ISSN 0097-8078, Water Resources, 2007, Vol. 34, No. 3, pp. 259–267. © Pleiades Publishing, Ltd., 2007. Published in Russian in Vodnye Resursy, 2007, published in Vodnye Resursy, 2007, Vol. 34, No. 3, pp. 281–289.

WATER RESOURCES AND THE REGIME OF WATER BODIES

Feasible Groundwater Allocation Scenarios for Land Subsidence Area of Pingtung Plain, Taiwan* Po-Yi Choua, Cheh-Shyh Tingb a Research Assistant, Groundwater Resources Evaluation and Management Research Group Departmentof Civil Engineering, National Pingtung University of Science and Technology 1, Hseuh Fu Road, Neipu Hsiang, Pingtung, Taiwan b Associate professor (Corresponding author) Department of Civil Engineering National Pingtung University of Science and Technology 1, Hseuh Fu Road, Neipu Hsiang, Pingtung, Taiwan E-mail: [email protected] Tel: +886-8-7740238 Fax: +886-8-7740409

Received December 29, 2005

Abstract—Land subsidence basically is the deprivation of water and earth resources, further inducing social and economical undesirable impact. The principal direction of land subsidence prevention is properly management of groundwater. However groundwater management should be developed on the basis of combined technical, economical, social and institutional approaches to management that reflect local conditions and can be adapted and evolved. Therefore, the aim of this paper is to make land subsidence prevention strategies for government to refer. Before year 1969, agriculture was the main land utilization business in Pingtung Plain. Due to intensive development of fish breeding after 1970's, the aquaculture area along the estuary region of Pingtung Plain have been dramatically increased. Groundwater thus became the main fresh water resource for aquacultural water diluting and flushing because of the insufficiency in surface water supply. The uncontrolled development of groundwater resources has led to undesirable effects, especially in the south where aquaculture is concentrated. These effects are land subsidence, saline water intrusion, lowering of water tables and reductions in well yields. Government stressed on the improvement of breeding technology in the past, which mainly focused on the water quality control in order to raise the culture density, however, it neglected the impact to the environment and quantity control. This paper promotes a reasonable aquacultural water consumption policy aims at finding out the most suitable breeding species considering water consumption and its reasonable breeding area under the premise that it will not depress the original profit of aquatic products trading. DOI: 10.1134/S0097807807030037 *

INTRODUCTION Land subsidence basically is the deprivation of water and earth resources, further inducing social and economical ill impact. A very principal direction of land subsidence prevention is properly management, utilization and conservation of groundwater. Groundwater management should be developed on the basis of combined technical, economical, social and institutional approaches to management that reflect local conditions and can adapt and evolve (Molden, 2001). As we know that over-pumping of groundwater, oil, gas, etc. from underground aquifer would cause groundwater table to drop. The reduction of groundwater hydraulic head causes depression of upward stress to support the gravity of clay and silt beds. Indirectly, upper aquifer particles would compress and regional subsidence would occur. Land subsidence has already been a rather general but serious problem in many Asian major cities like Shanghai, Jakarta, Bangkok, Tokyo and Osaka. In Taiwan, owing to lack of management in the past, people pumped groundwater without limit, especially * The

text was submitted by the authors in English˛

the cities at estuary region. In order to get the cheaper and constant temperature freshwater supply, aquacultural proprietors pumped groundwater with a large scale, and that was the main cause of locally land subsidence. Due to land surface along estuary region was lower than sea level, every time when a typhoon passed over, powerful stormy waves always made these areas flooded. Government has to keep raising the height of dikes and using pump to maintain the land utilization with difficulty. Land subsidence has already repeatedly increased the expenditure of local government for drainage works. Although relative organizations in Taiwan proposed many approaches on land subsidence prevention including strictly groundwater-pumping restriction, but it is difficult to implement. Comprehensively restricted groundwater utilization should not be the only way of solving problem; therefore, this study aims at making a feasible and pliable groundwater stage by stage allocation scenario, in order to against land subsidence proceeding but still can be accepted by people. The methodology of this study is expressed as the flowchart below.

259

260

PO-YI CHOU, CHEH-SHYH TING

Background Data Collection

mm 600 500

Problem Identification 400

Hypothesis Drawing-Up

300 200

Literature Review 100 0

Decision Variables Identification

Analytical Approach Introduction

Solution Assessment (Reasonable or Not?)

Yes

No

Feasible Scenarios Discussion

Conclusions and Recommendation Scheme of study methodology

BACKGROUND INFORMATION AND PROBLEM HYPOTHESIS Background Information Causes of land subsidence are complex; it could be the result of a series of natural phenomena ex. volcano eruption, earthquake, diastrophism, sea level rising, etc. But people continually over extract groundwater, oil, natural gas and mining also cause the happening of land subsidence. Land subsidence is slow and usually irreversible, it is not like land slides or typhoons, which cause obvious and direct catastrophe in a very short time. On the contrary, it is not evident at the beginning. Only after several decade people would get conscious of seawater instruction, well water becoming saline and the occurring of various derivational problems. Before year 1969, agriculture was the main land utilization business in Pingtung Plain, Taiwan. Due to

I

II III IV V VI VII VIII IX X XI XII Month

Fig. 1. Mean monthly distribution of precipitation in Pingtung T.

intensive development of fish breeding after 1970s, the aquaculture area along the estuary region of Pingtung Plain have been dramatically increased, as the Fig. 1 presents. Because the trading of aquatic products flourishes regional economy, therefore the breeding area, aquaculture yield, species and value increased dramatically during 1970 to 1990. The rainfall in Pingtung Plain is abundant but quite uneven. According to the record of Taiwan Water Conservation Administration over the period of 1981–2001, 88% of the annual precipitation falls during rainy season (May to September) and only 12% of rainfall during dry season (October to next April); there is usually less than 100 mm rain per month in the dry season, as Fig. 1 presents. Groundwater thus became the main fresh water resource for aquacultural water diluting and flushing because of the insufficiency in surface water supply. The variation of groundwater level between rainy season and dry season is also large. During rainy season, groundwater level rises due to the recharge of precipitation and surface runoff. However, groundwater drops down quickly in dry season because of groundwater pumping. Table 1 illustrates the monthly groundwater level record (auto-recording) of four groundwater observation stations along the estuary area of Pingtung Plain. And Fig. 2 represents the location of them. From Fig. 2 it can be seen that station — (Linpien) and D (Wenfeng), both are located near the land subsidence area and it can be perceived that local densely pumping behavior caused the seasonal wide difference of groundwater level. The high density of fish and shrimp stock breeding along the estuary area is incredible. It is common using one m3 of water to produce twenty kilograms aquatic WATER RESOURCES

Vol. 34

No. 3

2007

FEASIBLE GROUNDWATER ALLOCATION SCENARIOS Month

(‡)

0 –6 VIII

Month

(c)

0 –8

261

–2

IX

X XI XII

VI VII

I

–12 V

–4 II

III

–6

IV

VIII

–8 –10

VI

IX X

VII

XI XII I

II III IV

V

–12

(d)

0 a

–1 b c

(b)

0

IX X XI XII

–2 –3

d

I

VIII

II

VII

–4

III

–0.4 –5 –0.8

IV

–6 V

–1.2

IX X XI XII I VIII

–1.6 –2.0 V

VI

VI

II III

VII

IV

Fig. 2. Groundwater level variations, m, at staticus (a) Hsinuan, (b) Tungkang, (c) Linpien, and (d) Wenfeng. The square is the zone of observations, the circle is the subsidance zone.

product [8]. However, people did not have correct conception about sustainable development and environmental protection, according to local farmers' habit,

part of them not only over pumped groundwater day and night, but also transported abundant brackish water to inland to breed special aquatic species. It caused seri-

Table 1. Groundwater level record, m, at four groundwater observation stations in different months Station

Years

V

VI

VII

VIII

IX

X

XI

XII

I

II

III

IV

Hsinuan

1995–2001 –2.69

–2.21 –2.14

–1.63

–1.81

–1.94

–2.08

–2.15

–2.46

–2.54

–2.69

–2.87

Tungkang

1982–1997 –1.88

–1.73 –1.62

–1.41

–1.13

–1.13

–1.17

–1.16

–1.21

–1.28

–1.57

–1.74

Linpien

1982–2001 –9.69

–7.51 –7.17

–5.56

–4.79

–5.16

–5.54

–6.1

–6.33

–7.01

–7.97

–9.13

Wenfeng

1984–2001 –5.52

–4.78 –3.1

–1.77

–1.16

–4.09

–1.12

–1.25

–1.75

–3.05

–4.06

–4.88

WATER RESOURCES

Vol. 34

No. 3

2007

262

PO-YI CHOU, CHEH-SHYH TING Income 0.819 0.737 0.655 0.573 0.492 0.410 0.328 0.246 0.164 0.082 0 (0,2,0,0,0,1,0,1) (1,2,0,0,1,0,0,0) (0,0,1,0,0,0,1,2) (1,1,0,0,1,0,0,1) (0,0,0,0,0,1,0,3) (1,0,0,0,1,0,0,2) (1,0,0,0,0,0,0,3) Area distribution

Fig. 3. Optimal area distribution (here and in Fig. 4, 1 unit is 1000 ha) for breeding different types of aquaculture.

ously saline-water intrusion and destroyed natural ecology along the estuary area. The uncontrolled development of groundwater resources has led to undesirable effects, especially the areas where aquaculture is concentrated. Land subsidence accumulated amount has more than 3 m, saline water intrusion, lowering of water tables and reductions in well yields [5]. People stressed on the improvement of breeding technology in the past, which mainly focused on the water quality control in order to raise the breeding density; however, it neglected the impact of environmental saturation and abundantly abstract groundwater. These areas used to be famous for processing and exporting aquatic products; local people became very rich in several years by groundwater utilization. However, people must use fortune to remedy the cost of land subsidence inversely nowadays. Therefore, a reasonable aquacultural water consumption structure is needed. Problem Hypothesis A reasonable aquacultural water consumption structure aims at finding out the most suitable breeding species considering water consumption and its reasonable breeding area under the premise that it will not depress the original profit of aquatic products trading. It is a multi-objective function problem. In another saying is: how to effectively decrease aquaculture water con-

sumption without affecting regional profit development? It would be truly an imperative problem need to deliberate and the modulation of water consumption structure between each breeding species is a possible solution. Many theses have ever focused on aquacultural water analysis but the most are still in the field of water quality improving. Seldom researches in Taiwan discussed the optimization scenario of aquacultural water like [3]. Yang et al. [3] proposed the fuzzy multi-objective function to resolve relative problems in Beishrliau district, Pingtung, and applied global optimization algorithm to find out suitable aquaculture scenarios in the demonstration district. Fuzzy logic was first invented to be a representation scheme and calculus for vague notions by L. A. Zadeh in 1965. It is a multi-valued logic that allows more human-like interpretation by resolving intermediate categories among an uncertain region. In another saying is multi-objective function represents no single optimum solution existing, but in contrast, there could be many possible solutions (noninferior solutions). Because of that, considerations can therefore be made in some ideal solutions, which can achieve the requirements as closer as possible. The result of Yang et al. [3] presented the most perfect scenario of breeding species, water demand and decrement, and the ideal profit. Although theoretically it offered the fisheries authorities as a perfect arrangement of aquaculture structure, but it is practically diffiWATER RESOURCES

Vol. 34

No. 3

2007

FEASIBLE GROUNDWATER ALLOCATION SCENARIOS

cult to enforce. Land subsidence is slow and usually irreversible. As the same way, it has to advance the work gradually and continually, and then it can truly terminate groundwater over utilization and earn the identification of local people. Therefore, a proposal on another reference direction of aquacultural water modulation in order to identify the ideal and unsuitable breeding species would be discussed below. DECISION VARIABLES IDENTIFICATION Aquaculture in Taiwan can be simply classified into marine culture and inland water culture. According to a survey of the Committee of Agriculture, Fisheries Administration (COAFA) in 2000, the area of inland water culture estimated 5105 ha; it was 91.7% of total aquacultural area in Pingtung County. Inland water culture can be further classified into brackish water and fresh water. During year 2000, 41.9% of inland water culture was brackish and the other 58.1% was fresh. According to the statistics of fisheries yearbook, the staple species (total breeding area larger than 100 ha) of inland aquaculture in Pingtung are Tilapia, Eel, Milkfish, Porgy, White-spotted reef-cod, Grass shrimp, Giant freshwater prawn, White shrimp and Soft-shell turtle. Relative variables of each species like breeding area, water consumption from groundwater pumping per unit area, annual production per unit area and the average price of each species are arranged as Table 2. Data of individual species unit area groundwater consumption comes from “Field Studies and Analysis on the Reasonable Water Use of Aquaculture (II)" issued by Liang et al in 1997 and the rest of data is issued by COAFA Fishery Yearbook in 2000. The table is merged Grass shrimp and White shrimp together as on item called “Sea water shrimp”.

263

The annual total water consumption of all species (ΣAiWi) is 354638000 m3/year, and annual total profit of all species (ΣYiPi) is 3758247000 NT$/year. Here only two objectives are concerned; one is the decrease of total water consumption and the other is the increase of total profit. In order to apply fuzzy theory, it is necessary to define two membership functions to identify the requirement of each objective. A membership function maps every element of the universe of discourse X to the interval [0,1], [6], according to the fundamental equation of Yang et al. [7], two membership functions are made as below

∑ ∑

∑ ∑

Ai W i – A *i W i µ con = ------------------------------------------------------, A i W i – ( A i )W imin

∑ ∑

∑

A *i Y i P i – Ai Y i Pi µ pro = ---------------------------------------------------------------------, ( A i ) ( Y i P i ) max – Ai Y i Pi

∑

µcon = membership Function of total water consumption; µpro = membership Function of total profit; Ai = original area of species i (ha); A *i = optimum area of species i (ha); Wi = water consumption of each species i (m3/ha/yr); W imin = 3 (m /ha/yr);

minimum water consumption of species i

Yi= annual production of each species i (kg/ha/yr); Pi = price of each species i ($NT/kg)

Table 2. Breeding area, groundwater consumption, annual production, and averaged price for each aquaculture species Area A, ha, [1]

Groundwater consumption, W per year, million m3/ha, [3]

Annual production Y, kg/ha, [1]

Price P, $NT*/kg, [1]

Tilapia

115

22740

7887

35

Eel

145

108580

9828

190

Milkfish

254

149380

7161

60

Porgy

336

36880

378

155

White-spotted reef-cod

567

65400

6093

220

Grass shrimp

382

128710

4236

380

2250

88350

3303

225

101

9000

16792

165

Species i

Giant freshwater prawn Soft-shell turtle * – 1$ = 34 $NT. WATER RESOURCES

Vol. 34

No. 3

2007

264

PO-YI CHOU, CHEH-SHYH TING Income 0.022 0.717 0.676 0.614

0.187

0.020

0.553

0.110

0.492 0.430

0.108 0.131

0.022

0.369

0.352

0.185

0.099

0.031

0.054 0.264

0.097

0.307

0.273 0.296

0.020 0.043

0.246

0.031

0.219

0.184 0.123

0.350

0.012

0.262

0.177 0.185

0.029 0.131

0.061 0.089

0.140

0.208

0.194

0.515

0.348 0.461

0.271 0.294

0.217 0.240 0.427

0.175 0.163 0.373

0.513

0.459 0.436

0 (0,1,0,0,1,0,2)

(0,0,0,0,1,0,3) (0,1,0,2,0,1,0) (0,0,0,1,2,1,0) (1,2,0,1,0,0,0) (1,1,0,0,2,0,0) (1,0,0,0,2,0,1) (2,1,0,0,1,0,0)

Area distribution

Fig. 4. Optimal area distribution for breeding different species except for solf-shell turtle.

ANALYTICAL APPROACH INTRODUCTION AND THE SOLUTION ASSESSMENT When the preference of each species to both consumer and government is the same, total water consumption and total profit are influenced only by the modulation of area. It is easy to discover from Table 2 that soft-shell turtle is the most ideal species among all, because its low water consumption and high profit. The program Borland Delphi is applied to present the possible solution distribution. In the compiling chart of the outcome, x-axis represents the expectation of total water consumption decreasing potential, the ideal scenario should get as closer to the right as possible, because that means the scenario will consume much less water than original situation; y-axis represents the expectation of total profit increasing potential. The ideal scenario should get as closer to the top as possible, because that means the scenario will increase much higher profit than original situation. See Fig. 3. The distribution of possible solutions is obviously influenced by the extreme value: soft-shell turtle. According to the experience of soft-shell turtle breeder, because soft-shell turtle has high environmental adaptation so it does not need very good water quality and usually breeder does not need to change water in whole year. On the other hand, although soft-shell turtle has very high production, however, its main purpose is the exportation outside of Taiwan. Therefore, if soft-shell turtle is neglected and the rest of species are taken into

consideration, annual total water consumption becomes 353729920 m3/year, annual total profit becomes 3478408785 NT$/year and then the possible area allocation will be as presented in Fig. 4. The label on x-axis is the distribution area of seven species in the order of Tilapia, Eel, Milkfish, Porgy, White-spotted reef-cod, Seawater shrimp, and Giant freshwater prawn. The label on y-axis represents µpro and the label above each solution point represents µcon. The outcome chart above presents 45 possible area distribution ways. And the description of each solution is arranged as shown in the Table 3 below. FEASIBLE SCENARIOS DISCUSSION Outcome chart here proposes another deliberation space for strategy making. According to Table 3, if total aquacultural area maintains as 4000 ha then the most ideal species is White-spotted reef-cod and the most unsuitable breeding species is Milkfish; due to its high water consumption and relatively low profit. Under the hypothesis of homogeneous preference between each species, if policy maker takes total groundwater consumption decreasing as prior consideration, and then the best area allocation would be 1000 ha of Tilapia, 3000 ha of White-spotted reef-cod and the rest with nothing. Total annual groundwater consumption of this scenario is 218940000 m3/year; it saves 135 Mm3 water (ca. 38.1% of original water consumption) per year. And the annual total profit increases to 4297 M WATER RESOURCES

Vol. 34

No. 3

2007

FEASIBLE GROUNDWATER ALLOCATION SCENARIOS

265

Table 3. Possible area distribution, ha, for breeding different aquaculture species (the total area is 4000 ha) µcon

µpro

Tilapia

Eel

Milkfish

Porgy

0.515 0.513 0.461 0.459 0.436 0.427 0.373 0.352 0.350 0.348 0.296 0.294 0.273 0.271 0.264 0.262 0.240 0.219 0.217 0.208 0.194 0.187 0.185 0.185 0.177 0.175 0.163 0.140 0.131 0.131 0.110 0.108 0.099 0.097 0.089 0.054 0.043 0.031 0.031 0.029 0.022 0.022 0.020 0.020 0.012

0.201 0.069 0.147 0.016 0.006 0.054 0.001 0.461 0.330 0.198 0.276 0.145 0.267 0.135 0.315 0.183 0.091 0.213 0.082 0.130 0.072 0.590 0.459 0.120 0.169 0.037 0.028 0.019 0.405 0.067 0.527 0.396 0.444 0.312 0.022 0.342 0.259 0.333 0.238 0.107 0.719 0.381 0.588 0.249 0.298

1000 2000 0 1000 2000 1000 0 0 1000 2000 0 1000 10000 2000 0 1000 0 0 1000 0 2000 0 1000 1000 0 1000 0 10000 0 0 0 1000 0 1000 0 0 0 1000 0 1000 0 0 1000 1000 0

0 1000 0 1000 0 0 0 0 1000 2000 1000 2000 0 1000 0 1000 2000 0 1000 1000 0 1000 2000 0 0 1000 1000 0 2000 0 0 1000 1000 2000 0 1000 2000 0 0 1000 2000 0 3000 1000 1000

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1000 1000 0 0 0 0 0

0 0 1000 1000 0 0 1000 0 0 0 1000 1000 0 0 0 0 2000 1000 1000 1000 0 0 0 0 0 0 2000 1000 1000 1000 0 0 0 0 0 1000 1000 0 0 0 0 0 0 0 0

WATER RESOURCES

Vol. 34

No. 3

2007

White-spotted reef-cod 3000 1000 3000 1000 1000 2000 2000 4000 2000 0 2000 0 2000 0 3000 1000 0 2000 0 1000 0 3000 1000 1000 2000 0 0 0 1000 1000 3000 1000 2000 0 1000 1000 0 1000 3000 1000 1000 2000 0 0 1000

Grass shrimp 0 0 0 0 1000 0 0 0 0 0 0 0 1000 1000 0 0 0 1000 1000 0 2000 0 0 1000 0 0 1000 2000 0 1000 1000 1000 0 0 0 1000 0 2000 0 0 0 1000 0 1000 0

Giant freshwater prawn 0 0 0 0 0 1000 1000 0 0 0 0 0 0 0 1000 1000 0 0 0 1000 0 0 0 1000 2000 2000 0 0 0 1000 0 0 1000 1000 3000 0 1000 0 0 0 0 1000 0 1000 2000

266

PO-YI CHOU, CHEH-SHYH TING

Table 4. Optimal scenarios of breeding the species Species Tilapia Milkfish Eel Milkfish Eel Milkfish Milkfish Giant freshwater prawn Milkfish Giant freshwater prawn Tilapia Milkfish Tilapia Eel Milkfish Grass shrimps Current state

Area distribution, Annual water consumption, ha million m3 1000 ha 3000 ha 2000 ha 2000 ha 1000 ha 3000 ha 4000 ha 2000 ha 2000 ha 1000 ha 1000 ha 2000 ha 1000 ha 1000 ha 1000 ha 1000 ha

$NT /year. The location of this solution is at the most right hand side in Fig. 4. Table 4 represents the suggested area allocation scenario based on total area maintaining the same. But if policy maker views total profit of aquaculture more important than the other, then the best area allocation would be 2000 ha of Eel, 2000 ha of White-spotted reef-cod and the rest with nothing. Annual total profit of this area will increase to 6416 million $NT/ year. Total annual groundwater consumption of this scenario is 347960000 m3/year; it saves about 1.7% of original water consumption per year only. The location of this solution is at the most top one. In case the policy maker is not biased against which objective then the possible solution would be either the combination between 1000 ha of Eel and 3000 ha of White-spotted reef-cod or 4000 ha of overall aquaculture area all for Whitespotted reef-cod. The two scenarios can save 13.8% and 26% of original total water consumption respectively. However, if the independent character of each variable, public opinion, government’s policy and market demand structure is considered, the modification result will be different. Therefore, aquaculture area and production are individually taken into consultation. (1) From the viewpoint of breeding area, Giant freshwater prawn occupies more than half of total area, in case the government decides to narrow the breeding area of Giant freshwater prawn, it will absolutely make enforcement difficult. Therefore, if the policy maker intends to keep 200 ha of Giant freshwater prawn without modification, then the best allocation will be

Water savings, %

Annual total income, million $NT

219

38.1

4297

348

1.7

3416

305

13.8

5889

262 308

26.0 13.0

5362 4167

242

31.6

3700

325

8.2

5094

354

0

3478

another 2000 ha for White-spotted reef-cod breeding. This scenario saves 13% (46 Mm3/year) of original groundwater per year. Or to keep half of Giant freshwater prawn breeding area and 1000 ha region for Tilapia and 1000 ha region for White-spotted reef-cod, then it can save 31.6% (112 Mm3/year) of groundwater consumption in a year. (2) From the viewpoint of production, Eel, Whitespotted reef-cod and Seawater shrimp have relatively considerable profit. If government takes annual profit of species as the reference of modification, policy maker has to retain these three species for further breeding. So the most ideal scenario will be 1000 ha of breeding region for Eel, White-spotted reef-cod. Sea water shrimp and Tilapia individually. Such modification can make annual total profit reach 5094 million $NT/year, and total water consumption also decreases to 325 Mm3/year. CONCLUSIONS AND RECOMMENDATIONS The process of policy-making has to consider of present situation, future development, stakeholder acceptability, government capacity etc. Especially aquaculture is a water utilization service with high subtractability, so it needs to be allocated wisely. This modulation strategy proposes government a feasible and pliable policy making scenario, during the processing of policy making and enforcement, government can review practical condiWATER RESOURCES

Vol. 34

No. 3

2007

FEASIBLE GROUNDWATER ALLOCATION SCENARIOS

tion of the citizenry and to modify policy requirement stage by stage. Presently it doesn’t consider of the effect of breeding density, water quality control, and the characteristic of individual species. These parameters are also important and valuable for more accurately analyze the total aquacultural water requirement. So it is recommended to involve them for the further research. REFERENCES 1. Committee of Agriculture? Fisheries Administration Fishery. Taiwan: COAFA, 2000. 504 p. 2. Estimation of Ground Water Recharge in Taiwan. MOEA/WRB-890049, December 2000. Chines: EITCO, 2000. 3. Liang, J.Y., Chen, H., and Lai, K.H., Field Studies and Analysis on the Reasonable Water Use of Aquaculture (II). Taiwan: COAFA, 1997.

WATER RESOURCES

Vol. 34

No. 3

2007

267

4. Molden, D., Groundwater. Conferences Report Water Policy, Official J. of the World Water Council, 2001, vol. 3, pp. 171–173. 5. Ting, C.S., Groundwater Resources Evaluation and Management for Pingtung Plain, PhD. Thesis Vrije Universiteit, Amsterdam: Re-published by Water Resources Bureau. Taiwan: MOEA, 2000, pp. 3–32. 6. Tsoukalas, L.H. and Uhrig, R.E., Fuzzy and Neural Approaches in Engineering, Canada: John Willey & Sons, Inc., 1996, pp. 10–30. 7. Yang, T.C., Yu, P.S., and Chang, K.C., A Study on Application of Fuzzy-Objective Functions To Aquaculture Scenario Simulation in Demonstration Districts, J. Of Taiwan Water Conservancy, 2001, vol. 49, no. 3, pp. 42– 50. 8. Yu, K.H., Fisheries and ecology along Pingtung country coastal area, 2nd Pingtung Research Seminar, Symposium of Business, Ecology, Architecture vs. Native soil. Pingtung: Pingtung county goverment. 2001. 9. Zadeh, L.A., Fuzzy Sets, Information Control, 1965, no. 8, pp. 338–353.