Resolving major discrepancies between predicted and monitored settlements of a highway embankment on soft soil

This case study discusses settlement predictions and how they compare to monitored settlements at an embankment constructed over soft soil forming part of the Woolgoolga to Ballina Pacific Highway upgrade on the north coast of New South Wales. The embankment height ranged from 3.4 to 7.7 metres with wick drains installed across most its 410-metre length. Settlement predictions ranged from 125 to 300 millimetres with a predicted construction and waiting time of 1.5 years. Settlement plates, inclinometers, magnetic extensometers, hydrostatic profile gauges and vibrating wire piezometers were installed to monitor settlements, pore pressures and embankment stability. It was found that after 3 months of construction and 9 months of combined construction and consolidation monitoring the soft soil had reached 95 to 100% consolidation, however the monitored settlement had exceeded predictions by up to nearly 300% in some areas of the embankment while in other areas the monitored settlement was less than 20% of prediction. Site investigation data was scrutinised and back-analysis was performed to match the monitoring data in a one-dimensional consolidation model. Back analysed parameters were compared to design parameters and possible reasons for the discrepancies in settlements and consolidation rates were discussed.