Analysis of subsurface strata of Colombo and Gampaha districts of Sri Lanka, based on geotechnical investigation data

Abstract

The subsurface exploration for geotechnical engineering applications in Sri Lanka is mainly based on borehole investigations. Several leading geotechnical engineering companies in the country have already done more than a couple of thousands of subsurface investigations for small- and large-scale civil engineering projects. Even though project-wise subsurface information is available, integration of available subsurface data in proximity, and development of subsurface three-dimensional (3D) models can hardly be seen in the country. This has been a significant disadvantage in the cost and planning of large-scale new investigation projects. Due to lack of information, most of the projects start even without knowing the tentative bedrock level of the area. Hence, in investigation cost estimations, pricing for most of the items is recorded as “rate-only”, hence the total cost could immensely be higher than the available budget with the client. However, if 3D modelling of subsurface strata can be done area-wise, based on already available borehole data, such uncertainties could be minimized. In this research, more than a thousand subsurface investigation reports were reviewed; data were recorded and analysed targeting to interpret the subsurface of the western province of Sri Lanka, and to develop a 3D subsurface model for the same. However, scattered data had to be excluded in data analysis, and eventually, the study was confined to Colombo and Gampaha districts, in which more than sufficient data could be found. By using interpolation methods, surface strata were interpreted in between borehole locations by matching similar geological features. In addition, artificial neural networks were used to forecast borehole data in exceptional cases for a few locations. This helped to improve the spatial coverage and accuracy of the 3D model developed by means of “Surfer” software. The 3D model developed for the study area well demonstrates the subsurface strata and facilitates taking of cross sections in any direction within minutes. Hence, the findings of this research will enhance the outcome of general geotechnical investigation practice in Sri Lanka. This will also be immensely beneficial in planning and budgeting of future large-scale geotechnical investigation projects, more accurately than in the past, saving energy and time.