A novel approach to estimate surface refractivity with simple meteorological data

This study analyzes surface refractivity patterns and proposes six empirical models for its estimation in Houston, Texas, a subtropical climate region of the United States, using 18 years (2006–2023) of meteorological data. Monthly and yearly variations revealed a seasonal pattern peaking in summer (372.4 N units in July) and reaching its lowest in winter (327.8 N units in December). An analysis of the contributions of the dry and wet components shows the dry component dominating in the winter (61–63 %) and the wet component prevailing in the summer (56–60 %). The proposed models were developed using air temperature, atmospheric pressure, and relative humidity as input variables. Models’ performances were evaluated using statistical metrics and the Akaike Information Criterion (AIC). The models' residuals closely followed a normal distribution, indicating robust and reliable predictive capabilities that enhance understanding of surface refractivity. Model 3, a two-variable linear model incorporating air temperature and atmospheric pressure ($N_i=-33.7689+1.3234\left(T\right)-0.0069\left(P\right)$, was identified as the best performing model with the lowest AIC (6.829), AIC_C (12.543), and ΔAIC_C (0.000) values and deviations between −1.80 % and 2.38 % from the measured values. Model 1, a simpler single-variable linear model using only air temperature ($N_i=-41.2088+1.3250\left(T\right)$), also showed strong performance, with deviations ranging from 0.12 % to 6.36 %. Comparative analysis indicated that locally developed models significantly outperformed the standard refractivity equation, highlighting the importance of location-specific empirical models for accurate surface refractivity estimation. Importantly, these models rely solely on readily accessible air temperature and atmospheric pressure measurements, eliminating the need for solar radiation data and enabling their easier application across diverse regions. These findings have implications for modeling radio wave propagation, radar systems, telecommunications planning, and atmospheric studies in similar climate regions.