spatially, high-yield maize areas were concentrated in the southeast, rice productivity peaked along river basins, and soybean yields displayed a distinct north-high-south-low gradient; (4) crop type transitions contributed to a net yield increase of 35.9177 million tons, dominated by soybean-to-maize conversions (50.41% contribution), while maize-to-soybean shifts led to a 2.61% yield reduction. This study offers actionable insights for optimizing planting structures and tailoring grain production strategies in Northeast China, while providing a methodological framework for crop yield estimation in analogous regions.