The Liquefaction potential of soils on the Cape Flats established by means of empirical correlation

Schoeman, Jurgens Johannes (2018-12)

Thesis (MEng)--Stellenbosch University, 2018.

Thesis

ENGLISH ABSTRACT: Soil Liquefaction is a problem associated mostly with earthquakes and occurs in areas with relatively shallow water tables. During an earthquake, the soil loses strength and stiffness due to the shaking. This results in settlement of silty sands, forcing groundwater out of its pores and up to the surface. The soil is put into a temporary state where the structure of the soil is distressed, resulting in soil particles losing contact and ultimately behaving like a liquid. The Cape Flats, situated in the Western Cape of South Africa, is an area known for flooding in low lying areas during the winter months, especially in informal settlements. The majority of the area is characterised by flat plains, and heavy winter rain cannot drain away due to the soil being saturated and draining canals usually blocked or absent. The soil in this area is also predominantly fine and medium grained silty sand, mostly cohesionless and uncemented, and potentially prone to liquefaction if ground shaking occurs. The Milnerton Fault Line runs through the Cape Flats and was the cause of the large destructive earthquake that struck Cape Town in 1809 and the years thereafter. Volcanoes of bursting sand and mud were reportedly seen all over the low-lying areas. Therefore, it is important to know whether liquefaction can occur in the Cape Flats during seismic events. Empirical correlation methods from proven studies were used to determine if the soils from the Cape Flats are susceptible to liquefaction during shear, induced by earthquakes. The testing methods used were the Piezocone Penetrometer and the Standard Penetration Test. Critical State concepts were also applied. Determining areas where the soils might be potentially dilative or contractive were established from literature. Soils that are potentially contractive are generally loose and potentially liquefiable. Soils that are dilative in nature are generally dense and should resist flow liquefaction. The use of the Dynamic Probe Super Heavy (DPSH) test was also included in the study due to limited access of SPT testing done on sands within the study area. The DPSH test also provides an equivalent SPT-N value and was essentially used in the same calculations. Results from the SPT and CPT tests indicated that the majority of soils in the study area are susceptible to liquefaction within the first 1.0 – 2.0 metres below ground. Soil samples obtained from within the study area as well as pre-graded soil samples were tested on a custom made vibrating table. Samples were graded to form uniformly graded, gap graded and well graded samples. The soils were saturated to 100% and vibrated at accelerations of 0.15g and 0.25g. These are Peak Ground Accelerations that can be expected for an earthquake in South Africa. The fact that denser soils have a larger or longer resistance to liquefaction was also proven in the laboratory on the vibrating table. Samples that were in a loose state liquefied much quicker compared to samples that were compacted and denser. Well graded and gap graded samples also resisted liquefaction for longer compared to that of uniformly graded samples.

AFRIKAANSE OPSOMMING: Grondvervloeïng is ‘n probleem wat meestal gepaardgaan met aardbewings en gebeur in areas met redelike vlak watertafels. Sanderige grond verloor sterkte tydens `n aardbewing. Dit veroorsaak dat sanderige grond versak en lei daartoe dat water uit die porieë forseer word. Die grond gaan dan deur `n tydelike fase waar dit soos `n vloeistof reageer. Die Kaapse Vlaktes, geleë in die Wes-Kaap van Suid Afrika, is `n area alombekend vir vloede in laagliggende dele tydens winter. Die grond in die area is ook meestal fyn en medium gegradeerde sand, los en moontlik vatbaar vir vervloeiing tydens `n aardbewing. Die Milnerton-foutlyn strek deur die Kaapse Vlaktes en was die oorsaak van een van Kaapstad se grootse aardbewings in 1809. Vulkane van sand was glo gesien, en dit is dus belangrik om te bevestig of grondvervloeiing hier kan plaasvind. Empiriese vergelykings van erkende studies was gebruik om te bepaal of grond op die Kaapse Vlaktes grondvervloeiing kan ondergaan tydens `n aardbewing. Die toets metodes was die van die Standaard Penetrasie Toets (SPT) en die die Keël Penetrasie Toets (CPT). Kritiese fase konsepte was ook toegepas. Dig en goed gekompakteerde grond kan uitsit en moontlik vervloeiing weerstaan. Los grond kan inkrimp of versak tydens `n aardbewing en moontlik vervloeiing ondergaan. Beide die twee fases was ook ondersoek. Dinamiese Penetrasie Super Swaar toetse (DPSH) was ook ingesluit by die SPT data omdat net n beperkte hoeveelheid van SPT data beskikbaar was. Die DPSH lewer `n soortgelyke waarde as die van die SPT, sodat altwee die toetse se data in die tesis gebruik was. Data dui daarop aan dat al die studie areas grondvervloeiing kan ondervind tydens `n aardbewing, veral in die eerste 2 meter onder grondvlak. Grondmonsters was getoets op n vibrasie tafel, spesiaal gebou vir die studie. Monsters was gegradeer om verskillende verhoudings van fyn tot grof te kry. Die monsters was dan tot 100% versadig, met water, en gevibreer teen n versnelling van 0.15 en 0.25g. Laasgenoemde is die verwagte Piek Grond Versnelling (PGA) wat verwag kan word tydens ‘n aardbewing in Kaapstad en Suid-Afrika. Digter grond het tydens toetse langer vervloeiing weerstaan. Monsters met a groter aantal fyn en growwer verhouding het ook langer vervloeiing weerstaan vergelykend met medium grote eenvormige gegradeerde sand.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/105044
This item appears in the following collections: