Assessing local and global displacements from tunnelling and water extraaction in jakarta

Abstract

This thesis investigates the significant ground settlements that have occurred in Jakarta due to excessive groundwater pumping to a much lesser degree tunnelling. Particular attention is focussed on the effects of the groundwater extraction, which have been identified from monitoring groundwater levels and settlements. The settlements from tunnelling are 2 - 3 orders of magnitude smaller. Several locations in the northern part of the city and not too far from the vicinity of the Jakarta Mass Rapid Transit (JMRT) tunnel alignment were selected for study. The overall aim is to gain a better understanding of the mechanisms, magnitudes and rates of settlements caused by the lowering of the water table which has led to extensive consolidation, and also to predict how much more might occur pumping continues or stops. Two site investigations (SI) consisting of geotechnical boreholes and piezocone penetration tests (CPTU) were conducted as part of this research. The findings from there were integrated within the overall geological setting determined from desk studies. Comprehensive laboratory test campaigns on samples from the SI were carried out in Imperial College with major components involving an extensive series of oedometer tests and the development of Soil-Water-Retention-Curves (SWRC) as well as physico-chemical tests. Both intact and reconstituted specimens were prepared and tested. Different salt concentrations were used in order to investigate the effect of saline water on the compressibility characteristics and unsaturated response of the soils. After detecting salt diffusion during the oedometer testing of intact samples of Jakarta clay, an investigation into the effects of salts was run. Tests on reconstituted samples of kaolin and Montmorillonite prepared with fresh and saline water showed that the effect on kaolin was minor. However, the saline montmorillonite samples exhibited markedly lower ratios and compressibility but higher permeability and hence rates of consolidation. It seems that cations from the salt replace the need for such a thick diffusion double layer, leading to these changes in consolidation characteristics. This effect reduced with increasing stress level. Test were performed on intact samples of Jakarta clay surrounded by a saline water bath and also on reconstituted samples of the same clay prepared with fresh or saline water. In these case the results were not so marked as the effect depends on the relative proportions of kaolin and montmorillonite in each sample, both of which were identified from diffractograms for specific samples. The effect was also obscured in some cases by the effect of structure evident in the intact samples. Secondary compression effects have not be investigated. Clay samples from the Quaternary sediments are of extremely high plasticity and have significant organic content. The SWRC results confirm the presence of salts within the soils in significant concentrations near the shoreline reducing proportionally with distance from it. Physico-chemical tests confirm the presence of salts and this trend of higher concentrations near the shoreline in northern Jakarta. Primary drying and wetting scanning curves exhibit significant hysteresis. There was no indication of the shrinkage limit being reached up to the suction limit of 30 MPa suggesting that this would not inhibit the development of settlements from pore water pressure reduction without changes in total stress. Results from CPTU and oedometer tests relating to the Tongkol SI location in northern Jakarta were used to back-analyse settlements to date, making comparisons with settlement monitoring that started in 1990, and to make future predictions. The CPTU data allowed the complex stratigraphy to be broadly grouped into aquifers and aquicludes and provided supplementary coefficient of consolidation, c h , values. Appropriate consolidation characteristics were then obtained from the oedometer tests on intact and reconstituted samples, relating to stress changes deduced from monitoring well data. A range settlements of values between 25 and 32 cm was determined, depending on assumptions made. It is estimated that for this location if groundwater extraction continues, the incremental settlement (from 1990) will double to 80 cm by 2100. However, it is important to recognise that overall settlement at this location since 1950 are about 2.0 m and are greater in some areas closer to the shoreline. Such magnitudes of settlements are of major concern regarding flooding of the city from the sea and flood protection measures are necessary.