See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/365847293

Correlation between effective cohesion and plasticity index of clay

Article in Geologica Balcanica · November 2022

DOI: 10.52321/GeolBalc.51.3.45

CITATIONS READS

4 2,241

2 authors:

Boriana Tchakalova Plamen Ivanov

Bulgarian Academy of Sciences 46 PUBLICATIONS 211 CITATIONS
21 PUBLICATIONS 43 CITATIONS
SEE PROFILE
SEE PROFILE

All content following this page was uploaded by Plamen Ivanov on 02 January 2023.

The user has requested enhancement of the downloaded file.

 GEOLOGICA BALCANICA 51 (3), Sofia, December 2022, pp. 45–49.

Correlation between effective cohesion and plasticity index of clay

Boriana Tchakalova, Plamen Ivanov

Geological Institute, Bulgarian Academy of Sciences, Acad. G. Bonchev Str., Bl. 24, 1113 Sofia, Bulgaria;
e-mails: [email protected]; [email protected]
(Received: 01 November 2022; accepted in revised form: 28 November 2022)

Abstract. Correlations of engineering properties are a useful tool in geotechnical engineering practice. This
paper aims to provide a correlation between the effective cohesion and plasticity index for natural, undis-
turbed clay soils from the Kozloduy area (NW Bulgaria), based on the results from laboratory tests. It has
been demonstrated that there is a strong correlation between the plasticity index and the effective cohesion.
The derived regression equation can be used to estimate the effective cohesion as first approximation in pre-
liminary design of engineering projects of Pliocene and Quaternary clays encountered in northwest Bulgaria.

Tchakalova, B., Ivanov, P. 2022. Correlation between effective cohesion and plasticity index of
clay. Geologica Balcanica 51 (3), 45–49.
Keywords: effective cohesion, plasticity index, correlation.

INTRODUCTION limits represent the plasticity characteristics of soils

and are essential in the classification of fine-grained
The effective cohesion (cʹ) of soils is one of the most soil. They are also used to calculate the plasticity
important soil parameters that is evaluated in slope index (IP), which could be correlated with many soil
stability and suitability for building foundations. Ef- properties.
fective cohesion is considered as a part of the shear Correlations between the index properties pa-
strength that can be mobilized due to forces arising rameters and the strength and deformation proper-
at particle level and is independent from the effec- ties of cohesive soils are widely employed in geo-
tive stress (Lambe, 1960). As per Yong and War- technical engineering practices as first approxima-
kentin (1966), cʹ of soils is extremely dependent tion of the soil characteristics in the preliminary de-
on the interaction characteristics of the clay–water sign of geotechnical structures, and later as a mean
system. Thus, cʹ is affected by the Atterberg limits to validate the results of laboratory tests (Sørensen
of soils. and Okkels, 2013).
Essentially, the Atterberg limits are controlled Beneficial empirical equations associated with
by soil mineralogy, pore structure, and particle size various soil properties correlated with Atterberg
distribution and reflect the ability of fine-grained limits have been provided by many researchers,
soil to resist external shear loading (Seed et al., such as Fener et al. (2005), Dolinar and Trauner
1966). Atterberg (1911) derives seven limits that (2007), Mehta and Sachan (2017), Spagnoli and
describe changes in the behavior of cohesive soils Shimobe (2020), and others.
at varying water content. Nowadays, in practice, Obtaining of the Atterberg limits in a laboratory
only three are in use: liquid limit (wL), plastic limit setting is relatively simple to perform, quick, and
(wP) and shrinkage limit (wS). The liquid and plastic inexpensive compared to tests for the determination

© БАН, Геологически институт „Акад. Страшимир Димитров“, 2021 www.geologica-balcanica.eu

https://doi.org/10.52321/GeolBalc.51.3.45
Boriana Tchakalova, Plamen Ivanov

of soil strength parameters such as cʹ. Also, obtain- Table 1

ing cʹ requires undisturbed test samples, the deriva- Regression models
tion of which is time-consuming and costly. In that
Regression Model Equation
connection, this study was carried out to predict the
drain cʹ of fine-grained soils from their IP, which Simple linear y̌ = b0 + b1x
will be useful for the preliminary analysis of an en- Logarithmic y̌ = b0 + [b1 * ln(x)]
gineering project. The Atterberg limits and cohe-
sion of 33 soil samples were determined. Based on Quadratic y̌ = b0 + b1x + b2x2
statistical methods, an empirical equation, for the Exponential y̌ = b0 * e (b1 * x)
prediction of cʹ based on IP, has been obtained for
practical use.

RESULTS AND DISCUSSION

MATERIALS AND METHODS
The tested soils cover a wide range of classifica-
The soil tests were performed on 33 undisturbed tions. Based on ESCS, samples are classified as
soil samples of Pliocene and Quaternary clays from follows: low plasticity clay (ClL) – 13 specimens;
the Kozloduy area (NW Bulgaria). The soil sam- medium plasticity clay (ClM) – seven specimens;
ples were collected from borehole cores at differ- high plasticity clay (ClH) – six specimens; and
ent depths, mainly from 10 m to 25 m below the very high plasticity clay (ClV) – seven specimens
surface. Grain size distribution, particle density, (Fig. 1). Summary of the obtained results for IP and
plasticity limits, and cohesion were determined ac- cʹ is presented in Table 2.
cording to BDS EN ISO/ TS 17892. Classification Correlation and regression analysis were con-
of soil samples was performed according to the Eu- ducted with IP as an independent variable, and cʹ as
ropean Soil Classification System (ESCS). a dependent variable. The regression models of cʹ
The Atterberg limits were defined according to and IP are presented in a graphic form in Fig. 2. The
BDS EN ISO/ TS 17892–12:2018. The tests were first procedure was estimation of the strength of the
performed at a room temperature of 20 °C by the relationships between variables by Pearson’s cor-
same operator in order to reduce the possibility of relation test and the F-test related to it. The F-test
human error. The liquid limit was obtained as rec- was targeted to the significance of entire regression
ommended in the clause 5.4 of BDS EN ISO/ TS models and to detect if the independent variable
17892-12:2018 with Casagrande apparatus with a could be used to predict the models.
hard base percussion cup.
The effective cohesion was derived at saturated
consolidated-drained conditions from direct shear
test. The shear resistance envelope was obtained
by three shear tests, each using a different effec-
tive normal stress (100 kPa, 200 kPa, and 300 kPa),
performed on specimens from the same soil sample.
The least squares method was used to obtain the
corresponding cʹ values.
A variety of statistics was applied to explore the
relationship between cʹ (response variable) and IP
(predictor variable). The curve fitting procedure
was used for estimation of the regression model. In
regression analysis, curve fitting is the process of
specifying the model that provides the best fit to the
specific curve set of data.
Regression analyses were performed and single-
factor models were obtained, using the equations as
described in Table 1, where x is a predictor variable,
denotes the predicted value of the response variable
y̌ for a given x; b0, b1, and b2 – coefficients of the Fig. 1. Tested samples shown on plasticity diagram for the
independent variable; and e is Euler’s constant. ESCS classification.

46
 Correlation between effective cohesion and plasticity index of clay

Table 2 Table 3
Summary of obtained results for IP and cʹ Pearson correlation and F-test results

Number of Range of Range of cʹ, Regression F-test

Soil type R R2
specimens IP, % kPa model F p-value
ClL 13 11.8–14.6 14.8–23.5 Exponential 0.953 0.908 304.745 0.000
Linear 0.982 0.964 830.572 0.000
ClM 7 13.8–30.5 19.3–33.4
Logarithmic 0.962 0.925 381.482 0.000
ClH 6 32.0–46.3 36.4–53.0
Quadratic 0.983 0.966 420.450 0.000
ClV 7 44.2–78.2 47.7–76.0 Power 0.968 0.938 466.874 0.000

rectly related to adding predictor variable and could

not be accepted as the best-fitted model.
The results of the F-test show that the regression
modes are statistically significant (p value <0.05;
Table 3). This is sufficient evidence to conclude
that IP will provide a better fit than a model with
zero independent variable, and could be used to pre-
dict the models.
The next step of regression analysis is perform-
ing the T-test. The T-statistic measures the statisti-
cal significance of the coefficients of the independ-
ent variable in explaining the dependent variable y.
Generally, any t value greater than +2 or less than –2
is acceptable. The higher the t value, the greater the
confidence we have in the coefficient as a predic-
tor. Low t values are indications of low reliability of
the predictive power of that coefficient (Draper and
Smith, 1998).
The outcomes of the T test analyses are summa-
Fig. 2. Regression models of cʹ and IP.
rized in Table 4. All models passed the test with the
exception of the Quadratic one. The coefficient b2 of
the model is insignificant and it has been rejected.
The derived regression models, which passed
Table 3 provides the results from these tests. all statistical tests and their mean absolute errors
Pearson’s correlation coefficient (R) gives informa- (MAE), are presented in Table 5. In judging the
tion about the magnitude of the correlation, as well efficiency of the regression models, the lower
as the direction of the relationship. The obtained R MAE is the better model. A MAE of zero means
values range from 0.953 to 0.983. This indicates a perfect model. An advantage of MAE is the fact
that there is a very strong and positive correlation that its score is in the same units as the variable
between cʹ and IP. of interest.
The coefficient of determination (R2) was also Тhe exponential and the logarithmic regression
obtained, as it provides information of how good a equations show the highest MAE with values of
model fits the data. The derived values for R² are 3.75 kPa and 3.59 kPa, respectively. The MAE of
closer to 1 (Table 3), signifying that the regression the regression for the linear model (2.50 kPa) is
models cover most of the variance of values for the almost as low as the MAE for the power model
dependent variable (from 90.8% to 96.6%); the total (2.57 kPa). The difference between them is so small
variation of cʹ was explained by IP and just about that we could use either one for the best-fitted model
9.2% of it are unclear. if we did not take into account the R2 value.
The exponential model has the lowest R2 value Regression model selection criteria used in the
(0.908), but it still had a strong relationship. The current paper are the highest R2 value, least MAE
quadratic model has the highest R2 value, but it is di- and statistical significance of the model and its

47
Boriana Tchakalova, Plamen Ivanov

Table 4
T-test results of regression models

Regression T-test
model b0 t p-value b1 t p-value b2 t p-value
Exponential 14.613 21.011 0.000 0.022 17.457 0.000
Linear 8.476 8.487 0.000 0.776 28.820 0.000
Logarithmic –46.186 –11.205 0.000 24.254 19.532 0.000
Quadratic 6.520 3.322 0.002 0.915 7.439 0.000 –0.002 –1.156 0.257
Power 2.767 21.607 0.000 0.727 21.607 0.000

Table 5
Regression model equations

Regression Adjusted
Equation MAE,%
model R2
Exponential cʹ = 14.613 * e0.022IP 0.905 3.746
Linear cʹ = 8.476 + 0.776IP 0.962 2.502
Logarithmic cʹ = 24.254 * ln(IP) – 46.186 0.922 3.590
Power cʹ = 2.767 * IP0.727 0.936 2.570

Table 6
Statistical data of the best fitted model

F-test T-test
Equation R R2 MAE
F p value b0 p value b1 p value
cʹ = 8.476 + 0.776*IP 0.982 0.964 2.502 830.572 0.000 8.476 0.000 0.776 0.000

coefficients. The best-fitted regression model for

predicting cʹ of clay soils from their IP accord-
ing to these criteria is the linear regression model
(Table 6).
The uncertainty of the derived linear model is
calculated by a 95% confidence interval (blue band)
and 95% prediction interval (gray band; Fig. 3). The
limitation of the suggested equation is that it is ap-
plicable for the values of IP in the range of 11% to
78% and can be used just for preliminary estimation
of cʹ.

CONCLUSION

In order to provide a correlation between the ef-

fective cohesion and plasticity index for natural
undisturbed soils from NW Bulgaria, 33 clay soil
specimens were laboratory tested to determine their
index properties and effective cohesion. Correla-
Fig. 3. Uncertainty of the derived linear model. tion and regression analysis were performed with

48
 Correlation between effective cohesion and plasticity index of clay

plasticity index as an independent variable, and ef- Acknowledgements

fective cohesion as a dependent variable. The test
The authors wish to thank two anonymous review-
results confirm that the effective cohesion can be re-
ers for their constructive reviews and suggestions,
lated to the plasticity index. It was observed that the which improved the quality of the manuscript. This
linear regression model gives the best equation with work has been carried out in the framework of the
a coefficient of determination of 0.964 and mean National Science Program “Environmental Protec-
absolute error of 2.50 kPa. The derived equation tion and Reduction of Risks of Adverse Events and
likewise appears to be appropriate to a wide range Natural Disasters”, approved by the Resolution of
of clays with an IP value in the range of 11% to 78%. the Council of Ministers No. 577/17.08.2018 and
The authors believe the suggested correlation would supported by the Ministry of Education and Sci-
be a useful assessment tool for the preliminary de- ence (MES) of Bulgaria (Agreement No. DO1-
sign stages. 279/03.12.2021).

REFERENCES

Atterberg, A. 1911. Die Plastizität der Tone. Internationale Mehta, B., Sachan, A. 2017. Effect of mineralogical properties
Mitteilungen der Bodenkunde 1, 4–37. of expansive soil on its mechanical behavior. Geotechnical
Draper, N.R., Smith, H. 1998. Applied Regression Analysis. and Geological Engineering 35, 2923– 2934, https://doi.
Third Edition. Wiley & Sons Inc., New York, 736 pp., org/10.1007/s10706-017-0289-6.
https://doi.org/10.1002/9781118625590. Seed, H.B., Woodward, R.J., Lundgren, R. 1964. Fundamental
Dolinar B, Trauner L. 2007. The impact of structure on the aspects of the Atterberg limits. Journal of Soil Mechan-
undrained shear strength of cohesive soils. Engineer- ics and Foundations Division, Proceedings of ASCE 90,
ing Geology 92 (1–2), 88–96, https://doi.org/10.1016/j. 75–105.
enggeo.2007.04.003. Sørensen, K.K.; Okkels, N. 2013. Correlation between drained
Fener, M., Kahraman, S., Bay, Y., Gunaydin, O. 2005. Cor-
shear strength and plasticity index of undisturbed overcon-
relations between P-wave velocity and Atterberg limits
solidated clays. Proceedings of the 18th ICSMGE, Paris,
of cohesive soils. Canadian Geotechnical Journal 42 (2),
673–677, https://doi.org/10.1139/t04-102. France, 423–428.
Frost, J. 2020. Regression Analysis: An Intuitive Guide for Us- Spagnoli, G., Shimobe, S. 2020. Statistical analysis of some cor-
ing and Interpreting Linear Models. Statistics by Jim Pub- relations between compression index and Atterberg limits.
lishing, State College, Pennsylvania, 358 pp. Environmental Earth Sciences 79, https://doi.org/10.1007/
Lambe, T.W. 1960. A mechanistic picture of shear strength in s12665-020-09272-0.
clay. Proceedings of the Research Conference on Shear Yong, R.N., Warkentin, B.P. 1966. Introduction to Soil Behav-
Strength of Cohesive Soils, Boulder, Colorado, 555–580. ior. The MacMillan Company, New York, 451 pp.

49

View publication stats