Divergent diurnal dynamics and regional applicability of GEMS-based ozone sensitivity indicators (FNR vs. GNR) across...

Satellite-based diagnosis of ozone (O₃) sensitivity often relies on a single daytime observation and fixed precursor-ratio thresholds, potentially obscuring intraday and regional variability. Using hourly GEMS observations from March to October 2023, we evaluated the formaldehyde-to-nitrogen-dioxide ratio (FNR) and glyoxal-to-nitrogen-dioxide ratio (GNR) across Central and Eastern China. Hour-specific thresholds were derived from LOESS relationships between precursor ratios and the probability of high-O₃ occurrence. Central estimates of both thresholds rose from morning to midday. The FNR interval shifted from [1.37, 2.07] at 10:00 to [2.39, 3.14] at 13:00, while GNR shifted from [0.06, 0.09] to [0.10, 0.17]. After 13:00, GNR decreased, whereas FNR remained relatively high. VOC-limited conditions clustered in urban and industrial areas, while NO_x-limited conditions were more widespread and changed markedly during daytime. Source related analyses indicated that FNR was more representative in Beijing-Tianjin-Hebei and Sichuan-Chongqing, whereas GNR was more applicable in the Yangtze River Delta and Pearl River Delta, where anthropogenic emissions, particularly aromatic VOCs emissions, were more pronounced. These results show that ozone sensitivity diagnoses require hour-specific thresholds and region-specific indicator selection for targeted NO_x-VOC control.