Wavelet analysis of the dynamic characteristics of saltwater intrusion - A case study in the Pearl River Estuary of China

The Modaomen Waterway (MW), a major outlet of the Pearl River Estuary, is a crucial source of water supply for Macao and Zhuhai City in Southern China. This waterway is frequently affected by saltwater intrusion from the South Sea, which has caused serious water shortage in recent years. For the planning and management of water supply at this waterway it is necessary to understand the characteristics of saltwater intrusion particularly in the low flow period of the year. However, as a result of multiple impacts from tide, river discharge, wind, topography, channel bathometry, and some other factors, the saltwater intrusion into this waterway is highly non-linear. In order to account for the non-linear characteristics, the wavelet method was used to analyze the period characteristics of tidal range, river discharge, and salinity and the impact of tidal range and river discharge on the salinity. Results show that the tidal range at Sanzao station and the salinity at Pinggang station both exhibit a stable 14.8-day period and the salinity in the MW is largely impacted by the tidal range during low stream flow periods. River flows from upstream of the waterway don't show any stable period characteristic, it, however, is critical in mitigating the salinity for water supply during saltwater intrusion. A minimum total flow of 2 500 m(3)/s from upstream at Sanshui and Makou stations has been practiced as a threshold for safe water supply. This study has found that the percent time when salinity is below the maximum permissible salinity of 250 mug for water supply increases as the river discharge from upstream increases and up to 50-70% of the time the salinity at water supply pumping station is below 250 mg/l when the upstream flow is greater than 2 500 m3/s. Planning and management of water supply operation at the MW is politically, economically and mostly importantly technically challenging, findings from this study can provide some guidance for operation and further research. For example, the phase lag of average salinity to tidal range was quantified to be 3.4-3.8 days. This phase lag time would suggest that the pumping operation to store freshwater in the backup storages needs to be at least 4-5 days ahead of the spring tide so as to avoid serious impact from saltwater intrusion. Safer water supply strategies can be developed by scheduling reservoir operations to maintain the threshold stream flow and the pumping. (C) 2014 Elsevier Ltd. All rights reserved.