利用衛星遙測技術研究近40年來棲地破壞對日本鰻資源量的影響

(1)

國立臺灣大學生命科學院漁業科學研究所碩士論文

Institute of Fisheries Science College of Life Science National Taiwan University

Master Thesis

利用衛星遙測技術研究近 40 年來棲地破壞對日本鰻資源量的影響

Study of the relationship between Anguilla japonica resource and habitat destruction by satellite remote sensing in the past 40 years

陳健澤 Jian-Ze Chen

指導教授：韓玉山博士 Advisor: Yu-San Han, Ph.D.

中華民國 102 年 7 月

July 2013

(2)

致謝

此論文能夠順利完成，要感謝韓玉山老師的悉心指導，不厭其煩一次又一次的指正論文上的缺失，也謝謝老師這兩年來提供許多的出差機會，讓我可以更了解鰻魚產業的現況，與產業有所接軌，這兩年真的受益良多；此外也要感謝黃祥麟老師在衛星遙測技術上的指導與意見，讓這本論文可以順利完成。

承蒙國立臺灣海洋大學終身特聘教授廖一久博士、國立台灣海洋大學環境生物與漁業科學學系王佳惠博士、國立台灣海洋大學環境生物與漁業科學學系黃祥麟博士，對於論文的細心審閱，並且給予寶貴的意見，使此論文更加的完善，

對此由衷的感謝。

在實驗室這兩年中，感謝怡誠學長、岱倪學姊、明惠學姊、小陶學姊、給予實驗上寶貴的意見；感謝皖之學姊在我們肚子餓，給予我們點心讓我們可以放鬆許多；感謝敬華學姊這一年來教導我帶領我很快的進入狀況，讓我事半功倍。另外也要感謝我的同學阿梅、阿旗、阿元、邦又、敏睿、鳳媖不只是實驗上給予意見，也在閒暇的時候一起聚餐、放鬆、談心讓心靈上有所慰藉。也感謝實驗室碩一、碩二的學弟妹的鼓勵與實驗上的幫助。也要感謝父母的支持與關懷讓我可以放心的在這個領域上學習。

在台大漁科所這兩年的日子裡我學習許多，也了解產業目前的所面臨的困境，

對此讓我有所使命感，希望在未來的日子裡可以貢獻所學，在鰻魚產業上能夠有所貢獻。

陳健澤謹誌中華民國 102 年 8 月

(3)

i

中文摘要

日本鰻是日本、韓國、台灣及中國重要養殖魚種，目前養殖業者的鰻苗完全

仰賴漁民在河口捕撈。然而其資源量從 1970 年代以後急遽下降；且近年來因各

國都市及工業不斷擴張，而使鰻魚的自然棲地受到嚴重的破壞。本研究擬探討從

1970 至 2010 年代，東亞四國日本鰻棲地改變的情形，與鰻魚長期資源量相互比

較，觀察是否有所關聯性。

本研究實驗方法使用衛星遙測技術，將其應用在土地覆蓋之改變 (Land

Cover Change, LCC).衛星遙測技術是一項分析時間空間改變有效率的工具，它可

以大範圍紀錄古今的地理樣貌，以提高效率以及精準度以利實驗分析。而本研究

區域是從東亞四國中，各選定四條主要鰻苗捕撈河川，藉由 USGS 下載各河川衛

星照片後，再由 Arc GIS 分析其棲地改變之情形。結果顯示，日本在 1970 至 2010

年代當中，河川天然棲地長度減少 21%，天然面積減少 27%，棲地品質指數

(habitat quality index, HQI) 減少 6%；韓國，天然棲地長度減少 46%，天然

面積減少 57%，HQI 減少 29%；台灣，天然棲地長度減少 22%，天然面積減少 53%，

HQI 減少 50%；中國，天然棲地長度減少 76%，天然面積減少 81%， HQI 減少 25%。

在鰻苗長期資源量方面，日本官方平均年產量在 1970 年代為 80.6 噸，1990 年

代為 35.9 噸，而近 5 年來為 6.6 噸，資源量在 1970 年代至現今減少 92%；而根

據福隆當地漁民資料，1984~1995 年間，年平均總產量 334096 隻，而 2007~2013

年間，年平均總產量 14190 隻，資源量從 1970 年代至今減少達 96%。

(4)

ii

結論而言，東亞地區在 1970 到 2010 年代，天然棲地總長度減少 47%，天然

棲地總面積減少 81%，HQI 減少 25%。與鰻苗資源量官方統計減少 92%及福隆當地

漁民統計減少 96%相互比較，鰻苗資源量應受到棲地破壞而減少。

關鍵字：棲地破壞、日本鰻、衛星遙測技術、東亞四國、鰻魚資源量

(5)

iii

Abstract

The Japanese eel is an important aquaculture species in Japan, Korea, Taiwan and

China. At present the only source of glass eels needed by fish farmers comes solely

from the catches made by fishermen at river mouths. However eel stocks have been in

rapid decline since the 1970s. Furthermore with urbanization and constant expansion of

industrialization in various countries, the natural habitat of eels have been severely

damaged. The aim of this study is to discuss the relationship between habitat changes in

the four East Asian countries and the long-term eel stock size from the 1970s to the

early 2010s.

The method of this study is using satellite remote sensing on land cover change

(LCC). Satellite remote sensing is an efficient tool for analyzing temporal and spatial

changes as it could record geographical features on a large geographical scales over

times to enhance efficiency and accuracy to facilitate data analysis. Present study focus

on four major eel-catching rivers in each of the four East Asian countries. Then satellite

images of those rivers were downloaded from the USGS website and fed to ArcGIS to

analyze the condition of habitat change in each of them.

The result of this study shows that in the period of 1970s~2010s in Japan, the

length of natural habitats in the rivers decreased by 21%, loss of natural areas is 27%,

and the habitat quality index, (HQI) decreased by 6%. In Korea, the length of natural

habitats in the rivers decreased by 46%, the loss of natural areas is 57%, and the HQI

(6)

iv

decreased by 29%. In Taiwan, the length of natural habitats in the rivers decreased by

22%, the loss of natural areas is 53%, and the HQI decreased by 50%. In China, the

length of natural habitats in the rivers decreased by 76%, the loss of natural areas is

81%, and the HQI decreased by 25%.

In terms of long-term glass eel stock sizes, Japanese official data shows 80.6 tons

of annual production in the 1970s, 35.9 tons in the 1990s and in the recent 5 years it is

6.6 tons. The eel stock size decreases by 92% from the 1970s to the present day.

According to local fisherman in Fulung, eel stock data in the period of 1984~1995, the

mean annual catch of glass eels is 334096, however, in the period of 2007~2013, the

total number of glass eels is 14190 on average per year and the stock size decreases by

96% from the 1970s to the present day.

In summary, in East Asia from the 1970s to the 2010s, the total length of natural

habitats decreased by 47%, the loss of total natural habitats is 81%, and the HQI

decreased by 25%. Compared with the official eel stock decrease of 92% in Japan, and

local fisherman data of a decrease of 96% in Fulung, this shows eel stock size decline

should be related to habitat loss.

Key words: habitat destruction, Japanese eel, satellite remote sensing, East Asia four

countries, eel stock size

(7)

v

Abstract ... iii

Contents ... v

Figure legend ... viii

Table contents ... xi

Introduction ... 1

Materials and methods ... 6

Study area ... 6

Landscape image collection and processing ... 6

Remote sensing image classification ... 7

Eel resource data ... 8

Result ... 9

Remote sensing image ... 9

Japan ... 9

Ten-ryu River (天龍川) ... 9

Ohyodo River (大淀川) ... 10

(8)

vi

Ni-yodo River (仁淀川) ... 10

To-ne River (利根川) ... 11

Korea ... 12

Han River (漢江) ... 12

Geum River (錦江) ... 12

Yeongsan River (榮山江) ... 13

Nakdong River (洛東江) ... 14

Taiwan ... 14

Danshui River (淡水河) ... 14

Lanyang River (蘭陽溪) ... 15

Zhuoshuei River (濁水溪) ... 16

Kaoping River (高屏溪) ... 17

China ... 17

Minjiang River (閩江) ... 17

Pearl River (珠江) ... 18

Qiantang River (錢塘江) ... 19

Yangtze River (長江) ... 19

Eel resource data ... 20

Discussion... 22

(9)

vii

Eel catch data ... 22

Habitat quality index, HQI ... 23

The long term habitat change in East Asia ... 23

Habitat destruction ... 24

The other factors of eel resource decline ... 26

Conclusion ... 30

References ... 31

(10)

viii

Figure legend

Fig.1 The artificial buildings of habitat destruction dam (a), aquaculture zone (b),

harbor r(c), riverbank (d). ... 40

Fig.2 The four main rivers catching eel area in Japan; Ten-ryu River (天龍川), Ohyodo River (大淀川), Ni-yodo River (仁淀川), and To-ne River (利根川) ... 42

Fig.3 The four main rivers catching eel area in Korea; Han River (漢江), Geum River (錦江), Yeongsan River (榮山江) and Nakdong River (洛東江) ... 43

Fig.4 The four main rivers catching eel area in Taiwan; Danshui River (淡水河), Lanyang River (蘭陽溪), Zhuoshuei River (濁水溪) and Kaoping River (高屏溪) 44 Fig.5 The four main rivers catching eel area in China; Minjiang River (閩江), Pearl River (珠江), Qiantang River (錢塘江) and Yangtze River (長江) ... 45

Fig.6 The Ten-lyu River (天龍川) of Japan 1970s(A) 1990s(B) 2010s(C) ... 46

Fig.7 The Ohyodo River (大淀川) of Japan 1970s(A) 1990s(B) 2010s(C) ... 47

Fig.8 The Ni-yodo River (仁淀川) of Japan 1970s(A) 1990s(B) 2010s(C) ... 48

Fig.9 The To-ne River (利根川) of Japan 1970s(A) 1990s(B) 2010s(C) ... 49

Fig.10 The Han River (漢江) of Korea 1970s(A) 1990s(B) 2010s(C) ... 50

Fig.11 The Geum River (錦江) of Korea 1970s(A) 1990s(B) 2010s(C) ... 53

Fig.12 The Yeongsan River (榮山江) of Korea 1970s(A) 1990s(B) 2010s(C) ... 54

Fig.13 The Nakdong River (洛東江) of Korea 1970s(A) 1990s(B) 2010s(C) ... 55

(11)

ix

Fig.14 The Dansuie River (淡水河) of Taiwan 1970s(A) 1990s(B) 2010s(C)... 56

Fig.15 The Lanyang River (蘭陽溪) of Taiwan 1970s(A) 1990s(B) 2010s(C) ... 59

Fig.16 The Zhuoshuei River (濁水溪) of Taiwan 1970s(A) 1990s(B) 2010s(C) ... 62

Fig.17The Kaoping River (高屏溪) of Taiwan 1970s(A) 1990s(B) 2010s(C) ... 65

Fig.18 The Minjiang River (閩江) of China 1970s(A) 1990s(B) 2010s(C) ... 66

Fig.19 The Pearl River (珠江) of China 1970s(A) 1990s(B) 2010s(C)... 69

Fig.20 The Qiantang River (錢塘江) of China 1970s(A) 1990s(B) 2010s(C) ... 70

Fig.21 The Yangtze River (長江) of China 1970s(A) 1990s(B) 2010s(C) ... 71

Fig.22 The bar chart of four Japanese rivers nature length (A),area (B) and HQI (C) in 1970, 1990 and 2010; four Japanese rivers total HQI value in 1970, 1990 and 2010 (D) ... 72

Fig.23 The bar chart of four Korean rivers nature length (A),area (B) and HQI (C) in 1970, 1990 and 2010; four Japanese rivers total HQI value in 1970, 1990 and 2010 (D) ... 74

Fig.24 The bar chart of four Taiwan rivers nature length (A),area (B) and HQI (C) in 1970, 1990 and 2010; four Japanese rivers total HQI value in 1970, 1990 and 2010 (D) ... 76

Fig.25 The bar chart of four China rivers nature length (A),area (B) and HQI (C) in

1970, 1990 and 2010; four Japanese rivers total HQI value in 1970, 1990 and 2010 (D)

(12)

x

... 78

Fig.26 The bar chart of East Asia four countries HQI value in 1970, 1990 and 2010 . 80 Fig.27 The eel resource run chart of Japan ... 81

Fig.28 The eel resource run chart of Korea ... 82

Fig.29 The eel resource run chart of Taiwan ... 83

Fig.30 The eel resource run chart of China ... 84

Fig.31 the eel resource run chart of Fulung ... 85

(13)

xi

Table contents

Table 1 The historical Landsat data path/row of Taiwan. ... 86

Table 2 The historical Landsat data path/row of Japan. ... 87

Table 3 The historical Landsat data path/row of Korea. ... 89

Table 4 The historical Landsat data path/row of China. ... 90

Table 5 the four countries eel resource official data during 1970s to 2010s ... 92

Table 6 the Fulung eel resource data during 1980s to 2010s ... 94

Table 7A The nature habitat and artificial building length and area of Ten-lyu River (天龍川) of Japan ... 96

Table 7B The percentage of HQI value of Ten-lyu River (天龍川) of Japan ... 96

Table 8A The nature habitat and artificial building length and area of Ohyodo River (大淀川) of Japan ... 97

Table 8B The percentage of HQI change value of Ohyodo River (大淀川) of Japan ... 97

Table 9A The nature habitat and artificial building length and area of Ni-yodo River (仁淀川) of Japan ... 98

Table 9B The percentage of the HQI change value of Ni-yodo River (仁淀川) of Japan ... 98

Table 10A The nature habitat and artificial building length and area of To-ne River (利

(14)

xii

根川) of Japan ... 99

Table 10B The percentage of the HQI change value of To-ne River (利根川) of Japan

... 99

Table 11A The nature habitat and artificial building length and area of Han River (漢

江) of Korea ... 100

Table 11B The percentage of the HQI change value of Han River (漢江) of Korea

... 100

Table 12A The nature habitat and artificial building length and area of Geum River

(錦江) of Korea ... 101

Table 12B The percentage of the HQI change value of Geum River (錦江) of Korea

... 101

Table 13A The nature habitat and artificial building length and area of Yeongsan River

(榮山江) of Korea ... 102

Table 13B The percentage of the HQI change value of Yeongsan River (榮山江) of

Korea ... 102

Table 14A The nature habitat and artificial building length and area of Nakdong River

(洛東江) of Korea ... 103

Table 14B The percentage of the HQI change value of Nakdong River (洛東江) of

Korea ... 103

(15)

xiii

Table 15A The nature habitat and artificial building length and area of Dansuie River

(淡水河) of Taiwan ... 104

Table 15B The percentage of the HQI change value of Dansuie River(淡水河) of

Taiwan ... 104

Table 16A The nature habitat and artificial building length and area of Lanyang River

(蘭陽溪) of Taiwan ... 105

Table 16B The percentage of the HQI change value of Lanyang River (蘭陽溪) of

Taiwan ... 105

Table 17A The nature habitat and artificial building length and area of Zhuoshuei

River (濁水溪) of Taiwan ... 106

Table 17B The percentage of the HQI change value of Zhuoshuei River (濁水溪) of

Taiwan ... 106

Table 18A The nature habitat and artificial building length and area of Kaoping River

(高屏溪) of Taiwan ... 107

Table 18B The percentage of the HQI change value of Kaoping River (高屏溪) of

Taiwan ... 107

Table 19A The nature habitat and artificial building length and area of Minjiang River

(閩江) of China ... 108

Table 19B The percentage of the HQI change value of Minjiang River (閩江) of

(16)

xiv

China ... 108

Table 20A The nature habitat and artificial building length and area of Pearl River (珠

江) of China ... 109

Table 20B The percentage of the HQI change value of Pearl River (珠江) of China

... 109

Table 21A The nature habitat and artificial building length and area of Qiantang River

(錢塘江) of China ... 110

Table 21B The percentage of the HQI change value of Qiantang River (錢塘江) of

China ... 110

Table 22A The nature habitat and artificial building length and area of Yangtze River

(長江) of China ... 111

Table 22B The percentage of the HQI change value of Yangtze River (長江)of China

... 111

Table 23 The eel resource in 1970s, 1990s and 2010s (A) and the percentage of the eel

resource change value (B) of Japan ... 112

resource change value (B) of Taiwan ... 113

resource change value (B) of Korea ... 114

(17)

xv

resource change value (B) of China ... 115

Table 27 The eel resource of year average in 1984~1995 and 2006~2013 (A) and the percentage of the eel resource change value (B) of Fulung ... 116

Table 28 The percentage of the HQI change value of Japan ... 117

Table 29 The percentage of the HQI change value of Korea ... 118

Table 30 The percentage of the HQI change value of Taiwan ... 119

Table 31 The percentage of the HQI change value of China ... 120

Table 32 The percentage of the HQI change value of East Asia ... 121

(18)

1

Introduction

The catadromous eel (genus Anguilla) is an important aquaculture species in the

East Asia. Due to the large-scale artificial production techniques of eel fry (glass eel)

are unavailable; the fry source for eel aquaculture must be captured by fisherman in

the estuaries. However, the natural stock of glass eel, especially the Japanese eel

Anguilla japonica, European eel A. Anguilla and American eel A. rostrata, has been

significantly decreasing for the last three decades owing to overfishing, habitat

destruction, global climate change and other unknown factors (Tatsukawa, 2003;

Dekker, 2003; Casselman, 2003). The decline of natural eel resources causes

considerable impact on eel aquaculture industry in East Asia.

Around the world, there are 19 species of anguillid distributed in the coastal

areas of North Atlantic Ocean, Indian Ocean and West Pacific Ocean (Aoyama, 2009).

All Anguilla larvae are spawned in the tropical or subtropical ocean. After eggs

hatching, the eel larvae with transparent and leaf-like body are called leptocephali,

and they are transported by oceanic currents for several months to their growth

habitats. When leptocephali grow to specific maximum size, they then metamorphose

into glass eels when they drift near continental shelf (Miller, 2009). After

metamorphosis, glass eel enter the estuaries, rivers or lakes and appear pigment on the

body. They grow for several years and then become silver eel ready for seawater

(19)

2

spawning migration. They migrate thousands of kilometers back to the birth place to

spawn and then die (Tsukamoto, 2006).

However, there are many artificial buildings near the estuarine due to

urbanization and industrialization during past few decades. Urban growth, particularly

the movement of residential and commercial land use to rural areas at the periphery of

metropolitan areas, has long been considered a sign of regional economic vitality. But,

its benefits are increasingly balanced against ecosystem impacts, including

degradation of air and water quality and loss of farmland and forests, and

socioeconomic effects of economic disparities, social fragmentation and infrastructure

costs (Squires, 2002; Fei, 2005).

For example, dams (Fig. 1a) may limit the upstream movement of eels such that

eel numbers often decrease above dams (Goodwin, 1999; Machut, 2007) and increase

immediately below dams (Wiley, 2004; Machutet, 2007). Consequently, barriers may

influence stream community composition and population dynamics in upstream and

downstream directions. Upstream of dams, decreased eel densities may influence

stream fish communities by removing a native piscivore which could otherwise

comprise over 25% of the total fish biomass in streams (Smith, 1955; Ogden, 1970).

Freshwater mussel distributions may also be limited through restrictions of the fish

host movements that are necessary for upstream dispersal of mussel glochidia

(20)

3

(Williams, 1993; Watters, 1996). Downstream of dams, increased eel densities may

increase intraspecific competition and decrease per capita growth rates (Machut,

2007). Reduced access to headwater streams may also influence eel stock–recruitment

dynamics by decreasing the production of female eels (Krueger, 1999).

The rapid expansion of coastal aquaculture (Fig. 1b) has serious environmental

and socioeconomic consequences, which include large-scale removal of valuable

coastal wetlands, land subsidence, acidification, salinization of groundwater and

agricultural land, and subsequent loss of goods and services generated by natural

resource systems (Chua, 1992).

Not only dam, aquaculture but also harbor (Fig. 1c), riverbank (Fig. 1d) which

replace the mangrove and wetland by land reclamation were creating the eel habitat

destruction. Land reclamation has been a common practice to produce valuable land

in coastal areas. The impact of land reclamation on coastal environment and marine

ecology is well recognized and widely studied. It has not been recognized yet that

reclamation may change the regional ground water regime, which may in turn modify

the coastal environment, flooding pattern, and stability of slopes and foundations (

Jiao,

2001)

.

Land cover change (LCC) is caused by human disturbances and/or natural events

(Wen, 2011). Human activities now affect most of the terrestrial biosphere and are

(21)

4

increasing in intensity and extent (Jeremy, 2003). Ensuing habitat loss and

degradation impair ecosystem function (Defries, 1999) and reduce the value of

ecosystem services for humans (Daily, 1997), and natural events such as global

warming, flooding, rivers dry up etc. LCC at different scales from local to global,

especially quantitative analysis of LCC has been a main concern to scientists and

researchers in the past century, particularly the past few decades around the world

(Wen, 2011). However, traditional field ecological data do not translate readily to

regional or global extents, and models derived purely from such local data are

unlikely to predict the global consequences of human activities (Jeremy, 2003).

Monitoring land cover changes using multi-temporal remotely sensed data

provides an accurate evaluation of human impact on the environment (Abdullah,

2012). Importantly, remotely sensed imagery provides an efficient means of obtaining

information on temporal trends and spatial distribution of urban areas (Fei, 2005) near

the estuary and river. The long-term data record obtained from Landsat satellites are a

valuable resource for monitoring land use cover change (USGS, 2011). Accurate and

up-to-date land use information is essential for environmental planning, understanding

the impacts to terrestrial ecosystems (Wulder, 2007) and achieving sustainable

development (Alphan, 2003).

For example, the study of Abdullah, (2012) is to investigate the extent of ship

(22)

5

breaking activities in Bangladesh along the Sitakunda coast, various spatial and non-

spatial data were obtained by remote sensing imagery. The other study “River

pollution remediation monitored by optical and infrared high-resolution satellite

images (Paolo, 2013) “were to use high resolution satellite images combined with a

classical remote sensing methodology to monitor vegetation conditions along the

Bormida River.

The hypothesis of this study focus on whether the artificial buildings like dam,

harbor, aquaculture and riverbank cause the Japanese eel resource decline; thus this

study use the technology of satellite remote sensing to find the relationship between

Anguilla japonica resource and habitat destruction during the past 40 years.

(23)

6

Materials and methods

Study area

The main Anguilla japonica glass eel catching area is around East Asia

countries, including Japan, Korea, Taiwan and China. Choosing Ten-ryu River (天龍

川), Ohyodo River (大淀川), Ni-yodo River (仁淀川), and To-ne River (利根川) as

Japan study area (Fig. 2) is according to the "日本養殖新聞", due to the four rivers is

the main eel catching area in Japan. Choose Han River (漢江), Geum River (錦江),

Yeongsan River (榮山江) and Nakdong River (洛東江) as the Korea (Fig. 3) study

area due to Han River, Jin River Luo-Dung River are the top three rivers. Choosing

Rung-Shan River as the Korea study area by news from the Korean who working in

eels.

Take Danshui River (淡水河), Lanyang River (蘭陽溪), Zhuoshuei River (濁水

溪) and Kaoping River (高屏溪) as study areas because of they are the main Anguilla

japonica catching area of Taiwan (Fig.4).Take Minjiang River (閩江), Pearl River (珠

江), Qiantang River (錢塘江) and Yangtze River (長江) as study areas due to they are

the main Japanese eel catching area of China (Fig. 5).

Landscape image collection and processing

The main rivers of East Asia countries Taiwan, China, Korea and Japan are the

(24)

7

major Japanese eel catching area. Historical Landsat data covering the rivers and

estuaries of four countries were collected from USGS website (http://glovis.usgs.gov/)

and the path/row were recorded on Table 1~4. Each Landsat image contained three

color bands (R-red, G-green, B-blue) composited with ArcMap (ESRI, 2008) either

RGB231 for MSS data (Landsat 1, 3) or RGB742 for TM/ETM+ data (Landsat 5, 7)

to highlight the contrast between vegetative and urban landscape (Kerr, 2003; Merem,

2008; Castilla, 2009). All satellite images were first geo-referenced to clearly

identifiable landmarks with ArcMap 9.3 (ESRI, 2008; Huang, 2013). When the

images were processed, we calculate the habitat length, area and HQI by a tool called

“Measure”.

Remote sensing image classification

After processing the satellite images, this study classifies all river valleys under

study into two parts, artificial buildings and natural habitats, and then calculates their

percentage in length and area. Artificial buildings include 1. riverbanks 2. dams and

the upper dam areas 3. aquaculture zones 4. harbors. The remaining area is regarded

as a natural habitat. The boundary line of the river under study is restricted to the

plane because it is the main habitat of Japanese eels. After measuring the length and

area of the river, the area data was used to weight a value called HQI (habitat quality

(25)

8

index). According to the study Kimura, (2012) when the artificial revetment rate in

Japanese lakes is 100%, then the decreasing rate of fish catch is 25%; so this study

refers to the data and makes the natural habitat data multiplied by 1 and the artificial

building data multiplied by 0.75 based on the effect of riverbanks on eel stock

abundance (Kimura, 2012). Furthermore, the dam, will stop the eel migrating to the

upper stream, thus the upper dam valley by 0.

Eel resource data

The four countries official eel catch data was collected from 日本養殖新聞

during 1970s to 2010s and the original data was recorded on Table 5. We also

collected the eel catch data form Fulung fisherman during 1980s to 2010s and the

original data was recorded on Table 6.

(26)

9

Result

Remote sensing image

Japan:

1.Ten-ryu River (天龍川)

The length of the natural habitat of Ten-ryu River in the 1970s, the 1990s and the

early 2010s is 27km, 27km and 27km respectively with the corresponding area of 8

km², 9 km² and 6 km²; the HQI value in the 1970s, 1990s and 2010s is 8 km², 9 km²

and 6 km² (Table 7A); the percentage of HQI in the 1970s~1990s, the 1990s~2010s

and 1970s~2010s is 13%,-33% and -25% (Table 7B).

Although there were many artificial buildings near the river in the past 40 years,

in the 1970s, the 1990s and the early 2010s, there were many sandbanks in the middle

of the river (Fig. 6 yellow frame). Therefore the eel habitat area did not decrease

significantly and the data of this river shows no significance in the past 40 years.

However, the habitat area still changed as the sandbank area is changed by climate

factors like water quantity (Table 7).

Because the Scan Line Corrector (SLC) on Landsat-7 used to compensate the

forward motion of the whisk-broom sensor malfunctioned on May 31, 2003, there is a

problem with the images. As a result, the images acquired from Landsat-7 show data

gaps that occupy about 22% of the entire scene (楊, 2009). Fortunately, this study is

not affected by this problem and we can still calculate the habitat change values

(27)

10

accurately.

2. Ohyodo River (大淀川)

The nature habitat length of Ohyodo River in 1970s, 1990s and 2010s is 35km,

30km and 30km, and the area is 7 km², 4 km² and 4 km²; the HQI value in 1970s,

1990s and 2010s is 8 km², 7 km² and 7 km² (Table 8A); the percentage of HQI in

1970s~1990s, 1990s~2010s and 1970s~ 2010s is -10%, 0% and -10% (Table 8B).

There was no harbor near the estuary in the 1970s in this river,, but in the 1990s

and the early 2010s, there was a harbor near the estuary (Fig. 7 yellow frame) – the

main factor on habitat change in this river. However in the 1970s, not many artificial

buildings were in the river. Yet in the 1990s and the early 2010s, more artificial

buildings began to appear. Even so, there were no significant habitat changes between

the 1990s and the early 2010s (Table 8).

3. Ni-yodo River (仁淀川)

The nature habitat length of Ni-yodo River in 1970s, 1990s and 2010s is 15km,

17km and 17km,and the area is 3 km², 3 km² and 3 km²; the HQI value in 1970s,

1970s~1990s,1990s~2010s and 1970~ 2010 is 25%, 0% and 25% (Table 9B).

(28)

11

There is no obvious habitat change in the past 40 years in this river, but it is

worth mentioning that in the 1970s, there was no tributary flowing to the river, but in

the 1990s and the early 2010s, there was a tributary flowing to this river (Fig. 8). At

first, we thought there was no tributary because of the drought season in the 1970s, so

we checked the other same path/row remote sensing images of this area to make sure.

We found there was really no tributary near the river. Therefore the total length and

area in the 1990s and the early 2010s are higher and bigger than those of the 1970s

because of the tributary. Indeed more and more artificial buildings begin to appear

year by year (Table 9).

4. To-ne River (利根川)

The nature habitat length of To-ne River in 1970s, 1990s and 2010s is 240km,

225km and 175km, and the area is 227 km², 179 km² and 166 km²; the HQI value in

1970s, 1990s and 2010s is 248 km², 237 km² and 233 km² (Table 10A); the

percentage of HQI in 1970s~1990s, 1990s~2010s and 1970s~2010s is -5%, -2% and -

6% (Table 10B).

More and more artificial buildings emerged year by year in this river in the past

40 years that caused the natural habitats to decrease year by year (Fig. 9). The data

also shows this trend (Table 10).

(29)

12

Korea:

1. Han River (漢江)

The nature habitat length of Han River in 1970s, 1990s and 2010s is 150km,

1970s, 1990s and 2010s is 190km², 182km² and 166km² (Table 11A); the percentage

of HQI in 1970s~1990s, 1990s~2010s and 1970s~2010s is -4%, -9% and -13% (Table

11B).

This study compares the two different parts of the river, the estuary and extended

city area. In the 1970s and the 1990s there were no artificial buildings like harbors

and aquaculture areas on the island near the river and the river estuary, but in the early

2010s there were many harbors and aquaculture areas on this island (Fig. 10 A-1, B-1

and C-1). The other part is the city extended significantly from the 1970s to the 2010s

as this area is Korea's capital, Seoul (Fig. 10 A-2, B-2 and C-2). This trend is recorded

in Table 11.

2. Geum River (錦江)

The nature habitat length of Geum River in 1970s, 1990s and 2010s is 55km,

1990s and 2010s is 71km², 59km² and 56km² (Table 12A); the percentage of HQI in

(30)

13

1970s~1990s,1990s~2010s and 1970s~2010s is -18%, -6% and -22% (Table 12B).

There was no harbor near the estuary in the 1970s in this river (Fig. 11 A-1 black

frame), but there was a harbor near the estuary in the 1990s and the 2010s (Fig. 11 B-

1 and C-1 black frame). Besides, the harbor was built by land accretion due to the fact

that the harbor built connected the estuary land and the outside island. So we

calculated the total length of the river in the early 2010s and found that it is higher

than in the 1970s and the 1990s because of harbor extension by land accretion (Table

12).

3. Yeongsan River (榮山江)

The nature habitat length of Yeongsan River in 1970s, 1990s and 2010s is 159km,

1990s and 2010s is 120km²,51km²and 41km² (Table 13A); the percentage of HQI in

1970s~1990s, 1990s~2010 and 1970s~2010s is -58%, -19% and -66% (Table 13B).

There were plenty of natural and territorial waters in the 1970s in this river ,

however in the 1990s and the early 2010s, parts of the territorial waters were replaced

by aquaculture areas, harbors and riverbanks (Fig.12 black frame), so the total area

declined significantly in the past 40 years. Especially for the period from the 1970s to

the 1990s, the HQI decreased by 66% (Table 13).

(31)

14

4. Nakdong River (洛東江)

The nature habitat length of Nakdong River in 1970s, 1990s and 2010s is 116km,

1970s~1990s, 1990s~2010s and 1970s~ 2010s is -5%, -5% and -9% (Table 14B).

40 years that caused the natural habitats to decrease year by year as well (Fig. 13).

The data also shows this trend (Table 14).

Taiwan:

1. Danshui River (淡水河)

The nature habitat length of Danshui River in 1970s, 1990s and 2010s is 103km,

1970s~1990, 1990s~2010s and 1970s~ 2010s is -69%, 1% and -69% (Table 15B).

There is a lake flowing to Danshui River in the 1970s (Fig. 14 A-2 black frame).

But in the 1990s and the early 2010s (Fig. 14 B-2 and C-2 black frame), the lake

shrunk due to land accretion leading to loss of natural habitats. Furthermore, this was a “winding” tributary flowing to the river (Fig. 14 A-2 black cycle), but in the 1990s

and the early 2010s, the tributary became “straight” (Fig. 14 B-2 and C-2 black cycle)

(32)

15

and this factor caused habitat loss and destruction between the period of the 1970s,

the 1990s, and the 2010s.

We also discover a habitat change factor. In the 1970s there was no weir on this

river (Fig. 14 A-1), but in the 1990s there were two weirs on the river (Fig. 14 B-1). In

2004, typhoon Aere destroyed one of the weirs, making the remote sensing images of

the 2010s to show just one weir (Fig. 14 C-1) (台灣自來水公司). For this reason, the

natural length and area value increased from the 1990s to the early 2010s. In the past

40 years, urbanization caused a 69% (Table 15) habitat loss because the river is near

the capital of Taiwan (Fig. 14 A-2, B-2 and C-2).

2. Lanyang River (蘭陽溪)

The nature habitat length of Lanyang River in 1970s, 1990s and 2010s is 55km,

1970s~1990s,1990s~2010s and 1970s~2010s is 6%, -49% and -46% (Table 16B).

There were not many artificial buildings near or on the river in the 1970s, and the

watercourse was extensive (Fig. 15 A-1). But in the 1990s and the early 2010s, there

were more artificial buildings near or on the river, making the watercourse thinner

(Fig.15 B-1 and C-1).

According to the data in Table 16 and Figure 24, the habitats change significantly.

(33)

16

Because in the period from the 1990s to the early 2010s, there were more land

accretion on the river. For example, in the river estuary there was land accretion that

caused habitat loss (Fig. 15 B-1 black cycle and Fig. 15 C-1 black cycle). In the other

example, in one of the tributaries in the 1990s, there was more water quantity in this

area. But in the early 2010s, the tributary disappeared (Fig. 15 B-1 black frame and

Fig. 15 C-1 black frame).

3. Zhuoshuei River (濁水溪)

The nature habitat length of Zhuoshuei River in 1970s, 1990s and 2010s is 84km,

1990s and 2010s is 108km², 69km²and 52km² (Table 17A); the percentage of HQI in

1970s~1990s,1990s~2010s and 1970s~2010s is -36%, -25% and -52% (Table 17B).

There were less natural and artificial structures near the estuarine area in the

1970s in this river (Fig. 16 A-1). But in the 1990s, there were many aquaculture zones

near the estuarine area that caused habitat loss (Fig. 16 B-1). Disastrously, there was

an industrial estate, The No. 6 Naphtha Cracker Complex (Mailiao) of Formosa

Petrochemical Corp (Fig. 16 C-1), near the coast and it not only caused habitat loss,

but also released industrial sewage to the water, leading to serious damage of the

estuarine area.

In addition, in the 1970s and the 1990s (Fig. 16 A-2 B-2), there was no dam on

(34)

17

the river. But in the early 2010s there was a weir, Chi-Chi (Yao, 2009) on the

upstream of Zhuoshuei River that made the river lose about 13km of habitat length

(Fig. 16 C-2 black frame) (Table 17).

4. Kaoping River (高屏溪)

The nature habitat length of Kaoping River in 1970s, 1990s and 2010s is 92km,

1970s~1990s,1990s~2010s and 1970s~ 2010s is -44%, 17% and -34% (Table 18B).

40 years. But from the 1990s to the early 2010s, the total habitat length decreased by

3km while the habitat area change is an increase of 9km (Fig. 17). The explanation for

this is that during the 20 years, there were many artificial buildings near the river that

caused the percentage of habitat length to decrease. However the sandbanks in the

river changed tremendously in this period and caused the percentage of habitat area to

increase (Table 18).

China:

1. Minjiang River (閩江)

The natural habitat length of Minjiang River in the 1970s, the 1990s and the

(35)

18

early 2010s is 92km, 79km and 31km, and the area is 86 km², 83 km² and 47 km²; the

HQI value in 1970s, 1990s and 2010s is 87km², 91km²and 94km² (Table 19A); the

percentage of HQI in 1970s~1990s, 1990s~2010s and 1970s~2010s is 5%, 3% and

9% (Table 19B).

Urbanization and habitat loss increased year by year in the past 40 years in this

river (Table 19). But the data shows the total length of the river became shorter and

the total area became larger because there was a tributary in the 1970s and the 1990s

(Fig. 18 A-1 and B-1 black frame). It is a natural, winding tributary with many

sandbanks. But in the early 2010s, the tributary became straight with fewer sandbanks,

and there were many artificial structures near the riverside (Fig. 18 C-1 black frame).

For this reason, the total area increased from the 1990s to the early 2010s but the total

length decreased.

2. Pearl River (珠江)

The nature habitat length of Pearl River in 1970s, 1990s and 2010s is 242km,

1970s, 1990s and 2010s is 1209km², 986km² and 772km² (Table 20A); the percentage

of HQI in1970s~1990s, 1990s~2010s and 1970s~2010s is -18%,-22% and -32%

(Table 20B).

In this river, there are two data changes worth discussing. One is the total length

(36)

19

became longer and the other one is the total area became smaller over the past 40

years. In the 1970s, the area faced less artificial destruction to the environment (Fig.

19A black frame), but in the 1990s (Fig. 19B black frame), land accretion area

increased. In the early 2010s, the situation had become more serious (Fig. 19C black

frame). So the total length became longer due to increased accretion area and the total

area became smaller because of the same factor (Table 20).

3. Qiantang River (錢塘江)

The nature habitat length of Qiantang River in 1970s, 1990s and 2010s is 224km,

1970s, 1990s and 2010s is 7124km², 6336km² and 5363km² (Table 21A); the

percentage of HQI in 1970s~1990s, 1990s~2010s and 1970s~ 2010s is -11%, -15%

and -25% (Table 21B).

In this river, the habitats saw a loss year by year in the past 40 years. In the

1970s there were fewer damages to the environment (Fig. 20A); but until the 1990s

there were many artificial structures near the riverside (Fig. 20B) and until the early

2010s the situation has become more severe than the 1990s (Fig. 21 C) (Table 21).

4. Yangtze River (長江)

The nature habitat length of Yangtze River in 1970s, 1990s and 2010s is 523km,

(37)

20

286km and 113km, and the area is 3199 km², 1637 km² and 291 km²; the HQI value

in 1970s, 1990s and 2010s is 3798km², 3399km²and 2928km² (Table 22A); the

percentage of HQI in 1970s~1990s, 1990s~2010s and 1970s~2010s is -11%, -14%

The habitat loss was also year by year in the past 40 years in this river. What is

worth mentioning is the total area (Table 22) from the 1990s to the early 2010s

decreased significantly due to the fact that in the 1970s and the 1990s there was not

much land accretion in the lake, Kao-Yu (Fig. 21 A and B yellow frame). But in the

early 2010s there were many artificial structures in the lake, making the area smaller

(Fig. 21 C yellow frame). Therefore the data shows the total area decreased

tremendously during the period of the 1990s to the early 2010s.

This study shows a bar chart of the four rivers of Japan, Korea, Taiwan and

China in the 1970s, the 1990s and the early 2010s in Figures 22~25. This study also

shows a bar chart of the four East Asian countries in the 1970s, the 1990s and the

early 2010s in Figure 26.

Eel resource data

According to the “日本養殖新聞” glass eel catch data of East Asian four

countries, Japan, the mean annual glass eel catch in 1970s,1990s and 2010s is 80.6

tons, 35.9 tons and 6.6 tons, respectively (Table 23A), and the percentage of the eel

(38)

21

catch value change in 1970s~1990s, 1990s~2010s and 1970s~2010s is -56%, -81%

Taiwan, the eel resource in 1990s and 2010s is 14.9 tons and 3.1 tons (Table

24A), and the percentage of the eel catch amount change in 1990s~2010s is -79%

(Table 24B).

Korea, the eel resource in 1990s and 2010s 8.7 tons and 3.4 tons (Table 25A),

and the percentage of the eel resource value change in 1990s~2010s is -60% (Table

25B).

China, the eel resource in 1980s, 1990s and 2010s is 43.2 tons, 36.7 tons and

38.6 tons (Table 26A), and the percentage of the eel resource value change in

1980s~1990s, 1990s~2010s and 1980s~2010s is -15%, 5% and -11% (Table 26B).

This study also made bar charts of each sites (Fig. 27~30).

As to local fisherman data in Fulung, this study lists data from 1984 to 1995, and

2007 to 2013 without data available from 1996 to 2006. The mean annual catch of

glass eels in the period from 1984 to 1995 is 334096 eels per year, and in the period

from 2007 to 2013 is 14190 per year (Table 27A). The percentage of eel catch change

value over this period, (1984~1995)~(2007~2013) is -96% (Table 27B). This study

makes a run chart in Figure 31.

(39)

22

Discussion

Eel catch data

This study uses eel stock data of the four countries of Japan, Korea, Taiwan and China from “日本養殖新聞”. All the eel stock data shows a decreasing trend (Fig.

27~30). However, it is worth mentioning that in 1978 China had a policy called “The

reform and opening-up policy” and the policy covered aquaculture (中投顧問 2010).

It meant the Chinese government wanted to develop the industry of aquaculture.

Therefore prior to 1978, there were not many cultured eels. But after the policy was

implemented, eel stock increased because many fishermen started to culture eels,

causing the requirement of glass eels to increase. Therefore the data of eel catch in

China shows no significant change.

This study refers to eel catch data in Japan as East Asian eel stock, because the

Japanese have been culturing eel for a long time, and their government demanded

strict records of the data, leading to complete data recording of the eel stocks in the

past 40 years, more importantly, according to Han, (2010), the population structures

of anguillid eels have long been considered panmictic. This is because sexually

mature stocks migrate and spawn in a single site, and their larvae are passively

transported back to their growth habitats by oceanic currents with a long larval

(40)

23

duration, making population genetic structuring quite impossible (Schmidt, 1925;

Tsukamoto, 1992; Avise, 1994; Tesch, 2003; Aoyama, 2009). Therefore, the Japanese

eel should be considered as a single management unit for conservation (Han, 2010).

Thus for this reason, Japanese catch data can reflect the whole East Asian eel stock

change. Generally speaking, eel stocks decreased heavily during the period according

to the official eel stock data.

Habitat quality index, HQI

There are many factors affecting the habitat, for example, riverbanks will cause

eel habitat destruction, dams will cut off the river and cause habitat loss, land

accretion will cause habitat loss too, thus, those factors appertain to water quantity.

Furthermore, the DO, PH, salinity, or other pollutions, and so on, appertain to the

water quality. In this study, we discuss the water “quantity” because the analytical

methods used in analyzing Landsat images in this study just show LCC and cannot

reflect the water quality.

The long term habitat change in East Asia

As a whole, in the past 40 years, Japanese natural habitats decreased not as

serious as those of the other East Asian countries. The percentage of HQI is -6%

(41)

24

(Table 28) (Fig. 22). As the Japanese government enacted a conscientious and careful

law about river management, and there are many professionals on environmental

conservation, furthermore, their people have a good concept of environmental

protection (曾). But the rate of habitat change in the other countries is higher because

their governments did not enact effective laws and instilled in their people the newest

concepts of environmental protection; For Korea, the percentage of HQI is -29%

(Table 29) (Fig. 23); For Taiwan, the percentage of HQI is -50% (Table 30) (Fig. 24);

For China, the percentage of HQI is -25% (Table 31) (Fig. 25) and the total HQI of

East Asia is -25% (Table 32) (Fig. 26)

Even though there were different decrease rates in the rivers of each country, as a

catadromous species Japanese eels would be transported to East Asian regions by

ocean currents after hatching (Miller, 2009). As such, the four countries of East Asia

is the mean habitat of the Japanese eel, because Japanese eel is a single panmictic

population of in East Asia thus (Han, 2010), if somewhere habitat loss or destruction

in East Asia, it will decrease the eel resource of whole East Asia.

Habitat destruction

The International Union for Conversation of Nature and Natural Resources

(IUCN) takes "habitat destruction" as the biggest reason for reducing biodiversity in

(42)

25

the last decade in the Red List of endangered species. The Red List shows 86% of

birds and mammals, 90% of freshwater fish as well as more than 30% of marine life

are directly affected by habitat destruction (NTNU). Therefore habitat destruction is

an important factor of organism resources.

In many tourist attractions like Jamaica, Dubai, Palau, etc (Chen, 2013), people

prefer to live the hotel witch near the coastline for ocean view and marine aqua

activities; and since time immemorial, the local resident live near the river for the

water or the other daily life. Human activities may cause dramatic changes to

landscapes, coastal line, and river habitat change (Wen, 2011).

Wetlands are the most productive natural environment, which gave birth to

countless species of aquatic, and is a very important link between terrestrial and

aquatic ecosystems, thus when people develop the wetlands by land accretion and

then it will cause the biodiversity decrease heavily (池), furthermore, people build

many riverbanks or revetments for flood defense, but it cause almost disappearance of

the supply of the detritus as food for the fish larvae and juveniles. These environment

changes are irreversible and permanently transform the native habitats and ecosystems

(Huang, 2013).

The other kind of habitat destruction, dam or weir, for example, the Atlantic

sturgeon depends on channel habitats for all life stages and on healthy freshwater

(43)

26

habitats for reproduction; such biological needs are in direct conflict with human

activities, such as dredging and dam construction, which alter habitats or reduce water

quality (Secor, 2000). And the other example, the anadromous members of shad,

herring, and menhaden are threatened by the addition of dams, which can prevent

them from reaching their spawning grounds. If positioned in key locations, water

withdrawal facilities—such as reservoir intakes—may pose a threat to freshwater

spawners in terms of egg and larval losses (Amanda, 2009).

The Japanese eel is a catadromous species, it will migrate into rivers for growth,

but when there are artificial buildings such as dams on the river, they will block the

river channel and the eel can-not migrate successfully (曾等, 2012). Furthermore,

many artificial buildings like harbors, aquaculture areas and riverbanks built by land

accretion or replacement of original natural habitats may cause eels to have less food

sources. For this we know habitat loss could cause eel stocks to decrease.

The other factors of eel resource decline

Except the habitat loss, there are other three main factors affecting the eel

resource, pollution, climate change and overfishing. Water quality changes associated

with increased levels of nutrients, sediments, and contaminants. For example, nutrient

loading leads to algal blooms, which can decrease the concentration of dissolved

(44)

27

oxygen (DO) in the water. Low DO can reduce the amount of suitable habitat for fish

and can impair fish growth and reproduction; and the factor, low DO, may effect eel

too (Amanda, 2009). Otherwise, the effects of persistent pollutants combined with the

eel's unusual life cycle may cause the decline in the eel population in northern Europe

in recent decades (Larsson, 1991); the other example, when the eel expose in the

contaminated environment, the strongly polluted eels detoxify less efficiently, have a

lower condition and might be less successful spawner (Feunteun, 2002).

About the climate change, global warming has affected the stability of the hence

produced shifts in plankton communities and food web structures. Two potential

sources of nutrition have been proposed for eel larvae; dissolved organic matter

(DOM) and particulate organic matter (POM) in the form of zooplankton fecal pellets

and larvacean houses (Otake, 1993; Mochioka, 1996; Pfeiler, 1999). Marine snow has

also been proposed as a potential source of nutrition (Knights, 2003); those primary

production has been considered to be a good proxy for leptocephali food (Bardonnet,

2005). Thus, recruitment declines in Japanese eel may also have been due to

starvation–advection problems (Karl, 2001; Knights, 2003).

There is also a factor about the climate change, the ENSO, Kimura (2001)

showed a certain synchrony between Anguilla japonica recruitment and salinity fronts

driven by ENSO in the Japanese eel spawning area; Kim (2007) also demonstrated

(45)

28

that the changing oceanic conditions associated with climate change have resulted in

decreased recruitment of Japanese eel.

The factor of eel resource decline worth to be discussed is overfishing. Tzeng

(1986) indicated that Japanese eel elvers have been overfishing for aquaculture in

Asian countries, thus the eel population is obviously decreased. Furthermore, Knight,

(2003) has also inferred that Japanese eel populations (and escapement of pre-

spawning silver eels) have been affected by overfishing.

The annual catch of glass eel in the 1970s in Japan about 80.6 tons and 334096

individual in Fulung in the period 1984~1995 on average; after that time, the catch

cleared showed a decrease although it was fluctuating. Annual catches in some local

fishing areas showed nearly synchronous fluctuations. This fluctuation may be caused

in part by oceanic current conditions (Kimura, 2001). On the other hand, the average

catch in Japan in the past 5 years were 6.6 tons and in Fulung in the period

2006~2013 is 14190 eels; this value was only 8% of the average catch in Japan in

1970s and was only 4% in Fulung in the period 2006~2013. It may be very difficult to

explain this rate of decline for about 30 years by only dynamic oceanic environmental

conditions (Tatsukawa, 2003). Because of some river fishermen have expressed

concern that the decline in catch might be caused by overfishing of glass eels year by

year. Furthermore, previous literature assumed that water pollution might have

(46)

29

affected the survival of glass eels too.

Although many factors affect eel stocks, unfortunately, it is difficult to separate

these potential factors. Thus we could not identify the main factor or the percentage of

each factors contributing to eel resource decline (Tzeng, 2004).

(47)

30

Conclusion

This study has demonstrated that the HQI of East Asia declines 25%. However,

the eel resource changes in Japan and Fulung of Taiwan are -92% and -96%,

respectively. Thus, although the habitat destruction should contribute to eel resource

decline to some extent, other factors such as water pollution, overfishing, and climate

change may also be important factors for the decreasing of the eel resource.

(48)

31

References

Abdullah, H. M., Mahboob, M. G., Banu, M. R., Seker, D. Z. (2012). Monitoring

the drastic growth of ship breaking yards in Sitakunda: a threat to the coastal

environment of Bangladesh.

Environmental Monitoring and Assessment. 185,

3839-3851.

Alphan, H. (2003). Land use change and urbanization in Adana, Turkey. Land

Degradation and Development. 14, 575-586.

Amanda, H., Julia, E., Fabrizio, M. C., (2009). Fisheries of the York River System.

Virginia Institute of Marine Science Gloucester Point. VA 23061 U.S.A.

Aoyama, J. (2009). Life history and evolution of migration in catadromous eels

(genus Anguilla). Aqua-bioscience monographs. 2, 1-42.

Avise, J. C., Helfman, G. S., Saunders, N. C., Hales, L. S. (1986). Mitochondrial

DNA differentiation in North Atlantic eels: population genetic consequences of

an unusual life history pattern. Proceedings of the National Academy of Sciences.

83, 4350-4353.

Bardonnet, A., Riera, P. (2005). Feeding of glass eels (Anguilla Anguilla) in the

course of their estuarine migration new insights from stable isotope analysis.

Estuar Coast Shelf Science. 63, 201-209.

Casselman, J. M. (2003). Dynamics of resources of the American eel, Anguilla

rostrata: Declinig abundance in the 1990s. Eel biology, Springer, Tokyo. 255-

(49)

32

274.

Castilla, G., Larkin, K., Linke, J., Hay, G. J. (2009). The impact of thematic

resolution on the patch-mosaic model of natural landscapes. Landscape Ecology

24, 15-23.

Chua, T. E. (1992). Coastal Aquaculture Development and the Environment The

Role of Coastal Area Management Marine. Pollution Bulhtin. 25, 1-4. pp. 98 103 Daily, G. C. (1997). Nature’s Services, Island Press.

Defries, R. S. (1999). Combining satellite data and biogeochemical models to

estimate global effects of human-induced land cover change on carbon emissions

and primary productivity. Global Biogeochem. 13, 803-815.

Dekker, W. (2003). Status of the European eel stock and fisheries. Eel biology,

Springer, Tokyo. 237-254.

ESRI, (2008). ArcMap 9.3. Environmental Systems Research Institute, Redlands,

California, USA.

Fei, Y. (2005). Land cover classification and change analysis of the Twin Cities

(Minnesota) Metropolitan Area by multitemporal Landsat remote sensing.

Remote Sensing of Environment. 98, 317-328.

Feunteun, E. (2002). Management and restoration of European eel population

(Anguilla anguilla): an impossible bargain. Ecological Engineering. 18, 575-591.

(50)

33

Goodwin, K. R., Angermeier, P. L., Orth, D. J. (1999). Assessing impacts of

hydropower dams on upstream migration of American eel. Final Report to the

Virginia Department of Game and Inland Fisheries, Richmond, Virginia.

Huang, S. L., Chang, M. Y., Wang, Y. T., Tzeng, W. N. (2013). Adverse impacts of

urbanization on the diversity integrity of fish larvae and juveniles in a Taiwan

river estuary. (Unpublished)

Jeremy. (2003). From space to species: ecological applications for remote sensing.

Ecology and Evolution. 18, 414-423.

Jiao, J. J., (2001). Analytical Studies on the Impact of Land Reclamation on Ground

Water Flow. Ground Water. 39, 912-920.

Karl, D. M., Bidigare, R. R. (2001). Letelier RM. Long-term changes in plankton

community structure and productivity in the North Pacific Subtropical Gyre: the

domain shift hypothesis. Deep Sea Res Part II, Topical Studies in Oceanography.

48, 1449-1470.

Kerr, J. T., Ostrovsky, M. (2003). From space to species: ecological applications for

remote sensing. Trends in Ecology and Evolution 18, 299-305.

Kim, H., Kimura, S., Shinoda, A. Kitagawa, T., Sasai, Y., Sasaki, H. (2007) Effect

of El Nino on migration and larval transport of the Japanese eel (Anguilla

japonica). Journal of Marine Science. 64, 1387-1395.

(51)

34

Kimura, S., Inoue, T., Sugimoto, T. (2001). Fluctuation in the distribution of low-

salinity water in the NEC and its effect on the larval transport of the Japanese eel.

Fish Oceanography 10, 51-60.

Kimura, S., Itakura, H. (2012). Environmental characteristics of the Japanese eel

migration from spawning grounds to nursery grounds. The 15^th East Asia eel

resource consortium

Knights, B. (2003). A review of the possible impacts of long-term oceanic and

climate changes and fishing mortality on recruitment of anguillid eels of the

Northern Hemisphere. The Science of the Total Environment. 310, 237-244.

Krueger, W. H., Oliveira, K. (1999). Evidence for environmental sex determination

in the American eel, Anguilla rostrata. Environmental Biology of Fishes. 55,

381-398.

Larsson, P. (1991). Factors Determining the Uptake of Persistent Pollutantsin an Eel

Population (Anguilla anguilla). Environmental Pollution. 69, 39-50.

Machut, L. S., Limburg, K. E., Schmidt, R. E., Dittman, D. (2007). Anthropogenic

impacts on American eel demographics in Hudson River tributaries, New York.

American Fisheries Society. 136, 1699-1713.

Merem, E. C., Twumasi, Y. A. (2008). Using spatial information technologies as

monitoring devices in international watershed conservation along the Senegal

(52)

35

River basin of West Africa. International Journal of Environmental Science and

Public Health. 5, 464-476.

Miller, M. J. (2009) Ecology of Anguilliform leptocephali:Remarkable transparent

fish larvae of the ocean sruface layer. Aqua Bioscience Monographs. 2, 1-94.

Mochioka, N., Iwamizu, M. (1996). Diet of anguilloid larvae: Leptocephali feed

selectively on larvacean houses and fecal pellets. Marine Biology Research. 125,

447-452.

NTNU, Holistic Education Proiect. http://hep.ccic.ntnu.edu.tw/browse2.php?s=905

Smith, M. W., Saunders, J. W. (1955). TheAmerican eel in certain freshwaters of the

maritime provinces of Canada. Journal of the Fisheries Research Board of

Canada. 12, 238-269.

Ogden, J. C. (1970). Relative abundance, food habits, and age of the American eel,

Anguilla rostrata (LeSueur), in certain New Jersey streams. Transactions of the

American Fisheries Society. 99, 54-59.

Otake, T., Nogami, K., Maruyama, K. (1993). Dissolved and particulate organic

matter as possible food sources for eel leptocephali. Marine Ecology Progress

Series. 92, 27-34.

Paolo, T., Maria, B., Walter, B., Marco, C., Matias, C. R., Sofia, B. L. (2013). River

pollution remediation monitored by optical and infrared high-resolution satellite

利用衛星遙測技術研究近40年來棲地破壞對日本鰻資源量的影響

國立臺灣大學生命科學院漁業科學研究所 碩士論文

Institute of Fisheries Science College of Life Science National Taiwan University

Master Thesis

利用衛星遙測技術研究近 40 年來棲地破壞對日本鰻資源量的 影響

Study of the relationship between Anguilla japonica resource and habitat destruction by satellite remote sensing in the past 40 years

陳健澤 Jian-Ze Chen

指導教授：韓玉山 博士 Advisor: Yu-San Han, Ph.D.

中華民國 102 年 7 月

July 2013

致謝

中文摘要

Abstract

Contents

Abstract ... iii

Contents ... v

Figure legend ... viii

Table contents ... xi

Introduction ... 1

Materials and methods ... 6

Result ... 9

Discussion... 22

Conclusion ... 30

References ... 31

Figure legend

Table contents

Introduction

Anguilla japonica, European eel A. Anguilla and American eel A. rostrata, has been

Jiao,

2001)

Anguilla japonica resource and habitat destruction during the past 40 years.

Materials and methods

Study area

japonica catching area of Taiwan (Fig.4).Take Minjiang River (閩江), Pearl River (珠

Landscape image collection and processing

Remote sensing image classification

Eel resource data

Result

Remote sensing image

Japan:

1.Ten-ryu River (天龍川)

2. Ohyodo River (大淀川)

3. Ni-yodo River (仁淀川)

4. To-ne River (利根川)

Korea:

1. Han River (漢江)

2. Geum River (錦江)

3. Yeongsan River (榮山江)

4. Nakdong River (洛東江)

Taiwan:

1. Danshui River (淡水河)

2. Lanyang River (蘭陽溪)

3. Zhuoshuei River (濁水溪)

4. Kaoping River (高屏溪)

China:

1. Minjiang River (閩江)

2. Pearl River (珠江)

3. Qiantang River (錢塘江)

4. Yangtze River (長江)

Eel resource data

Discussion

Eel catch data

Habitat quality index, HQI

The long term habitat change in East Asia

Habitat destruction

The other factors of eel resource decline

Conclusion

References

Environmental Monitoring and Assessment. 185,

3839-3851.

rostrata: Declinig abundance in the 1990s. Eel biology, Springer, Tokyo. 255-

japonica). Journal of Marine Science. 64, 1387-1395.

國立臺灣大學生命科學院漁業科學研究所碩士論文

利用衛星遙測技術研究近 40 年來棲地破壞對日本鰻資源量的影響

指導教授：韓玉山博士 Advisor: Yu-San Han, Ph.D.