长江上游生物完整性指数的年际变化.

长江上游生物完整性指数的年际变化

1朱迪2*常剑波

1 中国科学院水生生物生所

2 水利部中国科学院水工程生态研究所

*通讯作者

摘要根据长江上游特有鱼类区系而建立的一个包括5个不同类别的12个参数的适应性的生物完整性指数（IBI）。1997-2002年，每年在上游四个监测站点（宜昌－YC，合江－HJ，木洞－MD，宜宾－YB）开展商业性渔获物调查。包括未来的三峡库区，覆盖约1000KM的江段的四个监测站被选作代表受三峡大坝影响的长江上游地区。另外，历史数据通过跟近期的现场调查数据的对比，来反映流域的一些变化。对这四个江段每年的生物完整性进行计算，并划分为不同的等级，以此表明时空变化的情况。我们观察到自1997年以来，在四个江段的生物完整性指数值逐渐减小。因为所有的数据是在三峡水库蓄水以前收集的，该阶段影响长江上游关键因子显然应该是人类活动，特别是过度捕捞，而不是大坝建设的原因。

关键词鱼类群聚，生物完整性指数（IBI），三峡大坝（TGD），长江上游

简介

Karr（1981）最初提出的，Karr等（1986）创建的生物完整性指数最初是被应用在美国(Karr 1999a; Karr 1986; Karr 1999b)，现在已经日益被应用到别的地区，e.g. 欧洲((Simon 1999))。很多不同的生物类群被用作环境质量的指示物种。藻类、底栖无脊椎动物和鱼类是生物监测的经典种类(Matthews et al. 1982; R.F. Van Dolah 1999; Van Dolah 1999)。鱼类群聚被认为是评价河流生物完整性理想的物种，因为它们的高度的公众认识度，食物链的位置和对水质的高度敏感性(Karr 1981; Karr 1986)。

人类影响，例如，水化学或则自然栖息地的改变，通过破坏其结构和功能而改变鱼类群聚。鱼类群聚的改变可以被监测到，通过改变群落的各个成分，功能群落，物种多样性和相对丰度（Wootton，1990）。鱼类生物完整性指数（F－IBI）最初被应用于中西部的溪流和

河流(Karr 1981)，由于地区性的调整和校正，已经成为一个多参数指数家系，因为不同地区的不同的河流以及它们特有的鱼类群落(Kesminas and Virbickas 2000)。

IBI作为水状况评价的可靠的工具，已经在世界范围内被广泛应用和接受(Novotny et al. 2005)。虽然在IBI方面已有很多很突出的工作，但是在中国长江流域继续这项工作对于我们仍然有很重要的意义。本研究的目的是：a）发展长江上游的鱼类指示种的潜在参数b）量化三峡大坝建设的前六年鱼类群落的差异c）为未来水质评价提供一个参照基线。

材料和方法：

研究区域：

长江是世界第三大河流，全长6300公里，流域面积180 000平方公里。长江以其物种多样性和丰富度，和鱼类资源最大的组成部分而著称(Chang 1999; Wu 2003; Young 2003)。多数水资源受到水污染、筑坝和过度捕捞等人类活动的干扰而有不同比例的退化，过度捕捞现在已经成为中国内陆水体面临的主要问题(Chang J B 1999; Chang 1999)。三峡大坝（TGD）改变和阻断面积约58 000平方公里区域而形成一个1808平方公里的水库，而将展示出人类强大力量。1981年葛洲坝（在TGD下游38公里处）的构建已经导致长江上游洄游鱼类种群的锐减，特别是三种特有的古老鱼类，中华鲟(Acipenser sinensis)，湖鲟(A. dabryanus)和白鲟(Psephurus gladius)(Dudgeon 2000; Xie 2003; Young 2003)。

研究江段在长江干流大约1040公里长。在不同的江段设立4个监测站点，从上游到下游的顺序，分别是宜宾（YB）、合江（HJ）、木洞（MD）和宜昌（YC）（图1）。宜宾站位于宜宾县，监测江段包括金沙江下游约21公里的江段。木洞站点在距离重庆约50公里的木洞镇，位于三峡库区的库尾，监测江段30公里长。合江站点在四川省合江县，监测江段60公里长。宜昌站点位于宜昌市，监测江段是从葛洲坝到古老背约25公里的江段。

缺图？

图1 长江上游研究区域的选择和四个监测站点的分布

长江退化的模式

长江在过去几十年中经历了鱼类群落方面的巨大的改变（Chang 1999）。文献的记载历史案例反映了环境变化的影响。大体上看，主要有四种类型的人类活动导致环境发生了退化：大型水利工程建设，随着人口和工业的发展导致污染物的增加，农业的发展和造田，以及酷渔滥捕(Chang 1995; Young 2003)。众所周知，栖息地环境的改变，生态系统状态的由好到坏的转变深深地影响了鱼类种群，甚至那些定居性种类(Young 2003) 。

生境调查方法

在每个江段，一些生境特性，如平均河宽、深度、流速、水温、融氧和电导率，以及生境结构（底质、激流、深坑水湾、植物覆盖）等被调查和记录。水文学和天气状况由当地的水保局和气象部门提供。

数据收集和分析

自19世纪50年代以来，就开展了覆盖了长江不同江段的长江渔业调查和一些现场调查的工作，这些数据给我们的研究提供了非常重要的基础信息，这对于评价人类活动的生态影响具有很大的价值。中科院水生生物研究所TGD监测数据库中获得了1997-2002监测数据。大部分数据来自现场调查，其它的来自文献析出。从1997到目前，每年5-7月、9-12月期间在长江上游的四个监测站点开展两次调查。每次调查时间约20天左右。从渔船上和市场上收集渔获物信息，在每个监测站点根据随机的原理抽样调查商业捕捞渔船。渔船信息，网具和捕捞江段等信息同时被记录下来。

长江上游的渔获物调查中发现2种主要渔具(刺网和延长钓)，以及其它渔具（撒网，电捕鱼机，箍网和围网）。刺网的网眼在20-250mm之间不等，由于网具的选择性，也意味着渔获物的不同大小组成。延长钓通过或者无饵的鱼钩诱捕鱼类，或长或短，能捕到不同水层的鱼类。

假定追求商业利益最大化的渔民努力捕鱼，收集所有的渔获物并计数。每一尾标本被测量、称重、加上标签并保存在福尔马林溶液中。按比例抽取一部分鱼类样本，解剖检查寄生虫和畸形并获得详细的信息。实验室内，鱼类种类组成和营养结构被计算。根据获得鱼类数据，总的丰度可以被估算出来。

期望值的获得

期望值是评价水生系统健康或质量的关键因素(Karr 1981; Karr 1986)。期望值的确立建立在识别极少受到干扰的地点，它能代表最好的物理、化学和生物状况。本文中的期望值是从历史数据中析出的，来源于中科院水生所和部分文献析出数据加上我们自己调查的数据。

结果

鱼类群落

在四个监测江段出现的鱼类，种类数量记录在表1: YB江段有97种，HJ江段120种，MD江段91种，YC江段116种。上游的鱼类群落特性包括结构、耐受性、主要功能类型和出现江段被描述在表1中。

表1

缺表？

发展适合长江上游的IBI

根据原始的IBI和适应性应用的IBI确定候选参数(Hughes 1999; Lyons 2000; Oberdorff 1992)。考虑多种可能的IBI参数，包括一些已经存在湖泊IBI参数(Zhu 2004)和其它由于上

游独特的鱼类区系而使用的参数(Chang 1999; Young 2003)。早期的研究中使用的一些广泛适用的和较为稳定的参数以及和环境退化相关的参数被初步确定为本研究的参数。我们得到12个能最好的表达鱼类群聚特征的参数，详细描述在下面（表2）。

种类丰度和组成：不同站点种类的数量是衡量跟捕鱼努力程度相关的丰度比较可靠方法。我们根据Hughes和Oberdoff（1999）而不是Karr（1981），本参数采取本地种种类的数量而不是所有种类的数量，这是因为丰度的期望值包括所有的本地种。本地种的数目衡量生物多样性的尺度，随着水域中外来种入侵给环境带来不受欢迎侵扰的增加而降低。在长江中，鲤科、鳅科和鲿科鱼类是常见种类，有时包括在某种苛刻的条件下反而繁茂的种类（Zhu 2004）。

耐受/非耐受种：因为在多数河流的渔获物中常见的种类因而被认为是耐受种类，特别是鲫被认为对多种污染物表现出耐受性。这个参数是“非耐受性种类数量”的补充，跟Hughes and Oberdorff’s（1999）推荐的“耐受种的比例”具有一样的效果。象Karr（1981）最初使用的“绿鳞太阳鱼的比例”，它把低等水质和中等水质区分开。

非耐受种(Karr 1986; Lyons 2000)是先前必较多而现在因为环境退化而偶见的种类，也对多种类型的环境胁迫敏感在环境退化（悬浮物过多，温度升高、沉积增加，溶氧降低）存在时消失、当环境修复后又恢复的种类。渔获物中科的数量被选择衡量非耐受性鱼类，在生态系统良好的状态下，其数量增加。

营养功能类群：挑选的营养功能类群包括底栖昆虫食性、杂食性和鱼食性类群。这些划分的标准并没有被严格的界定。Karr（1981）推荐的杂食性种类的成鱼主要吞食植物和动物评价胁迫状况下食物网的中断。我们选择杂食性鱼类主要因为它们对恶劣条件具有耐受性。同样地，Yoder和Smith（1999）从丰富度参数中减去耐受种。顶级肉食性种类成鱼主要捕食其它鱼类和大型底栖动物用于评价营养类群多样性和关键种的缺失(Lyons 2000)。底栖昆虫食性用于评价第二级生产者环节的中断，因为无脊椎动物对河流中大部分有机物的处理起主要作用。破坏无脊椎动物的生物量和组成也能推测到该食性鱼类的减少。这些参数在IBI评价中的应用比较广泛(Hughes 1999)。

丰富度：单位努力捕捞量（CPUE）可以代替初始IBI中使用的“相关个体数量的多少”衡量来相对丰富度(Karr 1986; Zhu 2004), 这可以评价河流中鱼类种群的相对大小。差的水质预示着在相似的水域使用相同捕捞技术CPUE的值低于好水质的河流。它是一个可靠的参数，可以被广泛应用评价渔业资源。

非本地种类比例：如果没有干扰发生鱼类群聚被认为具有生物完整性。非本地种类的存在被认为是一个干扰因子(Ganasan 1998; Lyons 2000)。外来种类可能是由于养殖渔场的然逃逸，这些种类通过捕食和竞争被认为能改变本地种的群聚结构，有时能造成本地种的灭绝。因而，生物完整性指数应该跟存在的外来种的数量成负相关。

DELT种类的比例：在良好的水生生态系统中，畸形的鱼类很少，但是随着人类活动的加剧，其数量可能增加，所以我们继续使用该参数来评价鱼类个体健康状况。鱼体畸形包括畸形、鳍损伤、鱼体受损、肿瘤、疾病和寄生虫（DELT）。鱼体畸形反映亚致死的环境胁迫、间歇的胁迫、行为的胁迫或者化学污染的底质（Lyons 2000）。

表2 适合长江上游的IBI参数和赋分标准

缺表？

IBI值

在统计上，IBI的得分在六年的范围内降低了，多数IBI值被划分为好和一般两个等级

（图2）。线形回归分析的结果显示生物完整性具有明显降低的趋势，这种趋势的可能原因尚未确定，但是日益增加的人类活动肯定是非常重要的影响因子。

图2 长江上游四个监测站不同年度（1997-2002）的IBI得分以及IBI得分的线性回归结果

缺图

讨论

天然大河的物理、化学和生物特性从上游到河口是逐渐改变的，这些改变可以用纵向地理生物分布的概念来描述(Bram 2003)。多数鱼类只能在整个流域的一部分江段发现适宜生存的条件。19世纪的鱼类学家们已经把这个发现作为河流中生物地理分布体系的基础(Hol ?ík 1989)，按照这个理论，整个长江可以分成几个不同鱼类区带。长江上游被认为位于同一个鱼类区带中，具有特殊的生态学特征（Chang 1999）。在某个特别的地区，一个合适的监测人类干扰的方法是通过比较目前和期望值条件下生物群落的结构和组成的状况。该区域的参照状况是各个参照点和理想模型的期望值的综合，它可能包括历史状况的资料（Lyonns 2000）或者生态学原理的推断。

多数研究者，在淡水生态系统领域，使用鱼类IBI方法评价人类活动引起的环境退化(Araujo 2003; Francisco Gerson Araujo 2003; Joy 2004; Michael 2004; Paller 2000)。Karr最初（1981）在对IBI描述中，和他后面的(Karr 1999a; Karr 1986)论文中都强调了代表性采样的重要性。本文我们使用渔获物调查的方法代替标准化采样方案（Karr 1986）主要是因为下列原因：

长江的规模造成了采样的困难，由于人力物力的限制，我们只能对很小区域进行采样，而鱼类迁移性（水平和垂直）很大，在有限的时间内一些种类和很多个体由于深度而不能采集到（Simon 1999）。

我国淡水鱼类种类较多，但由于过度捕捞，各种的数量很少，短时间、小范围的现场采样不能代表整个研究河段。

对于大河来说，渔业渔获物的收集是鱼类群聚评价的首选方法。不同江段在不同时期积累的渔获物统计资料，有利于开展不同江段生物完整性时空变化的比较。

尽管采样方法没有标准化造成现有数据结构的局限，会使评价结论包含有一些不确定性，但现场采样的数据同样可能包含误差。因此确信在六年连续监测获得的数据对于反映长江鱼类群聚的真实情况是足够的；也同样确信在较长时期内获取的数据对于初次使用IBI评价大河生物完整性的时空变化也是十分充足的。结果的可靠性也为可持续管理提供了科学的依据。建立在渔获物调查上的IBI需要在不同生态地区进性进一步的检验。

该研究显示上游的生物完整性在1997-2002间急剧下降。河流筑坝对水生生态系统功能的影响已经被很多文献报道过(Baxter 1997; Young 2003)。大坝和水库改变了流态，从自由流动到滞留，阻碍鱼类越冬洄游和索饵洄游。上游干流上的TGD是世界上最大的水利发电站。TGD 的水利调节将改变水温和水化学特性，这反过来又会影响生物和化学过程的速率。

应该提前采取一些弥补措施，这些措施应该包括：a) 实施渔业管理以减少酷渔滥捕；b）建立过鱼设施，如，鱼道；c) 建立保育和保护中心d) 实施人工繁殖和放流。

References

Araujo, F. G., I., Fichberg, B. C. T., Pinto ,M. G., Peixoto. (2003). "A preliminary index of biotic integrity for monitoring the condition of the Rio Paraiba do sul, southeast Brazil." Environmental Management, 516-526.

Baxter, R. M. (1997). "Enviromental effects of dams and impoundments." Annual Review of Ecology and Systematics, 8, 255-283.

Bram, G. W. A. a. P., H. N. (2003). Fish zonations and guilds as the basis for assessment of ecological integrity of large rivers, Kluwer Academic Publishers, Netherlands.

Chang J B, W. X. C. (1999). "Fishery Significance of the River-Communicating Lake and Strategies for the Management of Fish Resources." Resources and Environment in the Yangtze Basin, 8(2), 153-157.

Chang, J. B., G. H., Zhang, Y. G., Xu. (1995). Chapter 8 Fish and Fisheries, Science Press, Beijing. Chang, J. B., W. X., Cao. (1999). "Fishery Significance of the River-Communicating Lake and Strategies for the Management of Fish Resources." Resources and Environment in the Yangtze Basin, 8(2), 153-157.

Dudgeon, D. (2000). "Going with the flow: large-scale hydrological changes and prospects for riverine biodiversity in tropical Asia." Bioscience, 50, 793-806.

Francisco Gerson Araujo, I. F., Benjamin Carvalho Teixeira Pinto and Magna Galvao Peixoto. (2003). "A preliminary index of biotic integrity for monitoring the condition of the Rio Paraiba do sul, southeast Brazil." Environmental Management, 516-526.

Ganasan, V., Hughes, R. M. (1998). "Application of an index of biological integrity (IBI) to fish assemblages of the rivers Khan and Kshipra (Madhya Pradesh). India." Freshwater Biology, 40, 367-383.

Hol?ík, J. (1989). The Freshwater Fishes of Europe, AULA-Verlag, Wiesbaden.

Hughes, R. M., Oberdorff T. (1999). Applications of IBI concepts and metrics to waters outside the United States and Canda, Levis Press, Boca Raton, Florida.

Ichthyologic Department, H. I. o. H., Chinese Academy of Sciences. (1976). The Fishes of Yangtze River, Science Press.

Joy, M. K., R. G., Death. (2004). "Application of the Index of Biotic Integrity Methodology to New Zealand Freshwater Fish Communities." Environmental Management, 34(3), 415–428.

Karr, J. R. (1981). "Assessment of biotic integrity using fish communities." Fisheries, 6, 21-27. Karr, J. R., E.W., Chu. (1999a). Restoring life in running waters: better biological monitoring, Island Press, Washington, D.C.

Karr, J. R., Fausch, K. D., Angermier, P. L., Yant, P. R., Schlosser, I. J. (1986). "Assessing biological integrity in running waters: a method and its rationale."

Karr, J. R. E. W. C. (1999b). Restoring life in running waters: better biological monitoring, Island Press, Washington, D.C.

Kesminas, V., and Virbickas, T. (2000). "Application of an adapted index of biotic integrity to rivers of Lithuania." Hydrobiologia, 422-423(0), 257-270.

Lyons, J., Altagracia Guti′errez-Hern′andez, Edmundo D′iaz-Pardo, Eduardo Soto-Galera, Martina Medina-Nava & Ra′ul Pineda-L′opez. (2000). "Development of a preliminary index of biotic integrity (IBI) based on fish assemblages to assess ecosystem condition in the lakes of central Mexico." Hydrobiologia, 418(1), 57-72.

Matthews, R. A., Buikema, J. A. L., Cairns, J. J., and Rodgers, J. J. H. (1982). "Biological monitoring : Part IIA--receiving system functional methods, relationships and indices." Water Research, 16(2), 129-139.

Michael, K. J., Russiell, G. Death. (2004). "Application of the Index of Biotic Integrity Methodology to New Zealand Freshwater Fish Communities." Environmental Management, 34(3), 415–428. Novotny, V., Bartosova, A., O'Reilly, N., and Ehlinger, T. (2005). "Unlocking the relationship of biotic integrity of impaired waters to anthropogenic stresses." Water Research, 39(1), 184-198. Oberdorff, T., R. M., Hughes. (1992). "Modification of an index of biotic integrity based on fish assemblages to characterize rivers of the Seine Basin, France." Hydrobiologia, 228(2), 117-130. Paller, M. H., M.J.M., Reichert , J. M., Dean, J.C., Seigle (2000). "Use of fish community data to evaluate restoration success of a riparian stream." Ecological Engineering, 15, 171-187.

R.F. Van Dolah , J. L. H., A.F. Holland , J.S. Rosen,T.R. Snoots (1999). "A benthic index of biological integrity for assessing habitat quality in estuaries of the southeastern USA." Marine Environmental Research, 48, 269-283.

Simon, T. P., Jankowski, R., and Morris, C. (2000). "Modification of an index of biotic integrity for assessing vernal ponds and small palustrine wetlands using fish, crayfish, and amphibian assemblages along southern Lake Michigan." Aquatic Ecosystem Health and Management, 3(3), 407-418. Simon, T. P., Sanders, R.E. (1999). Applying an index of biotic integrity based on great river fish communities: considerations in sampling and interpretation, CRC Press, Boca Raton, Florida.

Van Dolah, R. F., J.L. Hyland , A.F. Holland , J.S. Rosen,T.R. Snoots (1999). "A benthic index of biological integrity for assessing habitat quality in estuaries of the southeastern USA." Marine Environmental Research, 48, 269-283.

Wootton, R. J. (1990). Ecology of teleost fishes., Chapman and Hall, London.

Wu, J., Huang, J., Han, X., Xie, Z., Gao, X. (2003). " Three-Gorges Dam—Experiment in Habitat Fragmentation?" Science, 300, 241-248.

Xie, P. (2003). "Three-Gorges Dam: Risk to Ancient Fish." Science, 302, 1149-1150.

Yoder, C. O., M.A Smith. (1999). Using fish assemblages in a state biological assessment and criteria program: essential concepts and considerations, CRC Press, Boca Raton, FLorida.

Young, S. P., Chang, J., Leck, S., Cao, w., Brosse, S. (2003). "Conservation strategies for endemic fish species threatened by the three gorges dam." Conservation biology, 17(6), 1748-1758.

Zhu, D. C., J. (2004). "Evaluation on temporal and spatial changes of biological integrity for shallow lakes in the middle reach of the Yangtze River." Acta ecologica sinica., 24(12), 2761-2767.

Zhu, S. Q. (1995). Synopsis of Freshwater Fishes of China, Jiangsu Science and Technology Publishing House, Nanjing.

Annual variations of index of biotic integrity in the

upper Yangtze River

Di Zhu1 and Jianbo Chang2*

1 Institute of Hydrobiology, Graduate School, Chinese Academy of Sciences, 7# southern road of East Lake, Wuhan, Hubei Province,

China 430072

2 Institute of Hydroecology, Ministry of Water Resources and Chinese Academy of Sciences, 578# Xiongchu Avenue, Wuhan,

Hubei Province, China 430079

* Correspond author

Abstract An adaptive index of biotic integrity (IBI), including twelve metrics in five categories, was created according to the special fish faunas in the upper Yangtze River. Surveys on commercial fisheries were undertaken annually in four sections of the main channel of the Upper Yangtze (Yibin-YB, Hejiang-HJ, Mudong-MD, and Yichang-YC) from 1997 to 2002. These four monitoring sections covering a 1000 km reach including the future Three Gorges Reservoir were selected to represent the upstream area influenced by the construction of TGD. In addition, historical data were used to show changes in the watershed by comparison with field investigations recently. The biotic integrity of the four sections were calculated and classified into different levels annually for recognizing its spatial and temporal variations. It was observed that IBI scores were becoming lower diminishingly since 1997 in all the four sections. Because all the data were collected before the impoundment of the Three Gorges Reservoir, it is obvious that human activities, especially over-fishing, must be crucial factor instead of damming in the upper Yangtze River in that period.

Key words Fish assemblages, Index biotic integrity (IBI), Three Gorges Dam (TGD), Upper Yangtze River

Introduction

The Index of Biological Integrity (IBI) originally developed by Karr (1981) and established by Karr et al. (Karr 1986) had previously been used in the United States (Karr 1999a; Karr 1986; Karr 1999b) and became increasingly adaptive elsewhere, e.g. in Europe (Simon 1999). Many groups of organisms had been used as indicators to estimate environmental quality. Algae, benthic invertebrates and fish were typical species in biological monitoring (Matthews et al. 1982; R.F. Van Dolah 1999; Van Dolah 1999). Fish assemblages were considered to be an appropriate end-point for assessing stream integrity due to their high public visibility, their position in the food chain and high sensitivity to water-quality (Karr 1981; Karr 1986).

Human influences, such as changes in water chemistry or physical habitat modifications, could alter fish assemblages by disrupting their structures and functions. Varieties in fish assemblage could be detected through changes in components of the community, functional groups, species diversity, and relative abundance (Wootton 1990). The fish-IBI was originally developed for small America mid-western streams and rivers (Karr 1981), and had become a family of multi-metric indices that were regionally adapted and calibrated, because rivers of different regions, as well as their fish communities, were distinctive (Kesminas and Virbickas 2000).

The IBI was commonly used and accepted worldwide as a reliable tool to assess water condition now(Novotny et al. 2005). Despite many outstanding works on IBI had already been done(Novotny et al. 2005; Simon et al. 2000)，it was still of utmost importance for us to continue this work in Chinese Yangtze River basin. The objectives of this study were to: a) develop potential metrics of fish indicator for the upper Yangtze River, b) quantify fish assemblage differences in the early six years of Three Gorges Dam (TGD) construction, and c) provide a baseline for future water quality assessment in the upper Yangtze River.

Materials and methods

Study area

Yangtze River is the third longest river in the world, its total length was 6300 km with basin area 180 000 km2. Yangtze River is distinctive in diversity of species and abundance in number, and comprised the largest components of the fish resource(Chang 1999; Wu 2003; Young 2003). Most of water resources are disproportionately degraded by human activities such as water pollution, dam construction and over-fishing, which are now becoming the biggest concerns in Chinese inland waters (Chang J B 1999; Chang 1999). River damming is the most dramatic anthropogenic factor affecting freshwater environments (Baxter 1997; Dudgeon 2000). Three-Gorges Dam (TGD) is going to demonstrate the mighty power of humanity to change and fragment an area of about 58,000 km2 with the formation of a reservoir of 1080 km2 in Yangtze River.The Gezhou Dam (38 km downstream from the TGD) constructed in 1981 has led to sharp declines in the populations of migratory fish occurring in the upper Yangtze River, especially the three endemic ancient fish species, Chinese sturgeon (Acipenser sinensis), River sturgeon (A. dabryanus), and Chinese paddlefish (Psephurus gladius)(Dudgeon 2000; Xie 2003; Young 2003).

This study reach influenced by TGD is 1040 km long in the main channel of the upper Yangtze River. Four monitoring stations are set up at different reaches, from upper to lower, they are Yibing (YB), Hejiang (HJ), Mudong (MD) and Yichang (YC) respectively (Fig.1). Yibin (YB)

station is located in Yibin county, its monitoring reach is 21 km long including the lower reach of the Jinsha River. Mudong (MD) station, located in Mudong town with the distance of 50 km from ChongQin city, is the bottom of TGD Reservoir. The monitoring reach is 30 km long. Hejiang (HJ) station is located in Hejiang county of Sichuan province, its monitoring reach is 60 km long. Yichang (YC) station is located in Yichang city, the monitoring reach is 25 km from Gezhou dam to the Gulaobei.

Fig.1. Study areas selected in the upper Yangtze River in China and distribution of four monitoring stations: Yibing (YB), Hejiang (HJ), Mudong (MD) and Yichang (YC).

Patterns of degradation in Yangtze River

Yangtze River environment and fish assemblages had experienced major changes over the past decades (Chang 1999). Historical cases of well-documented faunas illustrated the impacts of environmental degradation. Holistically, there were four human activities induced environmental degeneration: a) the large damming project, b) pollutions increasing with a bulge in population and industrial development, c) agriculture and farming, and d) overfishing (Chang 1995; Young 2003). It was well known that those environmental changes occurring in the habitat, on the shift of good to poor status of ecosystem deeply influenced the fish populations, even non-migratory ones(Young 2003).

Field Methods for Habitat characterization

In each section, a number of habitat variables such as the mean channel width (m), depth (m), water velocity (m s-1), water temperature (℃), dissolved oxygen concentration (mg l-1), as well as habitat structure (bottom structure, presences of rapids, pits and coves, percent of bottom area overgrown with submerged vegetation) were measured and noted. Data of hydrology and weather were provided by local Water Conservancy Department and Weather Department.

Data collection methods and analyses

Fishery investigations had been conducted in Yangtze River from 1950’s till 2002 covering different reaches, together with the field investigations, these data provided very important baseline information, which were of tremendous value for assessment of ecological effects of human activities. The biological data (1997-2002) were obtained from the TGD monitoring database of the Institute of Hydrobiology (IHB), Chinese Academy of Sciences (CAS). Most of the data were obtained by field investigations and others were derived from literatures. Twice investigations were conducted in May-July and September– December from 1997 till 2002 at the four monitoring stations. The duration of investigation was 20 days every time. The catches were collected from fishing boats and fish markets, the commercial fishing boats were sampled according to the principle of random sampling at each station. The information of the fishing boat, fishing gear and fishing reach were recorded at the same time.

Two main fishing gears (gill net, long-line) among others (casting net, electrofisher, hoop net and

seine) were found in fishery catches investigations in the upper Yangtze River. The mesh sizes of the gill nets ranged between 20 - 250 mm while the size compositions of catches generally portrayed the relative selectivity of the different mesh sizes. Long-line fishing was the capture of fish by entrapment of fishhook with bait or no bait，there was floating and demersal long-line, the fish inhabiting in different depth of river could be caught.

Fishers who pursued maximal commercial benefits were believed to attempt to capture fish. All catches were collected and counted. Every fish specimen was measured, weighted, labeled and conserved in the formalin solution. Referrin g to the “synopsis freshwater fishes of C hina” (Ichthyologic Department 1976) and “The Fishes of Yangtze Rive”(Zhu 1995), fish specimens were identified. Fish were sampled and dissected proportionally to study the parasite and the availability of anomalies and get information in detail. In the laboratory, the proportion of fish species and trophic guilds were calculated. According to fish data obtained, total abundance could be estimated by calculating.

Reference conditions derived

Reference condition was a critical element of assessing the quality or health of the aquatic ecosystem(Karr 1981; Karr 1986). The establishment of reference condition was based on identification of minimally disturbed sites that represent the best physical, chemical, and biological conditions. The reference expectations used in the paper were derived from the historical data, which came from database of IHB, CAS and some came from literatures and unpublished data, including our own observations.

Results

Fish assemblages

Appearing within four river reaches the following numbers of fish species had been recorded: 97 in YB, 120 in HJ, 91 in MD, and 116 in YC (Table 1). The characteristics of fish assemblages of upper Yangtze River had been described including structure, tolerance, main trophic guilds, and appearance in investigations.

Table 1 Occurrence of native fish species in the upper Yangtze River basin with indication of their trophic guilds (BI = benthic insectivores, O = omnivores, C = carnivores, other = except BI/O/C feeding), environmental tolerance and others (in bold = species of commercial interests, E = endemic species, Y=yes, *= presence).

Development of an IBI for the upper Yangtze River

Candidate metrics of IBI h as been identified according to the original IBI as well as IBIs’ applications elsewhere (Hughes 1999; Lyons 2000; Oberdorff 1992). A variety of potential IBI metrics are considered, including many from existed lake IBI (Zhu 2004) and others suggested by the unique fish fauna of Yangtze River(Chang 1999; Young 2003). Preliminary metrics are selected based on their broad applicability and stability across the previous studies and their

consistent related to environmental degradation. 12 metrics, which best represent the fish assemblages’ characteristics and the patterns of degradation in upper Yangtze River, are obtained and described below (Table 2).

Species Richness and Composition The number of species is plotted against the sites, which give a more reliable measure of the species richness adjusted to employee fishing effort. We, therefore, followed Hughes and Oberdorff (Hughes 1999) rather than Karr(Karr 1981) in favoring only native species versus all species for this metric. This is because the total native species in the upper Yangtze River are included in the richness expectations. The number of native species is a measure of biological diversity that typically decreases with increased degradation for the undesirable influence of alien species in waters. Cyprinidae, Bagridae catfishes, and Cobitidae are very common in Yangtze River, sometimes comprising the groups of fishes thriving under such harsh conditions(Zhu 2004).

Tolerant/Intolerant Species These fishes are considered tolerant species because they are very common in the catches of most rivers, especially for the crucian carp are thought most tolerant to many pollutants. This metric is the complement to “number of intolerant species” and is comparable to Hughes and Oberdorff’s percent tolerant individuals (Hughes 1999). Like Karr’s original percent green sunfish metric (Karr 1981), it distinguishes between low and moderate water quality.

Intolerant species(Karr 1986; Lyons 2000) are that previously very abundant but presently occurs only occasionally because of environmental deterioration. Intolerant species are also sensitive to many types of environmental stressed and tend to be absent in the presence of environmental degradation (high suspended solids, increased temperatures and siltation, decreased dissolved oxygen), and are the last to reappear after restoration. Number of families in fishery catches is chosen as a metric of intolerance adversely, it will become more in a better quality of ecosystem.

Trophic Guilds Trophic groups selected include benthic insectivorous, omnivores, and carnivores. Criteria separating these groups are not sharply defined. Adults of omnivorous species proposed by Karr (Karr 1981) eat large proportions of both plants and animals for assessing disruption of the food web by stressors. We choose omnivores because they tolerate poor conditions. Similarly, Yoder and Smith(Yoder 1999) subtracted tolerant species from their abundance metric. Top carnivorous adults eat predominantly other fishes or large invertebrates for assessing loss of trophic diversity and keystone species(Lyons 2000). Percent of benthic insectivorous individuals is used for assessing disruptions in secondary production, because invertebrates are responsible for much of the processing of organic matter in rivers. Disrupted invertebrate biomass and composition is presumably reflected in disruptions in insectivorous fishes. Those metrics are also widely used in IBIs (Hughes 1999).

Abundance Catch per unit effort (CPUE) is a measure of relative fish abundance as a substitute of the relative number of individuals metric in the original IBI (Karr 1986; Zhu 2004), which evaluate fish populations within a stream or river site. Poor water quality will expectedly result in lower CPUE than stream sites of good quality when similar areas and sampling techniques are used. It is so reliable that it is widely used in evaluating the fishery resource.

Percent of Non-native Fish Species Fish assemblages are considered to be biotical

integerty if no disturbance occurred. The presence of non-native fish species is considered as a disturbance factor (Ganasan 1998; Lyons 2000). Alien species are mostly accidental releases from fish farms，these species could alter the structure of native fish assemblages by competition and predation, sometimes extirpate some native species. Consequently, the index of biotic integrity should be negatively correlated with the quantity of non-native fish species present.

Percent individuals with DELT Fish with abnormalities are rare in good ecosystem, but the number of abnormalities will increase with the human activities increases, so we continue to use that metric for assessing conditions of fish individuals. External anomalies includes deformities, eroded fins, lesions, tumors, diseases, and parasites (DELT)(Karr 1986). The presence of external anomalies indicates sublethal environmental stresses, intermittent stresses, behavioral stresses, or chemically contaminated substrates(Lyons 2000).

？

Table 2Metrics and scoring criteria used in the IBI for the upper Yangtze River

IBI values

The IBI scores were becoming lower statistically in the six-year scales and most IBI scores were classed as good and fair quality (Fig. 2). The result of linear regression analysis showed that the obvious degenerative trend in the biotic integrity for some uncertain reasons，but the increased human activities must be very important influencing factors.

Fig. 2. Observed IBI scores for four upper Yangtze River, YC(open circles), MD(squares), YB(left triangles) and HJ(right triangles) are plotted against years between 1997 and 2002. The straight line shows the linear regression between year and IBI scores.

Discussion

The physical, chemical and biological characteristics of natural large river changed gradually along its course from headwaters to estuary. Those downstream changes can be described by longitudinal zonation concepts (Bram 2003). Most fish species find suitable living conditions in only a selected stretch of the entire river. Already in the 19th century ichthyologists used this observation as the basis of a zonation system for river courses (Hol?ík 1989), in which the entire Yangtze River can be divided into several fish zones. The upper Yangtze River was thought within the same fish zone with its own ecological characterizations(Chang 1999). Within a specific region, a suitable way of

analyzing and monitoring the impact of human disturbance on the structure of the fish assemblages was by comparing community structure of the present state with that of the reference situation. The regional-based reference condition can be derived from an aggregate of reference sites or some empirical model of expectations and may include knowledge of historical condition(Lyons 2000) or extrapolation from ecological principles. Most researchers, in the field of freshwater ecosystems, evaluated the environment degradation induced by human activities using the fish-IBI methodology(Araujo 2003; Francisco Gerson Araujo 2003; Joy 2004; Michael 2004; Paller 2000). In his original description of the IBI (1981), as well as in later papers (Karr 1999a; Karr 1986), Karr emphasizes the need for representative samples. In this paper, we used the data from fishery investigations instead of the standardized protocols (Karr 1986) for the following reasons:

1) The size of Yangtze River created many problems, fishes were much more migratory (both longitudinally and horizontally) and at any given time some species and many individuals were inaccessible because of depth (Simon 1999);

2) There were many kinds of fresh water fishes in China, but human activities, particularly over-fishing, decreased the abundance. Field sampling with short time and small percentage of river can not represent the entire reach;

3) Fishery catches collection was a preferred method for the large river in the evaluation of fish assemblages. It had been conducted in four stations as a long-term monitoring method, it benefited the comparison of spatial and temporal changes of the biotical integrity.

Though these data might not be fully validated as the sampling methods had not been calibrated, the field survey itself may also contain some errors. Therefore it was believed that the data of six-year period continuous survey were sufficient enough to reflect the actual conditions of the fish assemblages in the Yangtze River. It was also understood that the data obtained from extensive time series were sufficient enough to allow the first attempt to a modified IBI for evaluation on temporal and spatial changes of biological integrity within a large river. The reliability of the results provided a scientific evidence for sustainable management. The IBI based on the fishery catches surveys need to be further tested and validated in many ecological areas with different fauna.

This research revealed that the biotic integrity of the upper Yangtze River declined dramatically during 1997-2002. The impact of river damming on aquatic ecosystem functioning had been documented (Baxter 1997; Young 2003). Damming and reservoir altered the flow regimes, from free-flowing to stagnant, and obstruct migratory fish for winter or food horizontally. Because all the data were collected before the impoundment of the Three Gorges Reservoir, it is obvious that human activities, especially over-fishing, must be crucial factor instead of damming in the upper Yangtze River in that period. In spite of that, the TGD in the main channel of the upper Yangtze River is still the biggest hydroelectric dam in the world and entirely modified the main channel of the upper Yangtze River. The regulation of the TGD will alter water temperatures and chemistry, which in turn will influence the rates of biological and chemical processes. The reparation measurements should be employed in advance. These include, a) reinforcing the fishing management to reduce the extremely fishing, b) building fish pass, e.g. the fish-ways, c)

constructing the nursery and conservation center; and d)implementing artificial propagation and release.

Acknowledgements