Abstract
In this work, a diagnostic scheme for lower atmospheric ducts was established based on the Weather Research and Forecasting (WRF) model. A 10-year simulation test was conducted for the China seas to investigate the spatio-temporal characteristics of the lower atmospheric ducts phenomenon. Compared with sounding data, the long-term simulations showed a high temporal correlation, and the root mean square error of the modified atmospheric refractivity remained between 4 M and 7 M. Significant regional differences in occurrence probability were detected from south to north. Surface ducts exhibited the highest occurrence probability, particularly in autumn and winter, with probabilities increasing at lower latitudes. In mid-latitude seas, strong ducts mostly occurred in spring and autumn, with single-layer ducts predominating. In lower latitude regions, the first duct layer existed year-round with weaker characteristics, while the second duct layer demonstrated stronger features.
Introduction
In a standard atmosphere, electromagnetic waves (such as radar or radio signals) bend slightly toward the Earth. However, specific weather conditions can create "atmospheric ducts"—layers where the refractive index drops so sharply that waves are trapped and guided along the Earth's curvature, much like light in a fiber-optic cable. While this can extend radar range far beyond the horizon, it also creates "blind spots" and signal interference. For maritime operations in the China seas, understanding the climatology of these ducts is vital. Previous research was often limited to short-term case studies or coarse reanalysis data that lacked the vertical resolution to capture these thin atmospheric layers. This study addresses this gap with a high-resolution, decade-long simulation.
Methodology
The researchers employed the Weather Research and Forecasting (WRF) model to conduct a 10-year simulation. To ensure the simulation remained constrained by reality, they implemented dynamic downscaling from ERA5 reanalysis data, four-dimensional grid nudging, and cyclic integration in 6-day segments.
A custom diagnostic scheme was developed to identify four specific duct types based on vertical profiles of modified atmospheric refractivity ($M$): simple surface, surface-based with a base layer, elevated, and composite ducts.
The model was validated against high-resolution radiosonde data from stations in Kowloon, Baoshan, and Dalian. Sea Surface Temperature (SST) data were also integrated to analyze the relationship between ocean conditions and duct formation.
Key Insights and Results
Regional and Seasonal Variability
The China seas exhibit a clear meridional (north-south) gradient in duct behavior. In the mid-latitude northern regions (Bohai and Yellow Seas), ducts are primarily seasonal and usually consist of a single, strong layer. Conversely, in the lower-latitude southern regions (South China Sea), complex two-layer "composite" ducts are much more common.
Seasonality also varies significantly with latitude. In mid-latitudes, duct occurrence peaks in spring and autumn. In low latitudes, peaks occur in early summer and early winter, lagging the northern peaks by approximately two months.
SST Correlation and Model Performance
A strong correlation (up to 0.95) was found between Sea Surface Temperature and the occurrence of simple surface ducts, as warm water enhances the humidity gradients necessary for wave trapping. The WRF model significantly outperformed the standard ERA5 reanalysis data, which often lacks the vertical resolution to detect sharp refractivity changes. The WRF diagnostic scheme achieved hit rates of 40% to 70% compared to balloon observations, whereas ERA5 often failed to detect ducts entirely.
Accuracy and Layer Inversion
The simulation demonstrated high reliability, with a Root Mean Square Error (RMSE) for modified refractivity between 4 M and 7 M and a temporal correlation exceeding 0.96. Interestingly, a "layer inversion" was observed: in the north, the first (lowest) layer is typically stronger, while in the south, the second layer often exhibits more pronounced strength and seasonal variation.
Discussion and Limitations
While the WRF model provides a robust climatological map, certain limitations remain. Numerical models tend to smooth atmospheric transitions, leading to a slight underestimation of duct frequency compared to instantaneous radiosonde data. Furthermore, the twice-daily radiosonde schedule may miss diurnal peaks in duct formation. Future work should explore coupled atmosphere-ocean models to better capture the dynamic feedback between the sea state and atmospheric conditions.
Summary updated