Guangdong Provincial Key Laboratory of Atmospheric Environment and Pollution Control, School of Environment and Energy, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou, 510006, China.
Guangdong Provincial Key Laboratory of Atmospheric Environment and Pollution Control, School of Environment and Energy, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou, 510006, China.
Environ Pollut. 2023 Oct 15;335:122291. doi: 10.1016/j.envpol.2023.122291. Epub 2023 Jul 30.
Ambient ozone (O) predictions can be very challenging mainly due to the highly nonlinear photochemistry among its precursors, and meteorological conditions and regional transport can further complicate the O formation processes. The emission-based chemical transport models (CTM) are broadly used to predict O formation, but they may deviate from observations due to input uncertainties such as emissions and meteorological data, in addition to the treatment of O nonlinear chemistry. In this study, an innovative recurrent spatiotemporal deep-learning (RSDL) method with model-monitor coupled convolutional recurrent neural networks (ConvRNN) has been developed to improve O predictions of CTM. The RSDL method was first used to build the ConvRNN within a 24-h scale to characterize the spatiotemporal relationships between the monitored O data and CTM simulations, and then incorporated the recurrent pattern to achieve 72-h multi-site forecasts based on a pilot study over the Pearl River Delta (PRD) region of China. The results showed that the RSDL method predicted O with high accuracy over this case study, with an increase of 27.54% in the correlation coefficient (R) average for all sites as well as an increase in R of 0.14-0.21 for all cities compared to CTM. Moreover, the regional distribution of CTM was further improved by the RSDL predictions with the data fusion technique, which greatly reduced the underpredictions of O concentrations, particularly in high O-level areas (concentrations >160 μg/m), with a 33.55% reduction in the mean absolute error (MAE).
环境臭氧 (O) 的预测可能极具挑战性，主要是因为其前体之间存在高度非线性光化学反应，而气象条件和区域传输会进一步使 O 的形成过程复杂化。基于排放的化学传输模型 (CTM) 被广泛用于预测 O 的形成，但由于排放和气象数据等输入不确定性，以及 O 非线性化学的处理方式，这些模型可能与观测结果存在偏差。在这项研究中，开发了一种具有模型监测耦合卷积递归神经网络 (ConvRNN) 的创新循环时空深度学习 (RSDL) 方法，以改进 CTM 的 O 预测。首先，RSDL 方法用于在 24 小时的时间尺度内构建 ConvRNN，以描述监测的 O 数据与 CTM 模拟之间的时空关系，然后结合递归模式，根据中国珠江三角洲 (PRD) 地区的试点研究实现 72 小时多站点预测。结果表明，该方法在该案例研究中对 O 进行了高精度预测，所有站点的相关系数 (R) 平均值增加了 27.54%，与 CTM 相比，所有城市的 R 增加了 0.14-0.21。此外，通过数据融合技术，RSDL 预测进一步改善了 CTM 的区域分布，大大减少了 O 浓度的低估，特别是在高 O 水平地区（浓度>160μg/m），O 浓度的平均绝对误差 (MAE) 降低了 33.55%。
