Expansive soils are rated as problematic soils because of their inherent potential to undergo large volume changes corresponding to changes in the moisture regime. They are considered potential natural hazard, which can cause extensive damage to structures if not adequately treated. When such soils absorb water they tend to expand and on the other hand, on drying they shrink and exhibit cracks. When such soils expand, they can exert enough uplift pressure on overlying structures, especially to lightly loaded structures such as single to double storeyed houses, pavements, floors and canal linings etc., causing severe damages in the form of uplifting and cracking of these structures. Reports on geotechnical investigations of various areas in Pakistan have revealed that considerable area of the country is covered with such soils and have caused severe damages to various types of structures. Identification of such soils at the investigation stage may help mitigating the detrimental effects of such soils. For preliminary identification of expansive soils by using empirical correlations between swelling parameters and index properties of the soils are very useful. Thus considering the problems associated with local expansive soils, there was a dire need to collect enough data of local soils and to determine swelling characteristics of such soils for the rating of their swell potential in order to develop specific mathematical models (based on index properties). Also, there is a common trend to develop zonation maps for the identification of such soils in different countries to enable better planning of infrastructure. Likewise, zonation maps for the local expansive soils were essentially required to facilitate the geotechnical engineers dealing with such soils. This research is mainly focused on the identification of local expansive soils and their characterization with respect to their swelling characteristics and mapping of such soils in different parts of the Punjab province of Pakistan. After collecting the data of basic classification tests from different agencies/organizations involved in geotechnical exploration in Pakistan, it has been established that the problem of swelling is mainly present in five districts of Punjab namely Dera Ghazi Khan, viii Chakwal, Sialkot, Gujranwala and Narowal. For detailed investigation, undisturbed and disturbed soil samples were retrieved from twenty eight different sites present in the potential swelling districts of Punjab. For geotechnical characterization of these soils, basic classification tests, X-ray diffraction analysis, expansion index tests, swell tests and compaction tests were performed.Expansion index (EI) test is considered to be an imperative classification test for the identification of the expansive soils. However, this test is quite time consuming and requires an elaborate testing arrangement. In this research, apparatus for performing the EI test was developed and based on extensive testing, an effort was made to correlate the EI parameter with plasticity index of the soil which can be determined through a simple Atterberg limit test. Through a series of EI tests on all samples including bentonite mixed samples (to have large variation in PI values), a linear relationship has been developed between plasticity index and expansion index parameter as given below. EI = 3.52 x PI----------------------------------------------------------------A-1 In order to explore the effects of various factors, which may affect the swelling properties of expansive soils such as initial moisture content, initial dry unit weight, plasticity index, surcharge pressure and mixing with non-swelling soil, a comprehensive laboratory testing programme was undertaken. Following are the salient findings of the laboratory investigations.· A series of swell tests were performed by varying remoulding initial moisture content and dry unit weight of the samples. It was revealed that both the swell potential (percent swell) and swell pressure are affected by changing initial moisture content of the samples remoulded on a specific initial dry unit weight. Swell potential decreased from 8.4% to 0.07% and swell pressure from 167 kPa to 5 kPa, when initial moisture content was increased from 0% to 20%. It implies that swelling problems can be mitigated by increasing moisture content. Moreover, swell potential increased from 0.07% to 8.4% and swell pressure from 5 kPa to 167 kPa when initial dry unit weight was increased from 14 kN/m3 to 18 kN/m3 ix showing that both swell potential and swell pressure values are more for the denser samples and less for the samples having low density.· Bentonite was added into the soils in increments from 0% to 100% (to have large variation in PI values) to investigate the effect of plasticity index. It has been observed that swell potential increased from 2.5% to 42.4% and swell pressure increased from 65.5 kPa to 850.6 kPa with the increase in plasticity index from 10% to 303%. · Swell tests on soil samples remoulded with different compaction energies revealed that when compaction energy was increased from 202 kJ/m3 to 2556 kJ/m3, swell potential increased from 0.4% to 8.8% and swell pressure increased from 10 kPa to 278 kPa. Moreover it has been observed that at a specific compaction energy level, for compaction moisture content dry of optimum, the swell potential tended to be significant and as the compaction moisture was more than OMC, the swell potential decreased. This finding implies that if the field compaction is performed slightly on wet of optimum, the swelling potential of the compacted soil can be effectively mitigated in field practice. · Series of swell tests with varying surcharge pressure on soil samples during saturation showed that when surcharge pressure was increased from 6.9 kPa to 27.6 kPa, both swell potential and swell pressure decreased from 9.4% to 0.2% and from 162 kPa to 8 kPa, respectively. This finding is very useful in controlling swelling of soils in the field. By increasing the vertical surcharge pressure on such soils in the form of foundation bearing pressure (within allowable bearing capacity of the soil), problems associated with swelling can be mitigated. · In order to investigate the effect of blending with non-swelling soil in swelling clays, a series of swell tests was performed by mixing locally available sand into the expansive soil samples. It was revealed that swell potential decreased from 9.4% to 0.4% and swell pressure decreased from 162 kPa to 7 kPa when the sand was added from 0% to 25%. Addition of about 20% locally available sand transformed high swelling soil into low x swelling soil. This finding may also be very useful in controlling the problems associated with swelling of expansive soils by mixing locally available sandy soil in proportion of 20~25%. By using the results of all swell tests performed on the samples collected from potential swelling districts of Punjab including bentonite mixed samples, a linear relationship between swell pressure and swell potential has been developed as given below. Swell Pressure (kPa) = 20.4 x Swell Potential ------------------------ A-2 Through a comprehensive data analysis of the swell tests investigating the various factors, it was revealed that the swelling parameters (swell potential and swell pressure) of the expansive soils are mainly affected by varying the plasticity index, field moisture content and field dry unit weight of the soils. In this connection, multiple linear regression analysis was performed on the investigated swell data and correlations between swelling parameters and the three affecting parameters, plasticity index, initial moisture content and initial dry unit weight, have been developed as given below. Sp =0.111 PI + 1.12 γd –0.208 w – 14.5 -------------------------------A-3 Ps = 2.22 PI + 28.2 γd – 1.87 w – 417 -------------------------------- A-4 whereSp = Swell potential (%), Ps = Swelling pressure (kPa), PI = Plasticity index (%) γd =Initial dry unit weight (kN/m3) and w =Initial moisture content (%) The above mentioned correlations given by equations A-3 and A-4 were duly validated by using the data of swell tests performed on all soil samples collected from various sites. It is to mention here that this data was not included in the development of the correlations. Comparisons of experimentally measured and predicted swelling parameters by equations A-3 and A-4 show that the proposed correlations can predict the swelling parameters of the soils within an accuracy of +10% which may be considered a reasonable prediction for preliminary purposes. Finally, zonation maps have been developed for the identification of expansive soils in Punjab province area based on plasticity index, liquid limit and depth of expansive soil using GIS (Geographic Information System) software with the help global coordinates of the sites investigated. Moreover, using the proposed models by equations A-3 and A-4, zonation maps for predicting swell potential and swell pressure have also been developed using the modified Proctor compaction test parameters (γdmax & OMC) and the plasticity index data of the investigated areas. As an outcome of this research, it can be said that the proposed mathematical models between swelling parameters and the basic properties of the soils will be a source of quick prediction of swelling characteristics of the expansive soils present in the Punjab province of Pakistan. Moreover zonation maps developed using GIS software based on Atterberg limits data and the proposed mathematical models (using the PI values and the field compaction parameters) can be used effectively to predict the swell potential and swell pressure at various locations. The findings of this research may provide a tool for quick identification and prediction of swelling parameters of expansive soils present in the Punjab region which will be quite useful during preliminary design stages of any civil engineering project." xml:lang="en_US