Longterm nitrous acid evolution modulates ozone and secondary aerosol variations in global megacities (DeepSurf-HONO dataset)

Nitrous acid (HONO) is the dominant precursor of HOx radicals in urban air, fueling the formation of surface ozone (O3) and secondary fine particles (PM2.5). However, uncertainties in HONO chemistry and the sparsity of in-situ observations restrict understanding on its long-term trends and impacts on urban air quality evolutions. Here, we develop a deep neural network (DNN) well trained and tested on worldwide campaigns to reconstruct spatiotemporal distributions of global surface HONO. This DNN model is applied to generate gridded surface HONO over China and globally. We provide the daily averaged surface HONO mixing ratios across China (15km×15km, spanning 2013-2021), which is of relatively high confidence and finest spatial resolution. The predicted HONO mixing ratios are validated against multiple field campaigns, yielding a slope of 0.82 and an r2 of 0.89. This surface HONO dataset is referred to as DeepSurf-HONO (version 1.0). If you use this dataset and related materials for academic purpose, please cite this dataset and our work.