A multilayer approach is set up for local gravity field modeling based on the idea of multi-resolution representation merging heterogeneous gravity data. Different layers of Poisson wavelets' grids are formed to recover the signals at various levels, where the shallow and deep layers mainly capture the short- and long-wavelength signals, respectively. The depths of these layers beneath the topography are linked to the locations that the anomaly sources locate, estimated by the wavelet decomposition and power spectrum analysis. For testing the performance of this approach, a gravimetric quasi-geoid over the North Sea in Europe called QGNSea V1.0 is computed and compared with other existing models. The results show that the multilayer approach outperforms the traditionally used single-layer one in high-frequency bands, and the former fit the gravity data better, especially in regions with a tendency toward topographical variation. The evaluation with GPS/leveling data show the accuracies of QGNSea V1.0 modeled from the multilayer approach are improved by 0.3 cm, 0.6 cm and 0.8 cm in the Netherlands, Belgium and parts of Germany, respectively, compared to the original solution computed from the single-layer approach. Further validation with existing models show QGNSea V1.0 has the best quality, which may be beneficial for studying the ocean circulation between the North Sea and neighbouring waters.