The assembly of a groundwater flow model for the shallow aquifer in Tianjin Municipality is outlined in this paper. Tianjin Municipality was selected because of its complicated hydrogeological conditions and rich data, which could be used to test a refined groundwater flow model for the shallow aquifer. When a shallow groundwater flow model is being assembled its recharge and discharge functions need to represent inflows from rainfall, irrigation return flows, seepage from rivers and reservoirs, and lateral inflows and outflows from evaporation, abstraction of groundwater for irrigation and industrial, and urban use. When abstracting groundwater, the water exchanges between a shallow aquifer and a deep aquifer also need to be considered. The real irrigation areas of Tianjin Municipality were input into the groundwater flow model, and the rivers and reservoirs were refined to the level of secondary tributaries and small scale reservoirs. The model calibration was carried out based on consideration of representative parameter values and their spatial distribution, the groundwater flow fields, the temporal variation in groundwater heads and the water balance for the years 2006–2008. It was concluded from a comparison of the observed and simulated groundwater heads that the precision of the model is high and that the simulated groundwater levels align with the real groundwater conditions. It is also concluded that the groundwater flow model for the shallow aquifer in Tianjin Municipality will be a useful tool for further studies about the relationship between shallow and deep aquifers and the surface environment.

INTRODUCTION

Groundwater is an important water resource which is widely used in agriculture, industry and in domestic life due to its extensive distribution, good quality and stability of supply. It is the main water supply source in areas with scarce surface water resources. However, over-exploration of groundwater can create a series of environmental and geological problems, such as degradation of the environment and its ecology, creation of cones of depression, ground collapse, ground fractures and saline intrusion (Li et al. 2013). To reduce these negative impacts, it is necessary to implement strict groundwater resource management practices in over-exploited areas.

Groundwater flow models have been proved to be powerful tools which can provide a scientific basis for the management of groundwater (Praveena et al. 2012). Wen et al. (2007) developed a three-dimensional numerical groundwater flow model in the Zhangye basin using FEFLOW. The model was used to predict the drawdown from the years 2000 to 2030 under two different groundwater management scenarios based on recharge of groundwater by inflows from the Heihe River and irrigation water return flows. Zhang et al. (2012) assembled a three-dimensional regional groundwater flow model using MODFLOW to better understand the spatial–temporal variation of groundwater levels and to evaluate the groundwater resources based on rainfall, irrigation and canal seepage as the main sources of recharge, and the evaporation and abstraction from pumping wells as the main discharges. Nettasana et al. (2012) proposed a three-dimensional finite-difference flow model using MODFLOW. In this paper, the inflows from precipitation and the Chi River and Huai Toei reservoir losses were regarded as the sources of recharge of groundwater, and the baseflow of the Chi River and pumping from wells were regarded as the primary discharges. Gonçalves et al. (2013) investigated the Federal District of Brazil where the recharge of groundwater occurred by rainfall and the lateral flow of infiltrated water and proposed a three-dimensional groundwater flow model using OpenGeoSys to represent the groundwater levels and flow systems. Qian et al. (2014) developed a two-dimensional finite-element transient flow model in the confined fractured chalk aquifer in Zinder, Niger using an equivalent porous media model. This fractured aquifer was mainly distributed in the Echkar group aquifer which does not receive direct vertical recharge through precipitation, but rather relies on lateral supply as the main source of recharge.

The above-mentioned studies used modelling systems which are most widely used to model groundwater flows in order to evaluate and manage groundwater resources. However, in most of these studies, the shallow aquifer was not represented separately but rather was represented as the upper boundary of a deep groundwater aquifer. The representation of the shallow aquifer was usually simplified with only the major recharge and discharge components being considered. Consequently, these models are not refined enough to assess problems in the shallow aquifer, such as the impact of the environment on groundwater and the relationship between surface water and groundwater. The assessment of these processes requires a refined groundwater flow model in which all the inputs and outputs must be detailed.

Jusseret et al. (2009) assembled a three-dimensional shallow groundwater flow model using MODFLOW and assessed the relative importance of various recharge sources for the shallow aquifer system and estimated the exchanges between groundwater and the Red River. Chung et al. (2010) expressed the view that groundwater flow models are unreliable without a good estimate of recharge and its spatio-temporal distribution. They assembled an integrated surface water-groundwater model using SWAT-MODFLOW to estimate the impact of spatio–temporal distribution of groundwater recharge on groundwater levels in the Mihocheon watershed in Korea. Barron et al. (2013) studied the impact of urbanization on groundwater systems and their interaction with surface waters using a process-based coupled surface water–groundwater model (MODHMS), which fully considered the recharge of groundwater and estimated the likely changes in river discharge as a result of land-use change. Sallam (2015) assembled a groundwater model to investigate the changes to groundwater systems in shallow aquifers as a result of deep building foundations. These groundwater flow models for shallow aquifers were mainly used to evaluate the groundwater system response to changing environmental conditions. Their recharge components were detailed, however, the hydraulic connection between the shallow aquifer and the deep aquifer were ignored with the bottom of the shallow aquifer regarded as a confining bed. While the shallow and deep aquifers are typically separated by a relative confining bed, water exchanges can arise when the deep aquifer is interfered with by human activities (e.g. groundwater exploration), which can not be ignored when assembling groundwater flow models.

Due to a lack of surface water resources, the groundwater of Tianjin Municipality, which is located around 120 km southeast of Beijing, is abstracted from a deep aquifer and supplies more than 70% of the city's water supply (Dong et al. 2013). This level of groundwater abstraction has resulted in land subsidence. Saltwater has also intruded into the shallow aquifer under Tianjin Municipality. More than 77% of the total area of Tianjin Municipality is underlain by a brackish shallow aquifer (salt content >2 g/L). Most groundwater studies have concentrated on the deep aquifer (Pan et al. 2004; Wang et al. 2010; Li et al. 2012). However any water resource planning for the shallow aquifer including the restoration of the environment and the local ecology and possible management response to the level of saltwater intrusion requires a refined shallow aquifer groundwater flow model. This paper outlines the creation of a three-dimensional groundwater flow model of the shallow aquifer within Tianjin Municipality using the Groundwater Modelling System (GMS). The recharge and discharge components were fully considered as was the water exchange between the shallow and the deep aquifer which is a model output.

THE HYDROGEOLOGICAL SETTING OF TIANJIN MUNICIPALITY

Tianjin Municipality is located around 120 km southeast of Beijing. It covers an area which is 172 km from the south to north and 104 km from east to west with a total area of 10.6 × 103 km2. Tianjin is bounded by the Bohai Sea to the southeast and Yanshan Mountain to the north. The terrain is flat from the piedmont plain to the coastal plain and it is slightly tilted from northwest to southeast. The piedmont floodplain and alluvial fans vary in elevation from 50 m down to 10 m. The adjacent parts are an alluvial plain and a fluvial plain which vary in elevation from 10 m down to 2.5 m. The southeast region is coastal plain which varies in elevation between 1 to 2 m with saltern areas. Figure 1 shows the location of the study area.
Figure 1

Location of Tianjin Municipality.

Figure 1

Location of Tianjin Municipality.

The broad plain from the foothills to the coast is generally divided by the Baodi fracture zone, north of which the buried depth of bedrock is less than 300 m and with a storage of a fourth series of pore groundwater and groundwater in the bedrock, forms a fresh water zone. South of the fracture zone, seawater has intruded into the groundwater and has formed a saltwater zone. The hydrogeological cross section of Tianjin Municipality is shown in Figure 2.
Figure 2

A cross-sectional hydrogeological map of Tianjin Municipality.

Figure 2

A cross-sectional hydrogeological map of Tianjin Municipality.

CONCEPTUAL MODEL OF SHALLOW GROUNDWATER

The distribution and circulation of groundwater in Tianjin Municipality reflects the complex distribution of aquifer media. Based on the consideration of lithologic properties, the hydraulic characteristics of sediment, the level of groundwater development and utilization, the aquifer was classified into four hydrogeological layers. Layer one is the shallow aquifer while the remaining layers represent the deep aquifers. This paper focuses on the shallow aquifer (layer one) which varies in thickness from 30 to 100 m (the shallow aquifer is 80–100 m thick in the fresh water zone and 30–50 m thick in the salt water zone – see Figure 2).

Boundary conditions

In the horizontal plane, the northern boundary is the interface of the mountainous area (the Yanshan Range) and the plain which is represented as a constant flow boundary; the eastern boundary of aquifer is the Bohai Bay shoreline which is represented as a constant head boundary with a water level set at 0 m; the other boundaries were processed as time varying flow boundaries due to the hydraulic connection between Tianjin Municipality and the border regions (see Figure 3).
Figure 3

Initial spatial distribution of groundwater head (red circles indicate observation wells). Please refer to the online version of this paper to see this figure in color: http://dx.doi.org/10.2166/ws.2016.046.

Figure 3

Initial spatial distribution of groundwater head (red circles indicate observation wells). Please refer to the online version of this paper to see this figure in color: http://dx.doi.org/10.2166/ws.2016.046.

In the vertical plane, the free water surface of the shallow aquifer is designated as the upper boundary, which exchanges water with external sources and sinks, while the bottom of the shallow aquifer is designated as the lower boundary, which can exchange water with the deep aquifer. The upper boundary receives recharge from rainfall, irrigation return flows and seepage from rivers and reservoirs and discharges water by evaporation and by groundwater abstraction. In the salt water zone, the deep aquifer is over-exploited while there is almost no abstraction from the shallow aquifer, which leads to a large difference in the water head (7–64 m) between the shallow and the deep aquifer. These head differences increase the water exchange (leakage), consequently the leakage boundary in the salt water zone is generalized as a time varying flow boundary. In the fresh water zone, the presence of bedrock below the shallow aquifer means that the water exchange between the shallow aquifer and the deep aquifer can be ignored.

Initial conditions

Figure 3 plots the location of 74 observation wells which monitor groundwater levels in the shallow aquifer under Tianjin Municipality. There are 36 wells in the fresh water zone with the density 3.6 wells per 100 km2, and there are 38 wells located in the salt water zone with the density 0.4 wells per 100 km2. The density of the shallow aquifer observation wells is greater in the fresh water zone than in the salt water zone due to the level of groundwater abstraction in the fresh water zone in comparison to the reduced groundwater abstraction in the salt water zone. The initial spatial distribution of groundwater head was obtained from the 74 observation wells in January 2006 using Kriging interpolation.

Groundwater recharge and discharge

The recharge of groundwater varies considerably due to the differences of soil type, land-use, topography, etc. The main recharge components in the study area are infiltration from rainfall, lateral flow, irrigation return flow and seepage from rivers and reservoirs. The infiltration of rainfall and irrigation water are generalized as planar recharge sources and processed using the RCH (Recharge) module in the GMS by inputting the intensity of rainfall and irrigation and infiltration coefficients. Lateral inflow is calculated using Darcy's Law based on the water head and hydraulic conductivity. The seepage from rivers is calculated using a water balance method based on runoff data (Zheng 2007). Reservoir seepage is also calculated using the water balance method based on the reservoir storage, water consumption and water level (Yao & Zhang 2008). Figure 4 identifies the location of rivers and reservoirs in Tianjin Municipality.
Figure 4

Location map of rivers and reservoirs in Tianjin Municipality.

Figure 4

Location map of rivers and reservoirs in Tianjin Municipality.

The discharge of groundwater in the study area occurs by harvesting, evaporation and leakage. In the fresh water zone, the groundwater is harvested seriously. The depth of groundwater is from 5 to 10 m below the surface with almost no evaporation, consequently the harvesting is the main discharge of groundwater. However, the distribution and number of wells harvesting groundwater are difficult to survey, so the quantity of harvested groundwater is generalized as planar discharge sources and processed using the RCH (Recharge) (input as a negative recharge rate in the model). The quantity of harvested groundwater is divided into three categories based on its use for irrigation, industrial purposes or domestic purposes. In this paper, the water demand for irrigation is calculated using the planting area and irrigation water ration for crops; and the quantity of groundwater harvested for industrial and domestic purposes are calculated based on the output of industrial products and output values and the population of urban and rural areas, respectively.

In contrast to the fresh water zone, the groundwater in the salt water zone undergoes little harvesting and the depth of groundwater level is from 1 to 3 m below the surface. Evaporation is the main discharge, which is simulated using the Evapotranspiration module in GMS assuming that the local depth limit for evaporation is 4–5 m. At the same time, the groundwater in the deep aquifer is heavily harvested, leading to a big water head difference between the shallow aquifer and the deep aquifer. Consequently leakage from the shallow aquifer can't be ignored. The leakage rate is calculated using the following formula: 
formula
1
where Q is the leakage rate from the shallow to the deep aquifer (m3/d); F is the area of the leakage zone (m2); is the vertical hydraulic conductivity (m/d); M is the thickness of leakage layer (m) and is the head difference between the shallow and the deep aquifers (m).

SHALLOW GROUNDWATER FLOW MODEL

The groundwater flow in the Tianjin Municipality was modeled using the MODFLOW package (Harbaugh 2005) developed by the US Geological Survey. The pre- and post-processor GMS package (Owen et al. 1996) developed by the Environmental Modeling Research Laboratory at Brigham Young University was used to input data and output results.

Mathematical model

Based on Darcy's law and water mass balance, anisotropic and heterogeneous three-dimensional flow of groundwater is described by the partial differential equation as follows: 
formula
2
where H is the hydraulic head (m); W is the source and sink term (m/d); , , are the hydraulic conductivity in x, y, z axes, respectively (m/d); M is the thickness of the layer (m); u is the specific yield; is the initial head (m); is the specified head at boundary (m); n is the normal direction of the boundary surface; q is the groundwater flow rate at boundary (m3/d) and is the leakage zone.

Spatial and temporal discretization

The study area is represented using cells which are 500 × 500 m. The model grid consists of 346 rows and 248 columns with a total of 42,628 active cells. Based on the available data, the period from January 2006 to December 2008 was selected as the identification and calibration period for the model where each month is a stress period that was divided into two time steps (total 36).

Hydrogeology parameters

The hydrogeology parameters used in groundwater flow models are mainly divided into two categories. One category is the parameters used for the calculation of source and sink terms, such as the precipitation infiltration coefficient and the irrigation regression coefficient. Figure 5 is the distribution of precipitation infiltration coefficient while Figure 6 plots the distribution of the irrigation regression coefficient based on the areas under irrigation in Tianjin Municipality. These data were provided by the Geological Survey of Tianjin Municipality. The other category is the hydrogeology parameters for an aquifer, such as horizontal hydraulic conductivity , vertical hydraulic conductivity , specific yield . The initial values of these parameters were derived from the pumping tests. It was assumed also that horizontal hydraulic conductivity is 10 times larger than the vertical hydraulic conductivity .
Figure 5

Precipitation infiltration coefficient distribution.

Figure 5

Precipitation infiltration coefficient distribution.

Figure 6

Irrigation regression coefficient distribution.

Figure 6

Irrigation regression coefficient distribution.

Model calibration

A three-dimensional finite-difference transient flow model was assembled. Before the model was applied it was calibrated against water-level data recorded from January 2006 to December 2008. The calibration of groundwater flow models generally follows the following principles (Zheng 2007):

  1. The identified parameters accord with the real hydrogeological conditions.

  2. The simulated groundwater flow fields are in keeping with the real groundwater flow fields.

  3. The conditions of groundwater head change over time are the same as the observation wells.

  4. The water balance of simulated groundwater conforms with the real conditions.

These principles were applied as follows:
  1. Parameter identification. A method of trial-and-error was used to adjust the hydrogeological parameters and until a good match was achieved between simulated heads and observed heads over time and space (refer Table 1). The final spatial distribution of hydraulic conductivity and specific yield are plotted in Figures 7 and 8.

  2. Groundwater flow fields. The distribution of heads observed in December 2008 was obtained from 74 observation wells (refer Figure 3) using Kriging interpolation and was compared to the simulated heads (refer Figure 9). It can be seen that there is a close agreement between simulated heads and observed heads with the maximum difference between observed and predicted groundwater head being within 2 m.

  3. Groundwater head change over time. A scatter plot of observed heads and simulated heads given in Figure 10 discloses excellent agreement between observed and simulated heads, particularly in the range −5 to +5 m. Time series plots of the simulated heads and observed heads in Wells 1, 2, 3 and 4 are shown in Figure 11. The four observation wells are located in different regions of the study area and represent the changes of regional groundwater. The observed heads exhibit a greater variability than the simulated heads. That is because the shallow aquifer is subject to many variables such as rainfall, irrigation, evaporation and harvesting which vary within each month in comparison to the adopted monthly mean values. Although there are differences between the observed heads and simulated heads, the trends of the groundwater heads are the same.

  4. Water balance. The water balance in the shallow groundwater in years 2006–2008 was analyzed using the groundwater flow model with the results presented in Table 2. The mean groundwater recharge calculated by the model was around 12 × 108 m3/a, while the mean groundwater discharge is around 13.4 × 108 m3/a. Wang et al. (2010), who calculated the water resource of shallow groundwater and deep groundwater in the Tianjin Municipality from 1991 to 2003, determined the annual mean groundwater recharge to be 13.39 × 108 m3/a. The difference from the model results is attributed to the gradual decline in average annual rainfall from 1990 (refer Figure 12) and because rainfall is the main source of recharge in the study area (about 80% of total recharge). Based on the spatial distribution of the recharge modulus of the groundwater provided by the Centre of Hydrographic Water Resources Survey (Figure 13), the mean annual groundwater recharge is calculated to be around 13 × 108 m3/a, which is similar to the model results. It is concluded therefore that the water balance is reasonable.

Figure 7

Hydraulic conductivity distribution.

Figure 7

Hydraulic conductivity distribution.

Figure 8

Specific yield distribution.

Figure 8

Specific yield distribution.

Figure 9

Comparison of simulated and observed groundwater levels in 2008.

Figure 9

Comparison of simulated and observed groundwater levels in 2008.

Figure 10

Comparison of observed groundwater heads and the heads simulated by the transient-state calibration.

Figure 10

Comparison of observed groundwater heads and the heads simulated by the transient-state calibration.

Figure 11

Simulated and observed heads at observation Wells 1, 2, 3 and 4.

Figure 11

Simulated and observed heads at observation Wells 1, 2, 3 and 4.

Figure 12

Change trend of rainfall in Tianjin Municipality.

Figure 12

Change trend of rainfall in Tianjin Municipality.

Figure 13

Distribution of recharge modulus of groundwater.

Figure 13

Distribution of recharge modulus of groundwater.

Table 1

Series values of hydrogeology parameters for different lithology and plains

 Piedmont plain
Central-coastal plains
LithologyHydraulic conductivity (m/d)Specific yieldHydraulic conductivity (m/d)Specific yield
Coarse sand, sand gravel 100–120 0.18–0.23 40–80 0.10–0.16 
Intermediate fine sand, silt sand 6–8 0.09–0.18 15–30 0.05–0.12 
Silt 1.5–2.5 0.03–0.08 3–6 0.035–0.074 
Clay, silt clay 0.8 0.03–0.04 – – 
 Piedmont plain
Central-coastal plains
LithologyHydraulic conductivity (m/d)Specific yieldHydraulic conductivity (m/d)Specific yield
Coarse sand, sand gravel 100–120 0.18–0.23 40–80 0.10–0.16 
Intermediate fine sand, silt sand 6–8 0.09–0.18 15–30 0.05–0.12 
Silt 1.5–2.5 0.03–0.08 3–6 0.035–0.074 
Clay, silt clay 0.8 0.03–0.04 – – 
Table 2

Water balance analysis (x 108 m3) in 2006, 2007 and 2008

Water balance200620072008
Recharge component Rainfall infiltration 7.81 9.51 11.60 
Irrigation leakage 1.42 1.52 1.42 
Lateral inflow 0.54 0.55 0.57 
River and reservoir seepage 0.33 0.33 0.33 
 Total recharge 10.10 11.91 13.92 
Discharge component Harvesting −1.04 −1.07 −0.99 
Evaporation −9.72 −8.79 −9.56 
Leakage −2.78 −3.13 −2.87 
Lateral outflow −0.03 −0.08 −0.13 
 Total discharge 13.57 13.07 13.55 
 Storage variation −3.47 −1.16 0.37 
Water balance200620072008
Recharge component Rainfall infiltration 7.81 9.51 11.60 
Irrigation leakage 1.42 1.52 1.42 
Lateral inflow 0.54 0.55 0.57 
River and reservoir seepage 0.33 0.33 0.33 
 Total recharge 10.10 11.91 13.92 
Discharge component Harvesting −1.04 −1.07 −0.99 
Evaporation −9.72 −8.79 −9.56 
Leakage −2.78 −3.13 −2.87 
Lateral outflow −0.03 −0.08 −0.13 
 Total discharge 13.57 13.07 13.55 
 Storage variation −3.47 −1.16 0.37 

Comparison of models

Li et al. (2012) established a three-dimensional numerical groundwater flow model for the Tianjin Municipality by coupling a one-dimensional soil consolidation model with a MODFLOW model. While this model included the shallow aquifer and the deep aquifers, the shallow aquifer was generalized as the upper boundary of the deep aquifer. Consequently the components of recharge and discharge were not fully considered, which caused differences between the observed heads and simulated heads. It can be seen from Figures 14 and 15 that there was a poor agreement between simulated heads and observed heads in comparison with the level of agreement shown in Figures 9 and 10. It is concluded that the precision of the current model is greater than the Li et al. (2012) model.
Figure 14

Comparison of observed groundwater levels and the levels simulated by the Li et al. (2012) model.

Figure 14

Comparison of observed groundwater levels and the levels simulated by the Li et al. (2012) model.

Figure 15

Comparison of observed groundwater heads and the heads simulated by the Li et al. (2012) model.

Figure 15

Comparison of observed groundwater heads and the heads simulated by the Li et al. (2012) model.

CONCLUSIONS

The assembly of a groundwater flow model for the shallow aquifer in Tianjin Municipality is outlined in this paper. Tianjin Municipality was selected because of its complicated hydrogeologic conditions and rich data which could be used to test a refined groundwater flow model for the shallow aquifer.

The real irrigation areas of Tianjin Municipality were input into the groundwater flow model, and the rivers and reservoirs were refined to the level of secondary tributaries and small scale reservoirs. The water exchanges between the shallow and the deep aquifers could not be ignored, especially when the deep aquifer experiences heavy abstraction for irrigation, industrial and urban use. The lower leakage boundary of the shallow aquifer was generalized as a time varying flow boundary based on the head difference between the shallow and deep aquifer.

The model calibration was carried out based on consideration of representative parameter values and their spatial distribution, the groundwater flow fields, the temporal variation in groundwater heads and the water balance for the years 2006–2008. It was concluded from a comparison of the observed and simulated groundwater heads that the precision of the model is high and that the simulated groundwater levels align with the real groundwater conditions.

It is concluded that the recharge of groundwater in the study area is primarily from rainfall (around 80%), irrigation return water (around 12%), lateral flow (around 5%) and seepage from rivers and reservoirs (around 3%) while the primary groundwater discharge components from the shallow aquifer are evaporation (around 70%), leakage (around 22%), harvesting (around 8%) and lateral outflows (around 1%).

It is also concluded that the groundwater flow model for the shallow aquifer in Tianjin Municipality will be a useful tool for further studies about the relationship between shallow and deep aquifers and the surface environment.

ACKNOWLEDGEMENTS

The authors would like to acknowledge the financial support for this work provided by the National Natural Science Foundation of China (Grant no. 51579169), the Science Fund for Creative Research Groups of the National Natural Science Foundation of China (Grant no. 51321065), and the Ministry of Water Resources Special Funds for Scientific Research on Public Causes (201401041).

REFERENCES

REFERENCES
Barron
O. V.
Donn
M. J.
Barr
A. D.
2013
Urbanisation and shallow groundwater: predicting changes in catchment hydrological responses
.
Water Resour. Manag.
27
,
95
115
. Doi: 10.1007/s11269-012-0168-0.
Dong
D. L.
Sun
W. J.
Zhu
Z. C.
Xi
S.
Lin
G.
2013
Groundwater risk assessment of the third aquifer in Tianjin Municipality, China
.
Water Resour. Manage.
27
,
3179
3190
. Doi:10.1007/s11269-013-0342-z.
Gonçalves
T. D.
Fischer
T.
Gräbe
A.
Kolditz
O.
Weiss
H.
2013
Groundwater flow model of the Pipiripau watershed, federal district of Brazil
.
Environ. Earth Sci.
69
,
617
631
. Doi:10.1007/s12665-013-2400-5.
Harbaugh
A. W.
2005
MODFLOW-2005, The U.S. Geological Survey modular ground-water model – the Ground-Water Flow Process. U.S. Geological Survey Techniques and Methods 6-A16, variously p
.
Jusseret
S.
Tam
V. T.
Dassargues
A.
2009
Groundwater flow modelling in the central zone of Hanoi, Vietnam
.
Hydrogeol. J.
17
,
915
934
. Doi:10.1007/s10040-008-0423-x.
Li
W. Y.
Cui
Y. L.
Su
C.
Zhang
W.
Shao
J. L.
2012
An integrated numerical groundwater and land subsidence model of Tianjin
.
Journal of Jilin University (Earth Science Edition)
42
(
3
),
805
813
.
Li
F. W.
Feng
P.
Zhang
W.
Zhang
T.
2013
An integrated groundwater management mode based on control indexes of groundwater quantity and level
.
Water Resour. Manage.
27
,
3273
3292
. Doi:10.1007/s11269-013-0346-8.
Nettasana
T.
Craig
J.
Tolson
B.
2012
Conceptual and numerical models for sustainable groundwater management in the Thaphra area, Chi River Basin, Thailand
.
Hydrogeol. J.
20
,
1355
1374
. Doi:10.1007/s10040-012-0887-6.
Owen
S. J.
Jones
N. L.
Holland
J. P.
1996
A comprehensive modeling environment for the simulation of groundwater flow and transport
.
Eng. Comput.
12
,
235
242
. Doi:10.1007/BF01198737.
Pan
Y.
Pan
J. G.
Gong
H. L.
Zhao
W. J.
2004
Research on the relation between groundwater exploitation and subsidence in Tianjin proper
.
Earth and Environ.
32
(
2
),
36
39
(
in Chinese
).
Praveena
S. M.
Abdullah
M. H.
Bidin
K.
Aris
A. Z.
2012
Sustainable groundwater management on the small island of Manukan, Malaysia
.
Environ. Earth Sci.
66
,
719
728
. Doi:10.1007/s12665-011-1279-2.
Qian
J.
Zhou
X.
Zhan
H.
Dong
H.
Ma
L.
2014
Numerical simulation and evaluation of groundwater resources in a fractured chalk aquifer: a case study in Zinder well field, Niger
.
Environ. Earth Sci.
72
,
3053
3065
. Doi:10.1007/s12665-014-3211-z.
Wang
J. B.
Li
P.
Zhang
B. M.
Wang
J. B.
2010
Allowable groundwater withdrawal and its determination basis in Tianjin Plain
.
Earth Sci. Front.
17
(
6
),
221
226
(
in Chinese
).
Wen
X. H.
Wu
Y. Q.
Lee
L. J. E.
Su
J. P.
Wu
J.
2007
Groundwater flow modeling in the Zhangye Basin, Northwestern China
.
Environ. Geol.
53
,
77
84
. Doi:10.1007/s00254-006-0620-7.
Yao
D. H.
Zhang
Q.
2008
Calculation and analysis of seepage quantity of Luowutian Reservoir in Kuiyong town of Shenzhen City
.
Resour. Environ. Eng.
22
(
3
),
336
338
(
in Chinese
).
Zhang
Y.
Ma
J.
Chang
X.
van Wonderen
J.
Yan
L.
Han
J.
2012
Water resources assessment in the Minqin Basin: an arid inland river basin under intensive irrigation in northwest China
.
Environ Earth Sci.
65
,
1831
1839
. Doi:10.1007/s12665-011-1165-y.
Zheng
H. M.
2007
Groundwater Three-Dimensional Numerical Simulation of Tianjin
.
China University of Geosciences
,
Beijing
(
in Chinese
).