Comparison of spatial interpolation techniques for innovative air quality monitoring systems

Different integration methods were tested to integrate data from a dynamic road network (ROM) in which pollution measurement sensors were mounted over delivery vans. Two methods were purposely developed, the isoelliptical expansion -ISOE -method accounting for the wind convective transport of pollutants and the modified isoelliptical expansion -MISOE -method in which, furthermore, local specific deviation of the pollution are estimated from historical sequences of pollution levels. The results obtained by these methods were compared with the well-known inverse distance weighted -IDW -method, which is only based on the distance from the interpolation sources. The comparison of the errors between the estimated values and the available measures reveals that the MISOE model provides more accurate estimated values with a low associated error. The ISOE model is more complicate than the IDW but provides better estimations in windy days. The maps of the local adjusting coefficients estimated month by months are able to identify critical areas to address in local environmental policy decisions.