Wei et al. (2022) Daily and Annual PM2.5, O3, and NO2 Concentrations at ZIP Codes for the Contiguous United States, v1 (2000 – 2016)

The Daily and Annual PM2.5, O3, and NO2 Concentrations at ZIP Codes for the Contiguous U.S., 2000-2016, v1.0 dataset contains predictions of air pollution concentrations at the ZIP Code-level for 2000-2016. An ensemble framework consisting of three machine-learning models (Random Forest, Gradient Boosting, and Neural Network) was implemented to estimate the daily concentrations of fine particulate matter (PM2.5), ozone (O3), and nitrogen dioxide (NO2) in grid cells at 1-km resolution. Prediction variables included air monitoring data, satellite aerosol optical depth, meteorological conditions, chemical transport model simulations, and land-use variables. Annual predictions represent the average daily predictions in each grid cell for each year.

美国本土2000-2016年邮政编码级逐日及年际PM2.5、O3、NO2浓度数据集（v1.0）包含2000-2016年邮政编码层级的空气污染浓度预测数据。本数据集采用由随机森林(Random Forest)、梯度提升树(Gradient Boosting)与神经网络(Neural Network)组成的集成学习框架，以1公里分辨率对网格单元内的细颗粒物（PM2.5）、臭氧（O3）和二氧化氮（NO2）的日浓度进行估算。预测变量涵盖空气监测数据、卫星气溶胶光学厚度、气象条件、化学传输模型模拟结果以及土地利用变量。年际预测值为各年份下每个网格单元的逐日预测平均值。