- Trang Chủ
- Địa Lý
- On the influence of the soil and groundwater to the subsidence of houses in Van Quan, Hanoi
Xem mẫu
- VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51
Original Article
On the Influence of the Soil and Groundwater to the
Subsidence of Houses in Van Quan, Hanoi
Dinh Xuan Vinh
Hanoi University of Natural Resources and Environment, 41 Phu Dien, Tu Liem, Hanoi, Vietnam
Received 11 January 2020
Revised 14 April 2020; Accepted 22 August 2020
Abstract: The area of Van Quan, Hanoi before 2004 was the rice field. Nearby, Ha Dinh water plant
has well-drilled underground water for residential activities. Van Quan's new urban area after being
formed has detected many subsidences. The objective of this study is to assess the main causes of
the subsidence of the houses, based on groundwater and soil. This paper applied the regression
method to study the effect of soil and groundwater on the residential constructions in Van Quan
urban area, Hanoi. Subsidence monitoring was carried out for 4 consecutive years, from 2005 to
2009, including over 500 subsidence monitoring points with high-precision Ni007 and INVAR
gauges. A groundwater observation well is 30 meters deep at the site of the settlement. The results
show a small effect of groundwater on subsidence. The characteristics of the young sediment area
and the soil consolidation process are the main causes leading to serious subsidence in residential
constructions in Van Quan urban area. This paper provides a different perspective on the impact of
groundwater on the subsidence of residential structures within approximately 100 ha.
Keywords: monitoring, subsidence, residential houses, groundwater, soil.
1. Introduction activities, and construction process. urban floor.
In this paper, we want to explore the impact of
The situation of land subsidence in the groundwater on the upper floor and the
region due to various subjective and objective consolidation process of soil on shallow
causes that many scientists as Tuong The Toan, foundation constructions, in particular, houses
Tu Van Tran, Ty Van Tran [1-3] agreed as under 5 floors in Van Quan urban area, Hanoi.
follows: Characteristics of sedimentary basins We have built a groundwater monitoring well
during consolidation, denudation or accretion of with a depth of 30 meters in the survey area.
topographic surfaces, groundwater extraction Observation data of groundwater and subsidence
________
Corresponding author.
E-mail address: dxvinh@hunre.edu.vn
https://doi.org/10.25073/2588-1094/vnuees.4539
42
- D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51 43
of residential houses of Van Quan urban area type can be interpreted as internal causes and
were conducted regression analysis. Thereby we results within the system. This format includes
assess the influence of each cause to the multiple regression (MR) model, stepwise
settlement of the houses on the young regression (SR), principal component regression
sedimentary basin. (PCR), partial least square regression (PLSR)
Some studies use the method of Terzaghi as and artificial neural network (ANN). The second
Ty Van Tran, Hiep Van Huynh [3], or the Finite model is based on the statistical rule of
Element method as Tu Van Tran et al [2], based dependent variables ie using linear statistical
on groundwater monitoring data to forecast models themselves, not by other environment
ground subsidence. In this study, we use the variables. They do not establish a model between
groundwater monitoring data in the subsidence cause and effect. This type includes Time series
area (about 100 hectares) and the subsidence (TS series), Gray system (GS). The deformation
monitoring data of the houses according to prediction model is based on information drawn
national Class II leveling Regulation. from the deformation monitoring data series,
Conducting the regression analysis for each these processes are performed in different ways.
cause of subsidence. The first is groundwater. Parameter model based on the analysis of
The second is the during consolidation monitoring data by continuous mechanical rules.
subsidence of the soil because Van Quan urban First, determine the relationship between the
area is located on a young sedimentary basin [2]. dependent variables and the independent
variables built on mechanical rules. Next, linear
statistics are applied to correct the assumed
2. Research Methods and Data calculation values or parameters throughout the
The raw monitoring data including calculation. This model type has a Kalman filter
appropriate measurements is a very important [4].
part of the building safety data. Based on the Regression analysis is a statistical method
monitoring data, one can recheck the design plan where the expected value of one or more random
as well as the construction process and the variables is predicted based on the condition of
operation of the building. The raw data provide other (calculated) random variables. Regression
valuable information that sheds light on the analysis is not just about curve matching
stability of the building. However, the raw data (choosing a curve but best matching a set of data
cannot reveal the shifting field or the points); it must also coincide with a model of
deformation trend of the building. A deterministic and stochastic components. The
comprehensive analysis is therefore needed to defined component is called the predictor and the
accurately and comprehensively identify various random component is called the error term.
deformations from a large volume of raw data. Regression analysis is both a mathematical-
Two types of dynamic models are formulated to statistical method and a deformation physics
analyze deformation monitoring test data, non- explanation, so it can be used to predict
parametric models based on mathematical- deformation. Calculation of univariate or
statistical theory, and principles-based parametric multivariate regressions is the solution a system
models major of continuous mechanics. of linear equations based on the least-squares
Non-parametric model based on principle the functional model is represented as
mathematical - statistical prediction algorithms. a matrix.
The first model is based on a functional 𝑌 = 𝑋𝛽 + 𝜀 (1)
relationship between the independent variables In this model, Y is a dependent variable, that
(the environment variables) and the dependent is, the vector of deformation measurement,
variables (are the deformations). Models of this matrix representing the component of the
- 44 D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51
dependent variable is 𝑌 𝑇 = (𝑦1 , 𝑦2 , … , 𝑦𝑛 ), n is For multivariate linear regression equations,
the amount of measurement; Equation (1) has we find the estimate 𝛽̂ by the least-squares
many variables x and each variable has a parameter method so that
β that needs to be estimated; The vector of random 2
error ε is the deviation of measured value (RMS ∑(𝑦𝑖 − 𝑦̂𝑖 )2 = ‖𝑌 − 𝑌̂‖ = ‖𝑒‖2 = 𝑚𝑖𝑛
measured value), 𝜀 𝑇 = (𝜀1 , 𝜀2 , … , 𝜀𝑛 ). Where 𝑖
the measurements are random components and
We obtain vector
follow the standard distribution rule 𝑁(0, 𝜎 2 ),
we can apply the Gauss - Markov procedure. The 𝛽̂ = (𝑋 𝑇 𝑋)−1 𝑋 𝑇 𝑌
random model is
and posterior accuracy
∑ 𝜀𝜀 = 𝐸{𝜀. 𝜀 𝑇 } = 𝜎 2 𝑄𝜀𝜀 ∑𝛽̂𝛽̂ = 𝜎02 𝑄𝛽̂𝛽̂ = 𝜎02 . (𝑋 𝑇 𝑋)−1
} (2)
𝑄𝜀𝜀 = 𝐼 Elements on the diagonal of the covariance
X is a matrix of the form matrix ∑𝛽̂𝛽̂ are the variances of the estimates 𝛽𝑗
1 𝑥11 𝑥12 ⋯ 𝑥1𝑚 ie 𝑞𝛽̂𝛽̂ = 𝑆𝛽2𝑗 .
1 𝑥21 𝑥22 ⋯ 𝑥2𝑚
𝑋=[ ] (3) Post-regression values
⋮ ⋮ ⋮ ⋮ ⋮
1 𝑥𝑛1 𝑥𝑛2 ⋯ 𝑥𝑛𝑚 𝑌̂ = 𝑌 + 𝑉 = 𝑋𝛽̂ = 𝑋(𝑋 𝑇 𝑋)−1 𝑋 𝑇 𝑌 = 𝐻𝑌
Matrix (3) shows m deformation-causing The H-matrix is called the "hat" matrix [5].
factors, each deformation-causing factor The principles of a multivariate linear
represents a measure of an independent variable regression model and solutions are consistent
or its function, they form the elements of the with the indirect adjustment model and the
matrix X, similar for the dependent variable there common solution in surveying, but different in
are all n groups; that: the number of causes of deformation
β is the regression coefficient vector, 𝛽 𝑇 = influence in the multivariate linear regression
(𝛽0 , 𝛽1 , … , 𝛽𝑚 ). Where: model has not been predetermined, it is
𝛽0 is the coordinate origin coefficient; necessary to use a certain method to defined
regression, making the optimal regression model.
𝛽1 is the slope coefficient of Y according to
In linear regression analysis, we include the
the variable 𝑥1 and keeping the variables
following concept: Residual Sum of Square (Q),
𝑥2 , 𝑥3 , … , 𝑥𝑚 constant;
Total Sum of Square (S) and Explained Sum of
𝛽2 is the slope coefficient of Y according to Squares (U). We have
the variable 𝑥2 and keeping the variables 𝑌− = +(𝑌̂ − 𝑌̄) (4)
𝑥1 , 𝑥3 , . .. , 𝑥𝑚 constant;
,... The concepts are defined as follows:
𝑛
𝛽𝑚 is the slope coefficient of Y according to 𝑆 = (𝑌 − 𝑌̄)𝑇 (𝑌 − 𝑌̄) = ∑(𝑦𝑖 − 𝑦̄ )2
the variable 𝑥𝑚 and keeping the variables 𝑖=1
𝑥1 , 𝑥2 , . .. , 𝑥𝑚−1 constant. 𝑛
𝑇
The slope coefficient 𝛽1 represents the 𝑄 = (𝑌 − 𝑌̂) (𝑌 − 𝑌̂) = ∑(𝑦𝑖 − 𝑦̂)2 = 𝑉 𝑇 𝑉
change in the mean of Y per unit of change of 𝑥1 𝑖=1
𝑛
regardless of the change of 𝑥2 , 𝑥3 , . .. , 𝑥𝑚 , so the 𝑇
𝛽𝑗 is also called partial regression coefficients. 𝑈 = (𝑌̂ − 𝑌̄) (𝑌̂ − 𝑌̄) = ∑(𝑦̂𝑖 − 𝑦̄ )2
𝑖=1 }
- D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51 45
Where the deformation-cause factor and the measured
𝑛 deformation values are not related to each other.
1 If the coefficient is close to -1 or +1, the
𝑦̄ = ∑ 𝑦𝑖
𝑛 deformation-cause factor and measured strain
𝑖=1
𝑦̂𝑖 is the regression value of the dependent value have a great relationship. We have
variable.
𝑅 2 = 𝑈⁄𝑆
Can prove that: 𝑆 = 𝑄 + 𝑈
We have conducted a groundwater
In regression, the correlation coefficient (R)
monitoring of a well built in the urban area of
is a statistical index that measures the degree of
Van Quan (Figure 1). Simultaneously with
correlation between deformation-cause factors
monitoring the subsidence time of the houses
and measured deformation values [6]. The
(Figure 2), we conduct monitoring the
correlation coefficient is close to 0, meaning that
groundwater level (Figure 3).
Figure 1. Groundwater monitoring well. Figure 2. Cracks on Van Quan houses.
-10,300
-10,350
-10,400
-10,450
-10,500
-10,550
-10,600
-10,650
9/12/05
2/14/06
11/4/06
11/6/06
05/09/2005
23/09/2005
28/10/2005
12/11/2005
25/11/2005
23/12/2005
08/01/2006
3/3/06
12/02/2007
(26/11/2007)
(21/01/2008)
(25/03/2008)
(26/05/2008)
(26/07/2008)
(26/09/2008)
(24/11/2008)
23/1/2006
11/10/2006
11/12/2006
11/8/2006
(1/10/2007)
(31/3/2009)
09/10/05
(15/05/07)
(27/07/07)
(5/2/2009)
Figure 3. Groundwater level in Van Quan area during monitoring of subsidence.
- 46 D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51
Figure 4. The groundwater monitoring well, the points of measurements and the boreholes.
Monitoring data from May 2005 to March The regression model we build is based on a
2009. Subsidence monitoring is done by high- finite set of measurement data, so it may be
precision leveling Ni007 and Invar gauges. The affected by measurement errors ε. We have the
measurement technique complies with the following hypothesis
national grade II standard. Monitoring the 𝐻0 : 𝛽0 = 𝛽1 = 𝛽2 = ⋯ = 𝛽𝑚 = 0
groundwater level with the Piezometer gauge.
(Figure 4). 𝐻1 : Have at least one coefficient 𝛽𝑗 ≠ 0
If the assumption H0 is true, that is, all slope
coefficients are zero, then the regression model
3. Theory and Calculation
built has no effect in predicting or describing the
Methods of assessing the conformity of the dependent variable. Formulation
regression model according to mathematical 𝑈
statistics include: Calculating the correlation 𝐹𝑡𝑡 = 𝑚 (5)
𝑄
coefficient R, using statistical tests to evaluate (𝑛 − 𝑚 − 1)
the overall model, calculating standard errors of
estimates, statistical tests list each individual In this formula, U and Q are known, n and m
independent variable. In geodesy, we are are sample size (number of measurements) and
interested in testing the overall regression model independent variable (number of factors
and testing the dominance of each deformation affecting deformation into the model),
effect factor (such as temperature, time, respectively. The degree of freedom of the
pressure,...) on the dependent variable numerator f1 = m, the degree of freedom of the
(deformation values). denominator f2 = (n-m-1). Select the confidence
level for the F statistic with 95%, that is, the
- D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51 47
alpha level for the test is 5%. Look up ∆𝑄 = 𝑄𝑚 − 𝑄𝑚+1
distribution table F to find the limit value 𝐹𝑓1 ,𝑓2 ,𝛼 . {∆𝑈 = 𝑈𝑚 − 𝑈𝑚+1
If Ftt > F limited, reject the H0 hypothesis. The F ∆𝑄 = ∆𝑈
statistic must be used in combination with the
Thus, the residual sum of squares increases
significance level value when you are deciding if
by the reduction of the explained sum of squares
your overall results are significant.
after increasing the deformation-cause factor xm
Test the dominance of each factor affecting + 1, through which the regression equation also
deformation (such as temperature, pressure, reflects the contribution of the additional
time,... ) to the dependent variable (is the increase factor with the regression effect. The
measured deformation value). We have the predominance test for the added deformation-
following hypothesis cause factor is as follows
𝐻0 : 𝐸(𝛽̂𝑗 ) = 0 𝐻0 : 𝐸(𝛽̂ ′𝑚+1 ) = 0
𝐻𝐴 : 𝐸(𝛽̂𝑗 ) = 𝛽̂𝑗 ≠ 0 𝐻𝐴 : 𝐸(𝛽̂ ′𝑚+1 ) = 𝛽̂ ′𝑚+1 ≠ 0
Create the following statistics according to Forming the F statistical distribution
∆𝑄
the T distribution 𝐹=
𝛽̂𝑗2 𝑄𝑚+1
⁄(𝑛 − 𝑚 − 2)
𝑞𝛽̂𝑗𝛽̂𝑗
𝑇= < 𝑇𝑛−𝑚−1,𝛼 (6) ∆𝑄
𝑄 ⁄(𝑛 − 𝑚 − 2)
2
(𝑛 − 𝑚 − 1) = ~𝐹1,𝑛−𝑚−2 (7)
𝑄𝑚+1
qβ̂ β̂ is the jth element on the main diagonal of Taking the significance level of 5%, when
j j
the matrix 𝑄𝛽̂𝛽̂ , where 𝑞𝛽̂𝑗𝛽̂𝑗 is the variance of 𝐹> 𝐹1, 𝑛 − 𝑚 − 2, 𝛼, the original hypothesis is
accepted, that is, the increased deformation-
the regression coefficient estimates (𝑆𝛽2𝑗 ); Q is cause factor has a significant effect on the
the residual sum of square. Look at the house's deformation, in contrast. it should not be
distribution table of T, get significance level of added. In the regression equation, the influence
5%, dominance of deformation influence factors of deformation often correlate with each
coefficient 𝛽̂𝑗 is 95% respectively. If 𝑇 < other, that is, there is some relation to each other.
𝑇𝑛−𝑚−1,𝛼, then the corresponding deformation- The close correlation between the variables in
2 the regression model created a multicollinearity
cause factor 𝑥𝑗 has a very small effect on
phenomenon, making the variance of the
deformation, which can be removed from the regression coefficient estimates big valuable.
regression equation. The multicollinearity phenomenon also reverses
In the regression model, we must put the the regression coefficient, instead of positive
deformation-cause factors into the regression coefficients, that is, the high water level causes
equation. In the process of testing their the deformation of the dam to be large, resulting
dominance, if any factors do not pass the test, in negative results, the high water level makes
they will be removed, and other factors must be the dam less deformed.
included in the evaluation model. Assume a Based on the above test steps, it is possible
following multivariate linear regression equation to induce the following step regression:
𝑦̂ = 𝛽̂0 + 𝛽̂1 𝑥1 + ⋯ + 𝛽̂𝑚 𝑥𝑚 a) Prequalification of independent variables
affecting the deformation
The residual sum of squares and the
explained sum of squares is Qm + 1, Um + 1, now we b) Determine the first univariate linear
have regression equation. Assuming that m
- 48 D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51
independent variables affect deformation, each 𝑌̂ = 𝛽0 + 𝛽1 𝑥𝛾
of these independent variables creates a
The linear regression equation for time
univariate linear regression equation, for a total
of m equations. Calculate the residual sum of 𝑌̂ = 𝛽0 + 𝛽2 𝑥𝜃 + 𝛽3 𝑥2𝜃
squares Q of each equation. If the regression Based on the observed data series we have
equation with 𝑄𝑘 = 𝑚𝑖𝑛{𝑄𝑖 }, 𝑖 = ̅̅̅̅̅̅1, 𝑚, then the following regression equation
the regression equation with Qk is collected after - For the effect of groundwater on the
testing its according to equations (6) and (7). subsidence of houses
c) Determine the best two-variable regression 𝑌̂ = 9876.1124 + 309.3856 𝑥𝛾 + 56.5974
equation based on the univariate linear regression The correlation coefficient 𝑅 2 = 0.0628 =
equation, in turn increasing the independent 6.28%, that is, the water table affects only 6.28%
variables affect deformation, and have (m-1) two to the subsidence of the structure. The posterior
linear regression equations. Calculate (m-1) the error of regression is 56.5974 mm. The posterior
residual sum of squares ΔQ, consider the error of the estimated coefficient 𝛽1 is 𝑆𝛽1 =
difference ∆Q j = max{∆Q i }, i = ̅̅̅̅̅
1, m. 158.29. The test value according to (6) for 𝛽1 is
The jth incremental independent variable is T = - 1.9545, corresponding to the significance
the “waiting” independent variable, conducting level of 5.55%. The correlation coefficient is too
its test, if adopted, it will be included in the low and the post-estimation error 𝑆𝛽1 is too high,
equation. It is the best two-variable linear so we remove the groundwater element from the
regression equation. If not, then stop at the regression model.
univariate regression equation. - For the effect of soil consolidation time on
d) If two independent variables affecting the subsidence of the houses
deformation are dominant for dependent variable 𝑌̂ = 6694.9641-1.4108 𝑥𝜃 +0.0024 𝑥2𝜃 + 6.7862
Y (amount of deformation), then according to the
above method, continue to select independent The correlation coefficient 𝑅 2 = 0.9809 =
variables to affect the third and fourth 98.09%, ie the time of soil consolidation affects
deformation,... So on until it is impossible to 98% of the settlement of the building. The slope
increase the new independent variable and can coefficient 𝛽2 indicates the settlement rate and
not remove any independent variables selected, 𝛽3 indicates the settlement acceleration is 0.0024
then stop. As a result, we have the best mm2 /week. The posterior error of the regression
regression model. is 6.7862 mm. The posterior error of the
estimated coefficient 𝛽2 is 𝑆𝛽2 = 0.0422, the
The independent variable affecting
coefficient 𝛽3 is 𝑆𝛽3 = 0.0002. The test value
deformation is groundwater and time. The
observation time characterizes the deformation according to (6) for 𝛽2 is T = - 33,4372,
of the test point over time, so its first-order corresponding to the significance level of 6.6.10-
differential is the subsidence rate, its second- 74%, and 𝛽3 is T = 9.8588, corresponding to the
degree differential is the subsidence significance level of 2.3.10-16%, the value This
acceleration. Simultaneous time represents the is very small by our standards (5%).
level of consolidation of the soil under the
construction. It can be said that: the consolidation 4. Results and Discussion
subsidence time lasts correspondingly the soil Based on the results of regression analysis of
belongs to young sediments. the causes of subsidence of residential houses,
Develop a regression equation for the groundwater level and the time of consolidation
groundwater variable γ and for time variable θ. of the soil from 2005 to 2009, we can draw a
We have a linear regression equation for regression line of subsidence according to the
groundwater consolidation time of the soil background.
- D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51 49
6750
6700
6650
mm (Subsidence)
6600
6550
6500
6450
Actual Regression
6400
28/10/05
19/12/06
12/11/07
25/12/07
11/11/08
25/12/08
5/9/05
3/5/06
2/8/06
3/2/07
5/5/07
(25/06/07)
4/2/08
7/5/08
12/9/05
23/1/06
17/6/06
18/9/06
3/11/06
19/3/07
11/8/07
1/10/07
29/6/08
10/8/08
26/9/08
12/2/09
25/3/09
17/3/06
25/3/08
Figure 5. Soil consolidation plays a major role in subsidence of the Van Quan houses.
Although some scientific studies suggest that in the period 2005-2009. The fluctuations are
the groundwater level strongly affects the mainly recorded during the rainy and dry
background subsidence. But to consider specific seasons. Because of the relatively stable
residential constructions, when the soil groundwater level in Van Quan, it cannot cause
background is loaded with the houses under 5 the subsidence of residential houses.
floors with the foundation structure without For the young sedimentary areas, the
reinforced concrete piles. This case has shown consolidation element subsided over time,
that the cohesive subsidence factor of the soil is constructions from three floors should have
the main cause of the subsidence of the houses. reinforced concrete foundation piles, constructed
The underground water observation well in by the method of pressing piles. The depth of
Van Quan urban area is made of Tien Phong reinforced concrete piles should exceed the fill
plastic pipe with a diameter of 90 mm, a depth of and soft soil layers, for Van Quan area is about
30 m from the protective steel pipe mouth on the 15 m depth, based on the geological survey
ground, the bottom of the tube is in direct contact drilling boreholes (Figure 6).
with the soil and is not prevented way. Due to In fact, after 2008, most of the residential
insufficient funds, we could not build a deeper houses in VanQuan's new urban area have to
groundwater monitoring well, or have a higher reinforce their foundations with piles, increasing
standard. This aquifer is at the top of the construction costs, but ensuring stable and safe
aquifers, not surface water or affected by surface houses for a long time. This is also an experience
water. Monitoring data of groundwater level for civil engineering designers in delta areas with
directly at the well did not notice much change weak soil.
- 50 D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51
Sheet number: 1/2
CYLINDRICAL BOREHOLE
BOREHOLE No.3
Construction Residential houses Coordination:X
Position Van Quan - Ha Noi - Y:
Start day 25/01/2006 End day: 26/01/2006 The height of borehole,m:
0,000
Groundwater level, m : The depth of borehole,m:
53,90
Soil layer Samples SPT
Depth, m
Thickness,m
Height, m
Layer
Depth, m
Number of
Number
Number
Desc r i pt i o n Ex per i men t al c h a r t
Depth, m
hammers
`
From To 15 15 15 N 0 20 40 60 80 100
0
1 Land fill: Sand, clay Number, N
1 mixed with 1
2 construction waste
2
3
3,70 -3,70 3,70 3
4 U1 3,8 4,00
4,0 4,45 1 1 1 2 4 2
5
U2 5,80 6,00 5
6 6,00 6,45 1 1 2 3
6 3
7
U3 7,80 8,00 7
8 8,00 8,45 1 2 2 4
8 4
9
Clay, clay mixed with U4 9,80 10,00 9
10 10,00 10,45 1 1 2 3
dark gray color, mixed
10 3
11 2 with plant organic
matter, plasticity U5 11,80 12,00 11
12 flowing 12,00 12,45 1 2 2 4
12 4
13
U6 13,80 14,00 13
14 14,00 14,45 1 1 2 3
14 3
15
U7 15,80 16,00 15
16 16,00 16,45 1 2 2 4
16 4
17
U8 17,80 18,00 17
18 18,00 -18,00 14,30 18,00 18,45 3 4 3 7
18 7
19
19
20 D2 20,00 20,45 5 7 7 14
20 14
21
21
22 D3 22,00 22,45 5 9 11 20
22 20
23
23
Depth, m
Fine grained sand ash
24 3 D4 24,00 24,45 7 11 12 23
gray, gray, sometimes 24 23
25 mixed with organic,
medium compacted 25
26 state D5 26,0 26,45 7 10 12 22
26
27
27
Note: - M : Original form
- D : Disturbance form
Figure 6. Cylindrical of Borehole No. 3 at the Van Quan residential houses.
- D.X. Vinh / VNU Journal of Science: Earth and Environmental Sciences, Vol. 36, No. 4 (2020) 42-51 51
5. Conclusions References
Regression model is a traditional analytical [1] T.T. Toan. Forecast of land surface subsidence
method to evaluate the impact of independent due to lowering of groundwater level. Final report
of the project RD 9505, Hanoi. 1999. (in
causes on measured values. Groundwater level
Vietnamese).
and soil consolidation process over time are [2] T.V. Tu, H.N. Anh, D.D. Minh, N.M. Tung,
factors to consider when designing a building. Forecast of ground deformation in Ha Dong area
The study showed that the groundwater level in due to urbanization and groundwater extraction.
the upper floor fluctuated very small and 98% of Journal of Earth Sciences 35(1)(2013) 29-35. (in
subsidence of residential houses in VanQuan's Vietnamese).
new urban area was due to the weak soil. [3] T.V. Ty, H.V. Hiep. Current status of groundwater
This study case is only for residential extraction and correlation between water level
buildings from 3 to 5 floors with non-reinforced lowering and land subsidence: Research in Tra
Vinh and Can Tho city. Can Tho University
concrete foundation and only consider the top
Journal of Science. Topics: Environment and
aquifer. For buildings under 3 floors are not
Climate Change 1 (2017) 128-136. (in
covered by this study. Buildings above 5 floors Vietnamese).
often have foundations made of reinforced [4] D.X. Vinh, N.T. Nhung, N.V. Quang.
concrete piles up to a depth of 20 to 60 meters, Determination of Deformation of Construction
so they may be affected by deeper aquifers. More Using Parametric Modeling-Kalman Filter
comprehensive studies are needed on this issue Application and NonParametric Modeling-Time
to be clear about the impact of groundwater on Series Application. VNU Journal of Science:
the subsidence of buildings. Earth and Environmental Sciences 34(3) (2018) 1-
3. https://doi.org/10.25073/2588-1094/vnuees.4274.
Acknowledgments (in Vietnamese).
[5] P.J. Huber, E.M. Ronchetti. Robust Statistics.
The author thanks the support for monitoring Second Edition. Published by John Wiley & Sons,
data of Van Quan of HUDCIC Consulting Inc. Canada. 1981.
Investment and Construction Joint Stock [6] R.A. Maronna, R.D. Martin, V.J. Yohai. Robust
Company. The author also thanks the comments Statistics: Theory and Methods, John Wiley &
of reviewers who helped improve the content of Sons, Ltd. England. 2006.
this article.
nguon tai.lieu . vn
