Meteorological data prediction for service environments of bridge using spatial interpolation method

The long-term adverse effect caused by environmental factors is one of the main causes for performance degradation of bridges. Accurately obtaining the environmental meteorological data on the bridge site can provide basis information for analyzing and assessing the negative effects of environments on bridges. To improve the efficiency and reduce the cost of meteorological data collection, this paper proposed a method to predict the meteorological data of bridge site based on the available shared date of weather station near the concerned bridge. Focusing on bridge temperature effects in this study, two major meteorological parameters, air temperature and wind speed, are investigated for data predication. Meanwhile, different geomorphologic conditions at different regions in China are discussed. Four interpolation methods, inverse distance weighting, Kriging interpolation, radial basis function, and minimum curvature method, are investigated. Cross-validation has been used to evaluate the performance of these four methods. The results show that Kriging method has the best predication on air temperature, and the inverse distance weighted method and Kriging method are both good for wind speed, however, Kriging shows better performance as a whole. Spatial interpolation method is feasible to predicate the meteorological parameters of bridge site, which provides a higher efficiency and less cost approach to obtaining the on-site meteorological data for the bridge temperature effects analysis and assessment.