Improvement for the multi-scale periodic characteristics revealing of precipitation signals and its impact assessment on soil hydrological process by combining HHT and CWT approaches

This study conducts a detailed analysis of multiscale periodic features involved in the annual and seasonal precipitation signals at the typical coastal reclamation region in China by selecting the suitable continuous wavelet transform (CWT) and innovatively combining the improved Hilbert–Huang transform (HHT), and further deduces the precipitation trend and its impacts on the future soil hydrological process. The Morlet wavelet transform is proved suitable in revealing the precipitation signals broad-scale periodicities, however, the critical mode mixing problem in CWT causes the poor significance in the fine-scale periodicities, which cannot well match the climate background. By combining the HHT approach, the fine-scale mode mixing drawback in CWT is effectively eliminated, and the the studied precipitation signals multi-scale periodicities are accurately revealed. Consequently, an overall decreasing trend of annual and seasonal precipitation in future years is demonstrated. Furthermore, by novelly using the cross wavelet transform (XWT) and wavelet transform coherence (WTC) approaches, the prominent correlations between the precipitation dynamics and soil and groundwater salinities dynamics, it is demonstrated that the precipitation increase can effectively leach the surface soil salt downwards into the deeper soil layers and groundwater with 5–7-day lag in the new cultivated tidal land. The revealed future decreasing trend of precipitation, especially in spring and summer, may aggravate the soil salinization at the coastal reclamation region, thus some reasonable salt leaching and evaporation suppression measures need to be taken to prevent the possible soil secondary salinization process.