齐齐哈尔黑土地退化智能诊断方案

针对东北黑土退化监测难题，本方案以齐齐哈尔市为典型区，整合长时序多源地理时空数据与实地采样数据，构建了“天-空-地”一体化的土壤有机碳（SOC）智能预测体系。研究对比多种机器学习模型，确定随机森林（RF）为最佳预测算法，并识别出温度与水分是影响SOC的关键变量。时空分析显示，齐齐哈尔SOC呈“东北高、西南低”分布，虽2010年后有所恢复，但近40年整体仍呈退化趋势，尤其是东北部高背景值区域。驱动机制分析表明，初始SOC含量、土壤黏粒及地形起伏是主导因子，气候变化与农业管理措施亦起到协同作用。本方案精准厘清了区域黑土退化的时空特征与驱动根源，为黑土保护的数字化决策提供了科学支撑。Aiming at the challenges of black soil degradation monitoring in Northeast China, this study takes Qiqihar City as a typical study area, integrates long-term multi-source geospatial-temporal data and field sampling data, and establishes an integrated "satellite-aerial-ground" intelligent prediction system for Soil Organic Carbon (SOC). After comparing various machine learning models, Random Forest (RF) is selected as the optimal prediction algorithm, and temperature and moisture are identified as the key variables influencing SOC. Spatiotemporal analysis shows that the SOC in Qiqihar follows a distribution pattern of "high in the northeast and low in the southwest". Although SOC has shown a recovery trend since 2010, the overall degradation trend has persisted over the past 40 years, particularly in the high background value regions in the northeast. Driving mechanism analysis reveals that initial SOC content, soil clay content and topographic relief are the dominant controlling factors, while climate change and agricultural management practices also exert synergistic effects. This study accurately clarifies the spatiotemporal characteristics and underlying driving causes of regional black soil degradation, providing robust scientific support for digital decision-making in black soil conservation.