A physically-modified spectral model for prediction of sand wave migration

Abstract

Sand waves are rhythmic bedforms existing in shallow seas and typically induced by the interaction of currents and topography. In this study, a physically-modified spectral model for simulating small-scale sand wave migration is proposed by integrating non-linear spectral analysis, typhoon wind field and sediment transport theory. We introduce a power-average velocity to modify the spectral analysis, which gives explicit physical meanings to the morphological method and improves its stability and reliability. Then, the effects of typical and extreme weather are decoupled and the short-term migration caused by typhoons is removed from the total migration. Results show that a single typhoon will cause large sand wave migration, but the migration caused by multiple typhoons in a long period of time will cancel each other out in the measured data. Moreover, the bilateral reverse migration trend of one-group sand waves in a small scale (∼3 km) is captured, which is consistent with field data and previous researches. Compared with existing researches, the proposed model is able to identify the influences of extreme conditions on sand wave migration and provide a feasible and high efficiency way to predict sand wave migration when the available hydrodynamic data is limited.