Spatiotemporal Patterns and Random Forest Prediction of Tropospheric NO₂ in Ankara, Türkiye

This dataset contains high-resolution tropospheric NO2 column density predictions for Ankara, Turkey, for the year 2025. The data was generated using a multi-sensor data fusion approach, integrating Sentinel-5P TROPOMI atmospheric data with Sentinel-2 multispectral surface reflectance. A Random Forest (RF) regression model was employed to downscale the coarse Sentinel-5P data (5.5 km) to a high-spatial resolution (100m) grid. The methodology relies solely on satellite-derived spectral signatures, providing a scalable and cost-effective solution for urban air quality monitoring without the need for ground-based stations or auxiliary meteorological data.本数据集涵盖土耳其安卡拉市2025年的高分辨率对流层二氧化氮（NO2）柱密度预测数据。该数据集通过多传感器数据融合方法生成，融合了Sentinel-5P TROPOMI大气数据与Sentinel-2多光谱地表反射率数据。研究采用随机森林（Random Forest, RF）回归模型，将分辨率较低的Sentinel-5P原始数据（5.5千米）降尺度至100米的高空间分辨率网格。该方法仅依托卫星反演的光谱特征，无需地面监测站点或辅助气象数据，可为城市空气质量监测提供可扩展且经济高效的解决方案。