青藏高原大气颗粒物浓度与酸沉降数据集

青藏高原环境问题数据集包括1990-2017年间青藏高原PM2.5浓度以及酸沉降量（硝态氮沉降、铵态氮沉降与硫沉降）的时空变化模拟数据（1km×1km），该数据主要是基于卷积神经网络（CNN）模型，结合长时间卫星遥感反演数据，结合气象传送模型以及环境监测站实测数据进行模拟生成。该数据集的CNN模型训练目标数据来自四个主要监测网络，包括东亚酸沉降监测网络（EANEkg）、欧洲监测和评估计划（EMEP）、国家大气沉降计划（NADP）、以及加拿大空气与降水监测网络（CAPMoN）。此外，MERRA-2的气象变量与全球大气研究排放数据库（EDGAR）作为CNN的训练数据输入。

The Qinghai-Tibet Plateau Environmental Issues Dataset contains spatially and temporally resolved simulation data (1km × 1km) of PM2.5 concentrations and acid deposition amounts (including nitrate nitrogen deposition, ammonium nitrogen deposition and sulfur deposition) over the Qinghai-Tibet Plateau from 1990 to 2017. This dataset was primarily generated via simulations using a Convolutional Neural Network (CNN) model, combined with long-term satellite remote sensing inversion data, meteorological transport models, and in-situ measured data from environmental monitoring stations. The target training data for the CNN model of this dataset comes from four major monitoring networks, namely the East Asian Acid Deposition Monitoring Network (EANEkg), the European Monitoring and Evaluation Programme (EMEP), the National Atmospheric Deposition Program (NADP), and the Canadian Air and Precipitation Monitoring Network (CAPMoN). Additionally, meteorological variables from MERRA-2 and the Emissions Database for Global Atmospheric Research (EDGAR) were used as training data inputs for the CNN.

搜集汇总

背景与挑战

背景概述

该数据集包含1990-2017年间青藏高原PM2.5浓度及酸沉降量的时空变化模拟数据，空间分辨率为1km×1km，数据大小为3.94 GB。数据基于卷积神经网络（CNN）模型，结合卫星遥感、气象传送模型及环境监测站实测数据生成，来源包括多个国际监测网络和数据库。