Environ Monit Assess (2016) 188: 431 DOI 10.1007/s10661-016-5414-0
Zonal management of multi-purposes groundwater utilization based on water quality and impact on the aquifer Ching-Ping Liang & Cheng-Shin Jang & Ching-Fang Chen & Jui-Sheng Chen
Received: 18 October 2015 / Accepted: 12 June 2016 / Published online: 24 June 2016 # Springer International Publishing Switzerland 2016
Abstract Groundwater is widely used for drinking, irrigation, and aquaculture in the Pingtung Plain, Southwestern Taiwan. The overexploitation and poor quality of groundwater in some areas of the Pingtung Plain pose great challenges for the safe use and sustainable management of groundwater resources. Thus, establishing an effective management plan for multi-purpose groundwater utilization in the Pingtung Plain is imperative. Considerations of the quality of the groundwater and potential impact on the aquifer of groundwater exploitation are paramount to multi-purpose groundwater utilization management. This study proposes a zonal management plan for the multi-purpose use of groundwater in the Pingtung Plain. The zonal management plan is developed by considering the spatial variability of the groundwater quality and the impact on the aquifer, which is defined as the ratio of the actual groundwater
Electronic supplementary material The online version of this article (doi:10.1007/s10661-016-5414-0) contains supplementary material, which is available to authorized users. C.
extraction rate to transmissivity. A geostatistical Kriging approach is used to spatially delineate the safe zones based on the water quality standards applied in the three groundwater utilization sectors. Suitable zones for the impact on the aquifer are then spatially determined. The evaluation results showing the safe water quality zones for the three types of utilization demands and suitable zones for the impact on aquifer are integrated to create a zonal management map for multi-purpose groundwater utilization which can help government administrators to establish a water resource management strategy for safe and sustainable use of groundwater to meet multipurpose groundwater utilization requirements in the Pingtung Plain. Keywords Multi-purpose groundwater utilization . Zonal management . Water quality standard . Impact on the aquifer . Transmissivity
Introduction Groundwater is a crucial resource used by people worldwide. In Taiwan, groundwater comprises approximately 32 % of the annual water supply and is thus considered a vital resource. The safe and sustainable management of groundwater involves great challenges in Taiwan where land subsidence and sea water intrusion induced by groundwater overexploitation have reduced drainage capability, increased inundation due to storm surges, and caused permanent damages to buildings and infrastructure.
431 Page 2 of 17
In addition to long-term problems induced by overexploitation, groundwater generally contains a wide spectrum of naturally occurring inorganic and organic chemicals. The inorganic chemicals originate from the soil and mineral grains through which the groundwater flows, whereas the organic chemicals originate from rotting plant and animal material and microorganisms. In addition, groundwater may be contaminated by anthropogenic activities. Hazardous constituents can substantially restrict its use for drinking, irrigation, and aquaculture if they are present in concentrations higher than certain thresholds. For example, arsenic (As) is recognized as a toxicant and carcinogen. Numerous studies have indicated that exposure to As can cause various types of acute and chronic health effects including blackfoot disease (Tseng 1977), also known as an endemic peripheral vascular disease, cancers of the liver, kidney, bladder, prostate, lymphoid tissue, skin, colon, lung, and nasal cavity (Wu et al. 1989; Chen and Wang 1990; Chen et al. 1985), ischemic heart disease (Hsueh et al. 1998); hyperpigmentation; hyperkeratosis; diabetes (Tseng et al. 2000); and meningioma. The As from contaminated groundwater used for irrigation and aquaculture may accumulate in seafood and crops, posing a potential threat to human health (Liang et al. 2010, 2011, 2013). Chloride and sodium are the most common ions found in groundwater, particularly in coastal areas. In general, high levels of chloride and sodium ions in groundwater can considerably inhibit the growth of crops. Moreover, high levels of chloride and sodium ions can affect the taste of the water, making it unsuitable for drinking. Manganese and iron ions are commonly found in groundwater but if water in which concentrations are high is used for agriculture or aquaculture, it may inhibit the growth of cultivated plants or fish. In Taiwan, the residents of the Pingtung Plain are unusual, in that only approximately 45.8 % use tap water (while the average tap water coverage in Taiwan is 92.93 %). In that area, a substantial amount of groundwater, which is relatively abundant and inexpensive, is used as a source of water to meet drinking, agriculture and aquaculture requirements. However, overexploitation has caused significant reductions in groundwater levels, severe seawater intrusion, and subsidence in the Pingtung Plain over the past five decades (Ting et al. 1998). A long-term groundwater quality survey of the Pingtung Plain conducted by the Agricultural Engineering Research Center (AERC) and financially supported
Environ Monit Assess (2016) 188: 431
by the Taiwan Water Resource Agency (WRA) indicated that several water quality parameters exceeded the water quality standards for drinking, irrigation, and aquaculture (AERC 2009, 2010, 2011, 2012). Considering the hazard to human health posed by deteriorating groundwater quality and the potential environmental impact caused by overexploitation, establishing an appropriate management plan for multi-purpose groundwater utilization in the Pingtung Plain is imperative. The major concerns associated with managing the water supply for different purposes are to ensure that the quality complies with standards set by various water utilization sectors and that the impacts on the aquifer caused by groundwater extraction is minimized. In general, there are substantial variations in the quality and hydrogeological characteristics of groundwater. However, since it is so expensive and time-consuming to effectively measure water quality and estimate hydrological parameters in the field, geostatistical techniques are increasingly being used for spatial characterization of hydrogeochemical and hydrogeological parameters such as groundwater quality and hydraulic conductivity (Goovaerts 1997). For example, Lin et al. (2001) identified the spatial characteristics of transmissivity using simulated annealing and Kriging methods. Jang and Liu (2004) applied Kriging analysis and conditional simulation to estimate the spatial variability of hydraulic conductivity. Smith et al. (1993), Oyedele et al. (1996), Halvorson et al. (1996), and Diodato and Ceccarelli (2004) developed a multi-variate indicator Kriging (MVIK) for probabilistically evaluating soil quality. Lee et al. (2008) applied MVIK to characterize the groundwater quality and proposed a zonal management strategy for multi-purpose groundwater utilization in the Lanyang Plain, Northern Taiwan. Chica-Olmo et al. (2014) used a categorical indicator Kriging method to assess the risk of groundwater nitrate pollution. This study establishes a sound zonal management strategy for multi-purpose groundwater utilization in the Pingtung Plain. The zonal management strategy is achieved by simultaneously evaluating the spatial variability of groundwater quality and the impact on the aquifer, which measures the potential drawdown caused by groundwater extraction and is defined as the ratio of actual groundwater utilization to the transmissivity. The impact on the aquifer impact is used to measure the potential negative effect induced by groundwater extraction. First, a geostatistical approach is used to spatially estimate the water quality of each hydrochemical
Environ Monit Assess (2016) 188: 431
parameter, and safe zones for different utilization purposes are evaluated on the basis of water quality criteria for drinking, irrigation, and aquaculture. The impact on the aquifer is then spatially determined. A zonal management plan for multi-purpose groundwater utilization is established by integrating the results of the determination of safe zones for different purposes and the impact on aquifer induced by pumping.
Study area The Pingtung Plain is situated in Southwest Taiwan and covers an area of 1210 km2 (Fig. 1). It includes parts of Pingtung County and Kaohsiung City and is rectangular in shape. It is bounded by foothills and river valleys to Fig. 1 Map of the study area
Page 3 of 17 431
the north, the Fengshan Fault to the west, the Taiwan Strait to the south, and the Chaozhou Fault to the east. The Kaoping River is the largest river in the area and forms the largest drainage area in Taiwan. There are two other smaller rivers on the Pingtung Plain, the Tungkang River, and the Linbian River. The unconsolidated sediments of the Late Pleistocene and the Holocene that fill this basin contain abundant groundwater. Most of the sediments consist of coastal and estuarine sand and mud, with abundant shallow marine and lagoon shells and foraminifers. This plain is classified into a proximal-fan and a distal-fan. The investigation of the subsurface geology and hydrogeology were executed from 1995 to 1998. Subsurface hydrogeological investigation were completed to a depth of approximately 250 m. The plain deposits are
431 Page 4 of 17
divided into eight overlapping sequences, including four marine sequences and four nonmarine sequences, in the distal-fan (Taiwan CGS 2002). Nonmarine sequences with coarse sediment are considered to be aquifers, whereas marine sequences with fine sediment are regarded as aquitards. Aquitards are located mainly in the distal-fan but not in the proximal-fan. There are four usable aquifers, labeled aquifer 1, aquifer 2, aquifer 3, and aquifer 4, from top to bottom, at depths of 0–70, 40– 130, 90–180, and 160–250 m, respectively (Fig. 2). The principal source of freshwater in the plain is from annually renewable groundwater from natural rain infiltration and is supplied to wells by a principal, ancient Quaternary reservoir. The proximal-fan and the river valleys at the eastern and northern boundaries are a major region for aquifer recharging. In the Pingtung Plain, the period of maximum precipitation is from May to September (an average of 2493 mm per year), followed by considerably lower precipitation from October to December (winter) and January to April (spring). The accentuated seasonality of the precipitation has resulted in a reliance on irrigation involving both legal and illegal extractions of surface and groundwater. Renewable groundwater from natural rain infiltration is the principal source of freshwater which is supplied to wells from a principal ancient quaternary reservoir.
Environ Monit Assess (2016) 188: 431
The Pingtung Plain is a highly productive agricultural area in Southwestern Taiwan mainly used for cultivation of crops and aquaculture. Approximately 45.7 % of the plain is used for agriculture and 5.1 % for fishponds. Agricultural activities have continued to intensify over the last decade. The crop patterns are shown in Fig. 1. In the dry months and years, large amounts of groundwater are extracted to meet the water resource requirements for farmlands, fishponds, and households which has led to an increase in the salinity of the groundwater, a reduction in the pollution diluting capability of the surface water, and an increase in the occurrence of severe land subsidence and seawater intrusion (Ting et al. 1998).
Materials and methods General framework A general framework for multi-purpose groundwater management in the Pingtung Plain that considers both the spatial variability of water quality and the impact on aquifers is presented. First, the spatial distributions of the hydrochemical parameters are calculated using the Kriging geostatistic approach. Subsequently, the safe zones in terms of the water quality parameters for various utilization sectors are determined on the basis of standards for drinking, irrigation, and aquaculture. Suitable zones for the impact on aquifers are then spatially evaluated. Finally, safe zones for various utilization purposes and suitable zones for the impact on aquifers are integrated to delineate zonal management for multipurpose groundwater utilization. Groundwater samples
Fig. 2 Conceptual hydrogeological profile of the Pingtung Plain
A dense network of observation wells was established between 1995 and 1998 which the WRA used to characterize the hydrogeology of the Pingtung Plain and observe the long-term groundwater level. With financial support from the Taiwan WRA, the AERC has been conducting a long-term groundwater quality survey, including 31 items, since 2009 (AERC 2009, 2010, 2011, 2012). Dissolved oxygen, temperature, oxidation-reduction potential, pH, and electrical conductivity are measured in situ and the other items analyzed in the laboratory with standard analysis methods.
1.431 32.55 0.016 0.007 0.354 1.640 79.850 0.493 1.623 0.001 0.008 0.009 Number of wells sampled is 132
430.00 75th percentile
0.007
0.337
47.633
9.026
0.625 7.20 0.010 0.007 0.169 0.388 46.900 0.040 0.185 0.001 0.007 0.007 55.00 50th percentile
0.001
3.70
19,100.0 4.760
0.008 0.003
0.235 15.400
0.024 0.093
119.00 3550.0
11.625 0.010
55.80 12.500
0.035 0.001
0.004 0.091
0.006 0.007
0.016 0.544
0.001 10.00
890,000.00
25th percentile
1.40 0.000 0.000 0.001 0.004 0.30 0.010 0.001 0.000 0.000 0.001
1327.95
4455.44 0.487
0.098 0.008
0.020 1.472
0.444 2.836
11.393 514.736
173.720 2.073
7.174 2.219
1.261 0.001
Zn Cu Mn Fe SO42− NH4+-N NO3−-N Cd
0.001 0.013
0.010 0.008
0.002 0.065
0.018
0.000
Maximum
where c0 is the nugget effect, c is the sill, and a is the range. These models provide information about the spatial structure and input parameters for the Kriging
10.00
ð4Þ
Minimum
" #) 3h 2 γ ðhÞ ¼ c0 þ c 1−exp − f or a Gaussian model; a
11,118.11
(
79,938.83
ð3Þ
Standard deviation
3h γ ðhÞ ¼ c0 þ c 1−exp − f or an exponential model; a
Cr
ð2Þ
Pb
f or a spherical model;
As
h ≤a
E. coli
" 3 # h h −0:5 c0 þ c 1:5 γ ðhÞ ¼ a a > : c0 þ c h > a
Contaminants
8 > <
Table 1 Descriptive statistics of chemicals (mg/L) and E. coli (CFU/100) of the groundwater
where h denotes the lag, Z(xi) is the value of the regional variable of interest at location xi, and Z(xi + h) is the value of the regional variable of interest at location xi + h, N(h) is the number of pairs of sampling points separated by h. In practices, the probability of the distance between the sampling pairs being exact is low, and thus, h is represented by a distance interval. A semivariogram plot is obtained by calculating the values of the semivariogram at different lags. The semivariogram of the sampling data is then fitted to a theoretical model of γ(h) such as the spherical, exponential or Gaussian model. These theoretical models are mathematically written as follows (Goovaerts 1997).
Cl−
A geostatistical approach based on the regionalized variable theory is used, which states that the variables in an area exhibit both random and spatial structure (Journel and Huijbregts 1978). In general, the assumption of theoretical second-order stationarity is used in the geostatistical approach. Geostatistics provides a variogram of data with a statistical framework which can be used to quantify the spatial variability of random variables between two locations. The semivariogram, γ(h), is defined as follows: 9 8 N ðhÞ = 1
Average
SAR
Geostatistical approach
0.117
Page 5 of 17 431 3.771
Environ Monit Assess (2016) 188: 431
431 Page 6 of 17
Environ Monit Assess (2016) 188: 431
interpolation. Kriging is regarded as an optimal spatial interpolation method in which the values of the random field at an unsampling location x0 are estimated on the basis of the given measured values in a linear form
Z*ðx0 Þ ¼
N X
λi0 Z ðxi Þ;
ð5Þ
i¼1
where Z ∗ (x0) is the value to be estimated at x0; Z(xi) denotes the given measured values at xi; N is the total number of given measured values used for estimation; and λi0 is Kriging weight for Z(xi) to estimate Z ∗ (x0). Evaluation of aquifer impact The impact on the aquifer is generally evaluated by the drawdown caused by groundwater pumping. Physicallybased mathematical models can be effective tools for evaluating the drawdown induced by pumping. There are two main types of models: analytical and numerical. Analytical models are typically available for relatively simple geometries and processes. Despite the limitations, analytical models are popular because they require few input parameters and can be run quickly and efficiently. Analytical models always serve as screen-level tools. Numerical models can provide much more detailed simulation. The major limitations of the numerical models are that level of the hydrological characterization needed to fully parameterize the models is not always available.
Therefore, while the advanced numerical model can be used to simulate complex processes, they are always subject to uncertainty. The numerical models required substantial computation efforts to set up simulation. For evaluating the drawdown induced by pumping, we can apply a pumping rate at a location of simulated domain. Then, we can evaluate the drawdown due to pumping using a numerical simulation. However, it is noted that the radius of influence induced by the applied pumping at a well is finite. When we use a numerical model to evaluate drawdown response to a groundwater withdrawal at a fixed location, we will find that the variations of the drawdowns at the most nodes in the simulated domain are insignificant except for the nodes adjacent to the node that the pumping is applied. Therefore, it is not certainly required to build a regional-scale groundwater flow model for evaluating drawdown response to a groundwater withdrawal at a fixed location. Alternatively, we can perform the local-scale evaluation. Several analytical models are available for evaluating the drawdown response to a groundwater withdrawal in local scale. The most well-known analytical model is transient-state model presented by Theis (1935). and the steady-state solution developed by Theim (1906). From the mathematical expressions of the transient- and steady-state analytical solutions, we can find that the drawdown is proportional to the ratio of actual groundwater utilization to the transmissivity regardless of the dynamic or static approach.
Table 2 Water quality standards (chemicals (mg/L) and E. coli (CFU/100)) and the numbers of wells in which contaminants exceed the water quality standards of drinking, irrigation, and aquaculture Drinking water quality standard As
Pb
Cr
Cd
NO3−-N
NH4+-N
SO42−
Standard
0.01
0.01
0.05
0.005
10
0.1
Numberb
26
1
3
0
2
53
Contaminants a
Fe
Mn
Cu
Zn
Cl−
250
0.3
0.05
1
5
250
10
73
92
0
0
16
Zn
SAR
Aquaculture water quality standard E. coli
As
Pb
Cd
NH4+-N
Mn
Cu
Zn
Standard
10,000
0.05
0.1
0.01
0.3
0.05
0.03
0.5
Numberb
12
8
0
0
38
92
2
4
Contaminants a
Irrigation water quality standard As
Pb
Cr
Cd
SO42−
Fe
Mn
Cu
Standard
0.05
0.1
0.1
0.01
200.0
5.0
0.2
0.2
2.0
6.0
Numberb
8
0
0
0
11
12
56
1
2
16
Contaminants a
a
The water quality standards for drinking, irrigation, and aquaculture in Taiwan are available at http://w3.epa.gov.tw/epalaw/docfile/090040. doc, http://www.coa.gov.tw/files/web_articles_files/8655/1087.doc, and http://law.epa.gov.tw/zh-tw/laws/309417667.html
b
Numbers of well contaminant concentration exceeds water quality standard
Environ Monit Assess (2016) 188: 431
Page 7 of 17 431
For convenience, the steady-state Thiem solution is written herein as follows (Fetter 1994): s¼
Q r ln ; 2πT R
ð6Þ
where s is the drawdown [L]; Q is the extraction rate [L3/T]; T is the transmissivity [L2/T]; r is the radial distance from the center of a single extraction well [L]; and R is the radius of influence. It is implied that the drawdown is proportional to the ratio of actual groundwater utilization to the transmissivity. Thus, using QT as a representative index that measures the aquifer impact caused by groundwater
extraction is quite reasonable. Based on the data from the Taiwan Sugar Company (TSC), the spatial distribution of T is computed using the Blogarithmic Kriging geostatistical approach^ in which the kriging estimate is performed after the natural logarithmic transformation being applied to the data (Rendu 1987) A threshold is required to define the zones unsuitable for pumping based on QT. However, this threshold depends on political, economic, and social policies and is inevitably subjective. This study defines the low, medium, and high aquifer impact as <5, 5–10, and >10 m, respectively.
Fig. 3 Safe zones for drinking water. The safe zones are in gray color
F1
F2
F3
F4
431 Page 8 of 17
Results and discussion Determining water quality safe zones based on the standards for drinking, irrigation, and aquaculture Descriptive statistical analysis of data collected from the Taiwan Water Resources Agency (N = 132) from 2009 to 2013 in the Pingtung Plain is performed using SPSS. The descriptive statistics, including the maximum, minimum, mean, and standard deviation, for each hydrochemical parameter are summarized in Table 1. Table 2 summarizes the information on the number of observation wells in which different hydrochemical parameters exceed the water quality standards for drinking, irrigation, and aquaculture.
Environ Monit Assess (2016) 188: 431
Before calculating a semi-variogram, a histogram for each hydrochemical parameter is analyzed. The histograms show that the hydrochemical parameters can be more satisfactorily fitted with a lognormal distribution than a normal distribution. Accordingly, the logarithms of the observed hydrochemical parameters are included in the Kriging geostatistic estimation process. The calculated semi-variograms for the logarithms of all parameters are fitted to different theoretical models. The Gaussian models are selected as having the best fit. After determining the theoretical semi-variograms, the spatial distribution of each hydrochemical parameters is estimated by using exponents of the values, which are calculated using Eq. (5). Each aquifer is discretized into a grid of 34 × 72 cells with a spacing of 1000 m.
Fig. 4 Safe zones for irrigation water. The safe zones are in gray color
F1
F2
F3
F4
Environ Monit Assess (2016) 188: 431
Subsequently, the safe zones in all aquifers for different utilization sectors are evaluated by comparing the estimated hydrochemical parameters with the water quality standards for drinking, irrigation, and aquaculture. The safe zone for different utilization purposes is defined as when the estimated values of all the hydrochemical parameters do not exceed the thresholds of the specified water quality standards in a fixed cell. Figures 3, 4, and 5 show the safe zones in each aquifer for drinking, irrigation, and aquaculture purposes, respectively. The safe zones are in gray color. In aquifer 1, the safe zones for drinking water are primarily located in the northern, northeastern, and eastern parts of the Pingtung Plain. For aquifer 2, the safe zones for
Page 9 of 17 431
drinking water cover most areas of the northern Pingtung Plain and a small portion of the eastern Pingtung Plain. For aquifer 3, most of the northern areas are recommended as safe zones for drinking. For aquifer 4, quite a small portion of the northern Pingtung Plain is considered to have water safe for drinking. Thus, aquifer 2 covers the largest area for drinking water utilization, followed by aquifers 3, 1, and 4, in that order. The spatial distribution of the safe zones for irrigation purposes is quite similar to that for drinking, except for aquifers 2 and 4 in the northern Pingtung Plain. The water in aquifer 2, in the northeastern areas of the Pingtung Plain, is not acceptable for irrigation purposes but is suitable for drinking.
Fig. 5 Safe zones for aquaculture water. The safe zones are in gray color
F1
F2
F3
F4
431 Page 10 of 17
Conversely, a part of the northeastern areas is safer for irrigation than for drinking. In comparison with drinking and irrigation, the zones where the water is considered safe for aquaculture cover a large area of the Pingtung Plain in all aquifers. The unsafe zones include only local areas in the western Pingtung Plain for aquifers 1, 2, and 3. The safe zones in aquifer 4 are primarily situated in the northern regions. Basically, groundwater in the southwestern coastal region is not suitable for drinking, irrigation, or aquaculture because the water quality does not meet any of the standards.
Determining suitable zones based on the impact on the aquifer caused by the pumping of groundwater In addition to meeting the water quality standards for various groundwater utilization sectors, the appropriate management of groundwater resources requires that aquifers do not suffer a substantial negative effect (impact on the aquifer) caused by groundwater resource development. In this study, the impact on the aquifer is evaluated by the drawdown caused by groundwater pumping and QT which is treated as a representative index. It is noted that the Q/T index is not used to obtain an absolute measure of the impact on the aquifer by pumping but rather functions as a reference criterion to be used in decision-making. The current status of the actual regional groundwater utilization rate (Q) in the Pingtung Plain is unknown; therefore, we use the data for groundwater rights approved by the government to represent the regional utilization of groundwater resources. Water resource users, such as the irrigation associations, Taiwan Water Corporation, and agricultural and aquaculture corporations must apply for groundwater rights based on Taiwan’s legal requirements. The data for groundwater right applications are published for each township (Table 3). In this study, the average groundwater use of each township is assigned to individual cells. The spatial distribution of the aquifer transmissivity (T) used in this study is obtained from the Taiwan Sugar Company (1997) with the help of the logarithmic Kriging estimate. Figure 6 shows the spatial distribution of aquifer spatial distribution of the aquifer transmissivity (T). It is noted aquifer heterogeneity in individual cells may have a dominant influence on the resulting distribution of the pollutant, including their
Environ Monit Assess (2016) 188: 431 Table 3 Groundwater right applications (m3/day) of townships in the Pingtung Plain Township
Household (m3/day)
Agriculture (m3/day)
Meinong
1.926 × 102
1.706 × 103
3
Dashu
2.346 × 10
8.795 × 10
Daliao
4.180 × 102
6.202 × 102
Linyuan
0
0
Qishan
4.516 × 103
1.054 × 103
Pingtung
1.487 × 103
9.692 × 102
2
2.957 × 10
3.028 × 103
Donggang
0
0
Wandan
8.236 × 10−1
3.643 × 103
Changzhi
3.034 × 102
3.341 × 103
Jiuru
0
1.741 × 103
Ligang
4.282 × 10
2.434 × 103
Yanpu
3.736 × 10
4.376 × 103
Chaozhou
Gaoshu
2
1.319 × 10
2.866 × 103
Neipu
3.636 × 102
2.601 × 103
Xinpi
5.583 × 10
4.375 × 103
Fangliao
0
1.697 × 103
Xinyuan
0
1.982 × 103
Kanding
0
1.267 × 103
Linbian
5.006 × 102
2.146 × 103
Nanzhou
0
2.885 × 104
Jiadong
0
1.942 × 103
Wanluan
6.986 × 10
2.226 × 103
Zhutian
−1
1.481 × 10
1.405 × 103
Linluo
4.937 × 10−1
1.081 × 103
detected concentrations at the well (Pedretti et al. 2013; Pedretti et al. 2014; Molinari et al. 2015) In this study, two thresholds (Q/T = 5 m and Q/T = 10 m) are considered to classify the aquifer impact by pumping. The selection of the thresholds that define the suitable zones is subjective and will result in different area percentage of the classifications. The area percentages for three classifications are summarized in Table 4. From the physical point of view, the greater the Q/T index, the higher the aquifer impact. Since the transmissivity is defined as that the rate at which water is transmitted through a unit width of aquifer under a unit hydraulic gradient. If the Q/T value is larger, this means that the withdrawal rate may exceed the capacity of water supply of the aquifer. Figure 7 depicts the spatial distribution of the impact on the aquifer. In aquifers 1 and 2, the impact is
Environ Monit Assess (2016) 188: 431
Page 11 of 17 431
Fig. 6 Spatial distribution of the aquifer transmissivity
relatively higher in small areas of the northern and southeastern plain. In aquifer 3, the aquifer impact is relatively higher in smaller regions of the southern plain. In aquifer 4, the southeastern coastal areas have a high aquifer impact.
F1
F2
F3
F4
Zonal management of multi-purpose utilization based on water quality and aquifer impact A zonal management plan for various groundwater utilization sectors is delineated after integrating the results of
Table 4 Area percentage (%) of the three classifications for aquifer impact (Q/T) Q/T
F1 (%)
F2 (%)
F3 (%)
F4 (%)
F1+F2+F3+F4 (%)
<5 m
76.6
75.7
67.4
44.8
66.1
5∼10 m
7.4
11.4
19.0
31.3
17.3
>10 m
15
13.9
13.6
24.9
16.6
431 Page 12 of 17
Environ Monit Assess (2016) 188: 431
Fig. 7 Maps of the aquifer impact evaluated using the ratio of actual water utilization to aquifer transmissivity. The aquifer impact index is classified as <5, 5–10, and >10 m
determining the safe zones for groundwater quality and suitable zone for the impact on the aquifer. Figure 8 presents the first zonal management plan based on the safe zones for different groundwater utilization purposes and a low impact on the aquifer. Figure 9 depicts the second zonal management plan based on the safe zones for different groundwater utilization purposes and a medium impact on the aquifer. As can be seen in Figs. 8 and 9, the Pingtung Plain is divided into single-purpose (drinking (D), irrigation (I), or aquaculture (A), dualpurpose (D+I or D+A or I+A), and three-purpose (D+ I+A) areas. The aquifers safe for various purposes are mostly located in the proximal-fan and a few distal-fan areas with an impact on the aquifer of 5 m. However, given an impact on the aquifer of 5–10 m, the aquifers
F1
F2
F3
F4
safe for various purposes are distributed in the northern and southern regions (aquifers 1, 2, and 3), and northern and central regions (aquifer 4). The area percentage of the seven classifications for the first and second zonal management plans are tabulated in Table 5. The area percentage is defined as the ratio of the area of each classification to the total area of the Pingtung Plain. For the first zonal management plan, the area percentages for D, I, A, I+A, and D+I+A are 0–12.7, 14.7–33.7, 0–5.0, 0–7.8, and 5.6–25.3 %, respectively. For the second zonal management plan, the area percentages for I, A, I+A, and D+I+A are 2.9–9.9, 0–1.5, 0.0–0.8, and 1.5–7.3 %, respectively. A reduction in the amount of groundwater extracted is suggested for regions that are safe for a specific utilization purpose but with a medium impact on the
Environ Monit Assess (2016) 188: 431
Page 13 of 17 431
Fig. 8 First zonal management plan based on the safe zones for different groundwater utilization purposes and a low impact on aquifers
aquifer. The shortfall can be replaced with a supply of treated surface water. For areas deemed unsafe for groundwater utilization for any purpose, an alternative water supply system should be considered to avoid the utilization of groundwater unsafe for drinking, irrigation, or aquaculture. In addition, the aforementioned multi-purpose maps are useful for the local government to plan the safe use and sustainable management of groundwater resources. Comparison between our study and recent researches Recently, researches have been conducted to reach the goal of sustainable and reasonable groundwater utilization in Pingtung Plain. The researches integrated
F1
F2
F3
F4
the spatial variability of groundwater quality and quantity for improving groundwater utilization in Pingtung Plain. Jang et al. (2012, 2013) considered the spatial variability of groundwater quality and aquifer transmissivity to establish sustainable groundwater utilization plan for irrigation and aquaculture purposes, respectively. It is noted that the aquifer transmissivity only represents the capacity at which water is transmitted in the aquifer rather than the aquifer impact. In view that the pumping-induced drawdown can much more reflect the aquifer impact, Jang et al. (2016) combined the spatial variability of groundwater quality parameters and a groundwater flow model simulation for spatially establishing utilization strategies for groundwater and surface water in the Pingtung Plain.
431 Page 14 of 17
Environ Monit Assess (2016) 188: 431
Fig. 9 Secondary zonal management plan based on the safe zones for different groundwater utilization purposes and a medium impact on aquifers
The aforementioned studies all used the multivariate indicator kriging (MVIK) method to estimate occurrence probability. A greater occurrence probability represents that groundwater is more safe for drinking, irrigation, or aquaculture. However, the safe zones were determined by setting the occurrence probability being greater than the threshold. The selection of thresholds is inevitably subjective and significantly affects the extent of the safe zones. Different from previous studies, safe zones is determined by the criterions of groundwater quality for drinking, irrigation, and aquaculture in current study. The uncertainty of safe zones due to selection of threshold of occurrence probability can be excluded. We have compared the safe zones for water quality obtained from our study with those obtained from Jang
F1
F2
F3
F4
et al. (2016). We find that the spatial pattern of two studies is similar, whereas the extent of safe zones in our study is different from those in Jang et al. (2016). The difference in the extent of safe zones may be attributed to the selection of threshold of the occurrence probabilities. The numerical groundwater flow simulation highly relies upon a sufficient hydrological characterization, which is required to fully parameterize the model and easily suffer from uncertainty. Moreover, extensive computation effort is required to set up a simulation. Thus, using Q/T as a quick tool for measuring the aquifer impact caused by groundwater extraction is quite reasonable in screening-level assessment. However, the simple use of the Q/T has its limitation. Especially, it is not suitable for the evaluation of the aquifer impact in a highly
The area percentage is defined as the ratio of the area of each classification to the total area of the Pingtung Plain. Seven classifications are as follows: single-purpose drinking (D), irrigation (I), aquaculture (A), dual-purpose (D+I or D+A or I+A), and three-purpose (D+I+A)
14.1
17.2
1.5
25.2
0.0
5.0 9.5 14.7
33.7 0.0
0.0 0.0
0.0 0.0
0.0 0.0
0.0 0.0
0.5 0.0
0.0 0.0
0.0 0.0
0.0
5.6 F4
7.3 25.3 F3
2.5
9.9
0.0
58.9
9.83
4.69 49.7 0.0
0.0
2.9
0.0
28.7
21.0 0.0
0.0 12.7
0.0 0.0
0.0 0.0
0.0 0.2
0.8 1.6
7.8 0.0
0.0 0.0
0.0
19.4
2.6
21.0
F2
1.5
6.4
0.0
62.4
Page 15 of 17 431
F1
First zone Second zone First zone First zone First zone
Second zone
First zone
Second zone
First zone
Second zone
First zone
Second zone
First zone
Second zone
I (%) D (%) D+A (%) I+A (%) D+I (%) Aquifer D+I+A (%)
Table 5 Area percentage (%) of the seven classifications for groundwater use for first and second zonal management plans
Second zone
∑ (%) A (%)
Second zone
Environ Monit Assess (2016) 188: 431
heterogeneous aquifer. Recognizing the practical limitations of Q/T for dealing the highly heterogeneous aquifer, it is suggested that we can use more detailed local-scale numerical groundwater flow model that consider the aquifer heterogeneity for systematic evaluation of aquifer impact by pumping after determining the suitable zones determined by Q/T. During this stage, it is imperative to collect much more local-scale data on the heterogeneity of hydrological conditions and parameters so that simulation results can be accepted with confidence.
Conclusions Residents of the Pingtung Plain in Southwestern Taiwan use a substantial amount of groundwater for drinking, irrigation, and aquaculture, but land subsidence induced by overexploitation has become a long-term problem. Water quality monitoring has shown that parts of the aquifers have also been polluted by hazardous contaminants. To avoid the use of contaminated groundwater for drinking, irrigation, or aquaculture and maintain sustainability, a sound management plan for multipurpose groundwater utilization in the Pingtung Plain is imperative. The major concerns are to ensure that the water quality complies with the standards of the various water utilization sectors and that the aquifer impact caused by extraction is minimized. The water quality, hydrogeological characteristics, and actual groundwater utilization play crucial roles in multi-purpose groundwater utilization. Safe use and sustainable management of groundwater resources involves great challenges because of the spatial variability of water quality, hydrogeological conditions, and groundwater utilization. This study integrates the results of determining safe zones for water quality and suitable zones for the impact on the aquifer to propose the zonal management of multi-purpose use of groundwater in the Pingtung Plain. Based on the standards of different uses, the safe zones for various utilization purposes are spatially delineated using a Kriging geostatistical estimation method. The impact on aquifers is spatially determined using the ratio of the actual groundwater extraction rate to transmissivity. Based on the analytical results of the water quality and the impact on aquifers, a zonal management strategy is established. The results of this study can provide government administrators with a zonal management strategy for multi-purpose groundwater utilization in the Pingtung Plain.
431 Page 16 of 17 Acknowledgments The authors would like to thank the Ministry of Science and Technology of the Republic of China for financially supporting this work under Contract No. MOST 1032116-M-242-001.
References Chica-Olmo, M., Luque-Espinar, J. A., Rodriguez-Galiani, V., Pardo-Igúzquiza, E., & Chica-Rivas, L. (2014). Categorical indicator Kriging for assessing the risk of groundwater nitrate pollution: the case of Vega de Granada aquifer (SE spain). Science of the Total Environment, 470–471, 229–239. Diodato, N., & Ceccarelli, M. (2004). Multivariate indicator kriging approach using a GIS to classify soil degradation for Mediterranean agricultural lands. Ecological Indicators, 4(3), 177–187. AERC (Agriculture Engineering Research Center) (2009). Survey, analysis and assessment of groundwater quality in Taiwan areas in 2009. Water Resources Agency, Ministry of Economic Affairs, Executive Yuan, Taiwan, pp. 100–102. AERC (Agriculture Engineering Research Center) (2010). Survey, analysis and assessment of groundwater quality of Taiwan in 2010. Water Resources Agency, Ministry of Economic Affairs, Executive Yuan, Taiwan, pp. 151–158. AERC (Agriculture Engineering Research Center) (2011). Survey, analysis and assessment of groundwater quality of Taiwan in 2011. Water Resources Agency, Ministry of Economic Affairs, Executive Yuan, Taiwan, pp. 191–203. AERC (Agriculture Engineering Research Center) (2012). Survey, analysis and assessment of groundwater quality of Taiwan in 2012 – 2014 (1/3). Water Resources Agency, Ministry of Economic Affairs, Executive Yuan, Taiwan, pp. 85–101. Chen, C. J., & Wang, C. J. (1990). Ecological correlation between arsenic level in well water and age-adjusted mortality from malignant neoplasms. Cancer Research, 50(17), 5470–5474. Chen, C. J., Chuang, Y. C., Lin, T. M., & Wu, H. Y. (1985). Malignant neoplasma among residents of a blackfoot disease-endemic area in Taiwan: high arsenic artesian well water and cancers. Cancer Research, 45(11 Part 2), 5895–5899. Fetter, C. W. (1994). Applied hydrogeology (3rd ed.). New York: Macmillan College Publishing Company. Goovaerts, P. (1997). Geostatistics for natural resources evaluation (pp. 259–368). New York: Oxford University Press. Halvorson, J. J., Smith, J. L., & Papendick, R. I. (1996). Integration of multiple soil parameters to evaluate soil quality: a field example. Biology and Fertility of Soils, 21(3), 207–214. Hsueh, Y. M., Wu, W. L., Huang, Y. L., Chiou, H. Y., Tseng, C. H., & Chen, C. J. (1998). Low serum carotene level and increased risk of ischemic heart disease related to long-term arsenic exposure. Atherosclerosis, 141(2), 249–257. Jang, C. S., & Liu, C. W. (2004). Geostatistical analysis and conditional simulation for estimating the spatial variability of hydraulic conductivity in the Choushui River alluvial fan, Taiwan. Hydrological Processes, 18, 1333–1350. Jang, C. S., Chen, S. K., & Kuo, Y. M. (2012). Establishing an irrigation management plan of sustainable groundwater based on spatial variability of water quality and quantity. Journal of Hydrology, 414–415, 201–210.
Environ Monit Assess (2016) 188: 431 Jang, C. S., Liang, C. P., & Wang, S. W. (2013). Integrating the spatial variability of water quality and quantity to probabilistically assess groundwater sustainability for use in aquaculture. Stochastic Environmental Research and Risk Assessment, 27(6), 1281–1291. Jang, C. S., Chen, C. F., Liang, C. P., & Chen, J. S. (2016). Combining groundwater quality analysis and a numerical flow simulation for spatially establishing utilization strategies for groundwater and surface water in the Pingtung Plain. Journal of Hydrology, 533, 541–556. Journel, A. G., & Huijbregts, C. J. (1978). Mining geostatistics. San Diego: Academic. Lee, J. J., Liu, C. W., Jang, C. S., & Liang, C. P. (2008). Zonal management of multi-purpose use of water from arsenicaffected aquifers by using a multi-variable indicator kriging. Journal of Hydrology, 35, 260–273. Liang, C. P., Jang, C. S., Liu, C. W., Lin, K. H., & Lin, M. C. (2010). An integrated GIS-based approach in assessing carcinogenic risks via food-chain exposure in arsenic-affected groundwater areas. Environmental Toxicology, 25, 113–123. Liang, C. P., Liu, C. W., Jang, C. S., Wang, S. W., & Lee, J. J. (2011). Assessing and managing the health risk due to ingestion of inorganic arsenic from fish and shellfish farmed in blackfoot disease areas for general Taiwanese. Journal of Hazardous Materials, 186, 622–628. Liang, C. P., Jang, C. S., Chen, J. S., Wang, S. W., Lee, J. J., & Liu, C. W. (2013). Probabilistic health risk assessment for ingestion of seafood farmed in arsenic contaminated groundwater in Taiwan. Environmental Geochemistry and Health, 35(4), 455–494. Lin, Y. P., Tan, Y. C., & Rouhani, S. (2001). Identifying spatial characteristics of transmissvity using simulated annealing and kriging methods. Environmental Geology, 41, 200–208. Molinari, A., Pedretti, D., & Fallico, C. (2015). Analysis of convergent flow tracer tests in a heterogeneous sandy box with connected gravel channels. Water Resources Research, 51, 5640–5657. doi:10.1002/2014WR016216. Oyedele, D. J., Amusan, A. A., & Obi, A. O. (1996). The use of multiple-variable indicator kriging technique for assessment of the suitability of an acid soil for maize. Tropical Agriculture, 73(4), 259–263. Pedretti, D., Fernàndez-Garcia, D., Bolster, D., & Sanchez-Vila, X. (2013). On the formation of breakthrough curves tailing during convergent flow tracer tests in three-dimensional heterogeneous aquifers. Water Resources Research, 49, 4157– 4173. doi:10.1002/wrcr.20330. Pedretti, D., Fernàndez-Garcia, D., Sanchez-Vila, X., Bolster, D., & Benson, D. A. (2014). Apparent directional mass-transfer capacity coefficients in three-dimensional anisotropic heterogeneous aquifers under radial convergent transport. Water Resources Research, 50, 1205–1224. doi:10.1002/2013WR014578. Rendu, J. M. (1987). Kriging, logarithmic kriging and conditional expectation: comparison of theory with actual results. In T. J. O’Neil (Ed.), Applications of computers and operations research in mineral industry (pp. 199–212). New York: Soc. Min. Eng. Smith, J. L., Halvorson, J. J., & Papendick, R. I. (1993). Using multiple-variable indicator kriging for evaluating soil quality. Soil Science Society of America Journal, 57, 743–749. Taiwan CGS (Central Geological Survey) (2002) Hydrogeological survey report of pingtung plain, Taiwan (pp. 97–142).
Environ Monit Assess (2016) 188: 431 Taiwan: Central Geological Survey, Ministry of Economic Affairs, Executive Yuan. Taiwan Sugar Company (1997) Establishment and Operational Management of Groundwater Monitoring Network. Water Resources Bureau, Taiwan, pp2–12. Theim, G. (1906). Hydrologische (p. 56). Leipzig: Gebhardt. Theis, C. V. (1935). The lowering of the piezometer surface and the rate and discharge of a well using groundwater storage. Transactions, American Geophysical Union, 16, 519–524. Ting, C. S., Zhou, Y., de Vries, J. J., & Simmers, I. (1998). Development of a preliminary ground water flow model for water resources management in the Pingtung Plain, Taiwan. Ground Water, 35(6), 20–36.
Page 17 of 17 431 Tseng, W. P. (1977). Effects and dose-response relationships of skin cancer and blackfoot disease with arsenic. Environmental Health Perspectives, 19, 109–119. Tseng, C. H., Tai, T. Y., Chong, C. K., Tseng, C. P., Lai, M. S., Lin, B. J., et al. (2000). Long-term arsenic exposure and incidence of non-insulin-dependent diabetes mellitus: a cohort study in arseniasis-hyperendemic villages Taiwan. Environmental Health Perspectives, 108(9), 847–851. Wu, M. M., Kuo, T. L., Hwang, Y. H., & Chen, C. J. (1989). Doseresponse relation between arsenic well water and mortality from cancers and vascular diseases. American Journal of Epidemiology, 130(6), 1123–1132.
