Coello Fernando, Decorte Thomas, Janssens Iris, Mortier Steven, Sardans Jordi, Peñuelas Josep, Verdonck Tim
Universitat Autònoma de Barcelona, 08193, Bellaterra, Spain.
CREAF - Centro de Investigación Ecológica y Aplicaciones Forestales, Barcelona, 08193, Spain.
Sci Data. 2025 Jan 9;12(1):40. doi: 10.1038/s41597-024-04215-x.
As global fertilizer application rates increase, high-quality datasets are paramount for comprehensive analyses to support informed decision-making and policy formulation in crucial areas such as food security or climate change. This study aims to fill existing data gaps by employing two machine learning models, eXtreme Gradient Boosting and HistGradientBoosting algorithms to produce precise country-level predictions of nitrogen (N), phosphorus pentoxide (PO), and potassium oxide (KO) application rates. Subsequently, we created a comprehensive dataset of 5-arcmin resolution maps depicting the application rates of each fertilizer for 13 major crop groups from 1961 to 2019. The predictions were validated by both comparing with existing databases and by assessing the drivers of fertilizer application rates using the model's SHapley Additive exPlanations. This extensive dataset is poised to be a valuable resource for assessing fertilization trends, identifying the socioeconomic, agricultural, and environmental drivers of fertilizer application rates, and serving as an input for various applications, including environmental modeling, causal analysis, fertilizer price predictions, and forecasting.
随着全球肥料施用量的增加，高质量数据集对于全面分析至关重要，以便在粮食安全或气候变化等关键领域支持明智的决策和政策制定。本研究旨在通过采用两种机器学习模型，即极端梯度提升算法和直方图梯度提升算法，填补现有数据空白，以生成精确的国家层面氮（N）、五氧化二磷（P₂O₅）和氧化钾（K₂O）施用量预测。随后，我们创建了一个分辨率为5弧分的综合数据集地图，描绘了1961年至2019年13个主要作物组每种肥料的施用量。通过与现有数据库进行比较以及使用模型的SHapley值加法解释来评估肥料施用量的驱动因素，对预测结果进行了验证。这个广泛的数据集有望成为评估施肥趋势、识别肥料施用量的社会经济、农业和环境驱动因素以及作为各种应用（包括环境建模、因果分析、肥料价格预测和预测）输入的宝贵资源。
