大兴安岭多年冻土区森林地上生物量及碳氮磷储量数据集（2007-2010、2015-2023）

通过超高分辨率冠层高度模型（Canopy Height Model, CHM）和植被覆盖率（Fraction of Vegetation Cover, FVC）消除了ALOS-1/2 PALSAR-1/2合成孔径雷达（Synthetic Aperture Radar, SAR）数据中地形起伏与地面噪声的影响，并利用校正后的SAR影像估算了大兴安岭多年冻土区森林地上生物量（Aboveground Biomass, AGB）（2007-2010、2015-2023年）。在此基础上，基于时序植被指数构建了三角指数，该指数能很好地表征植被对养分的吸收与积累。随后，将三角指数与土壤变量、气候变量融合，结合实地测量的森林地上碳、氮、磷储量数据，采用高斯过程回归估算了研究区的森林地上碳、氮、磷储量地图（2007-2010、2015-2023年）。所有数据集均被重采样至250 m空间分辨率。

First, the effects of topographic relief and ground noise in ALOS-1/2 PALSAR-1/2 Synthetic Aperture Radar (SAR) data were removed using ultra-high-resolution Canopy Height Model (CHM) and Fraction of Vegetation Cover (FVC). Next, forest Aboveground Biomass (AGB) in the permafrost region of the Greater Khingan Mountains was estimated using the corrected SAR imagery for the periods 2007–2010 and 2015–2023. On this basis, a triangular index was developed from time-series vegetation indices, which can well characterize the nutrient uptake and accumulation of vegetation. Subsequently, the triangular index was fused with soil and climatic variables; combined with field-measured aboveground forest carbon, nitrogen, and phosphorus stock data, Gaussian Process Regression (GPR) was utilized to generate maps of aboveground forest carbon, nitrogen, and phosphorus stocks in the study area for the periods 2007–2010 and 2015–2023. All datasets were resampled to a spatial resolution of 250 m.