排序和广义线性模型与广义可加模型在植物种与环境关系研究中的应用

摘要/Abstract

摘要： 排序和广义线性模型(Generalized Linear Modd,GLM)与广义可加模型(Goneralized Additive Model,GAM)是研究植物种与环境间关系的重要方法.基于线性模型的排序方法应限定于环境梯度较短的植被数据,而基于单峰模型的排序方法更适用于梯度较长的情况.PCA、CA/RA系列和CCA系列是常用的排序方法.同时进行环境数据和植被数据分析的CCA系列,能清楚地得出植物种与环境间的关系.CCA改进后的DCCA和PCCA,是现今较理想的排序方法.GLM和GAM实质上是用环境变量的高阶多项式来拟合植物种与环境变量的关系.GLM和GAM扩展了植物种与环境变量之间的关系模型,能深入地探讨植物种与环境间的关系.GLM主要是模型决定的,而GAM主要取决于原始数据.一般来说,排序能得出研究区域的主要环境梯度,提供了物种聚集和植物群落的概略描述.GLM与GAM对于深入研究单个植物种与环境间的关系具有优势.在实际研究中,两种方法结合使用能互补不足.

关键词: 沟渠湿地, 挺水植物, 氮磷, 非点源污染物

Abstract: Ordination and GLM/GAM are both important methods to explore the relationship between plant species and environment.Generally,ordination methods,based on linear model,are suitable for dealing with the vegetation within a comparatively short environmental gradient,while those based on unimodal model are more suitable for tackling with the vegetation distributing along a longer gradient.PCA,CA/RA series and CCA series are the most commonly used ordination methods.Involving the environmental data in the analysis,CCA series could demonstrate the relationship between plant species and environment clearly.After modification,DCCA and PCCA are considered to be ideal ordination methods at present.Essentially,General Linear Model (GLM) and General Additive Modal (GAM) were using the high power polynomials of environmental variables to fit the response of plant species to environment.So,the models for delineating the relationship between species and environment have been enhanced through the introduction of GLM and GAM in recent years.GLM is basically driven by model,while GAM is more driven by data.In comparison,ordination can obtain the main environmental gradient and provide the general information of species assembling and plant community in the study area.GLM and GAM have advantages in probing deeply into the relationship between plant species and environment.In practice,the two kinds of methods can yield better results when used in combination.

Key words: Ditch wetland, Emerged plants, Nitrogen and phosphorus, Non-point source pollutants

中图分类号:

Q948

朱源, 康慕谊. 排序和广义线性模型与广义可加模型在植物种与环境关系研究中的应用[J]. 生态学杂志, 2005, (7): 807-811.

ZHU Yuan, KANG Muyi. Application of ordination and GLM/GAM in the research of the relationship between plant species and environment[J]. cje, 2005, (7): 807-811.

参考文献

[1] 王孝安,冯杰,张怀.1994.甘肃马衔山林区植被的数量分类与排序[J].植物生态学报,18(3):271～282.
[2] 王国宏.2002.祁连山北坡中段植物群落多样性的垂直分布格局[J].生物多样性,10(1):7～14.
[3] 朱源,邱扬,傅伯杰,等.2004.河北坝上草原东沟植物群落生态梯度的数量分析[J].应用生态学报,15(5):799～802.
[4] 江洪.1994.川西北甘南云冷杉林的DCA排序、环境解释和地理分布模型的研究[J].植物生态学报,18(3):209～218.
[5] 米湘成,张金屯,张峰,等.1996.山西高原植被与气候的关系分析及植被数量区划的研究[J].植物生态学报,20(6):549～560.
[6] 阳含熙,卢泽愚.1983.植物生态学的数量分类方法[M].北京:科学出版社.
[7] 张利权.1987.瑞典河漫滩草甸植被的数量分类和排序[J].植物生态学与地植物学学报,11(3):171～182.
[8] 张金屯.1991.植被数量分类和排序的发展[J].山西大学学报(自然科学版),14(2):215～224.
[9] 张金屯.1995.植被数量生态学方法[M].北京:中国科学技术出版社.
[10] 张金屯.1998.典范主分量分析及其在山西植被与气候关系分析中的应用[J].地理学报,53(3):256～263.
[11] 张新时.1991.西藏阿里植物群落的间接梯度分析:数量分类与环境解释[J].植物生态学与地植物学学报,15(2):101～113.
[12] 李绍忠.1987.辽东三块石天然次生林的排序[J].植物生态学与地植物学学报,11(4):264～275.
[13] 李斌,张金屯.2003.黄土高原植物群落生态关系研究[J].农业环境科学学报,22(4):471～473.
[14] 沈泽昊,张新时.2000.三峡大老岭地区森林植被的空间格局分析及其地形解释[J].植物学报,42(10):1089～1095.
[15] 沈禹颖,阎顺国,朱兴运.1994.河西走廊盐化草甸主要植物群落分布特点及其土壤环境特征[J].植物生态学报,18(1):95～102.
[16] 郑慧莹,李建东,祝廷成.1986.松嫩平原南部植物群落的分类和排序[J].植物生态学与地植物学学报,10(3):171～179.
[17] 施维德.1983.四川省缙云山森林群落的分类和排序[J].植物生态学与地植物学丛刊,7(4):299～312.
[18] 赵志模,郭依泉.1990.群落生态学原理与方法[M].重庆:科学技术文献出版社重庆分社.
[19] 郭水良,曹同.2000.长白山森林生态系统树附生苔藓植物分布与环境关系研究[J].生态学报,20(6):922～931.
[20] 钱宏.1990.长白山高山冻原植物群落的数量分类和排序[J].应用生态学报,1(3):254 263.
[21] 黄净,韩进轩,阳含熙.1993.长白山北坡阔叶红松林的DCA排序分析[J].植物生态学与地植物学学报,17(3):193～206.
[22] 潘代远,孔令韶,金启宏.1995.新疆呼图壁盐化草甸群落的DCA、CCA及DCCA分析[J].植物生态学报,19(2):115～127
[23] Austin MP. 1985. Continuum concept, ordination methods, and niche theory [J]. Ann. Rev. Ecol. Syst., 16:39～61.
[24] Austin MP. 2002. Spatial prediction of species distribution: An interface between ecological theory and statistical Modeling [J].Ecol. Model., 157:101 ～ 118.
[25] Goodall DW. 1954. Objective methods for the classification of vegetation Ⅲ. An assay in the use of factor analysis [J]. Aust.J. Bot., 3: 304～ 324.
[26] Guisan A, Weisis SB, Weisis AD. 1999. GLM versus CCA spatial modeling of plant species distribution [J]. Ecology, 143: 107～122.
[27] Guisan A, Edwards TCJ, Hastie T. 2002. Generalized linear and generalized additive models in studies of species distributions:Setting the scene [J]. Ecol. Model., 157: 89～ 100.
[28] Hill MO, Gauch HG. 1980. Detrended correspondence analysis:An improved ordination [J]. Vegetation, 42: 47 ～ 58.
[29] Hill MO. 1974. Correspondence analysis: A neglected multivariate method [J]. J. Roy. Statis . Soc. Series C, 23:340～354.
[30] Hill MO. 1973. Reciprocal averaging, an eigenvector method of ordination [J]. J. Ecol., 61: 237 249.
[31] Jongman RHG, Ter Braak CJF, Van Tongeren OFR. 2002. Data Analysis in Community and Landscape Ecology [M]. Wageningen: Pudoc.
[32] Kershaw KA, Looney JHH. 1985. Quantitative and Dynamic Plant Ecology (3rd edition) [M]. London:Edward Arnold.
[33] Oksanen J, Minchin PR. 2002. Continumm theory revisited:What shape are species responses along ecological gradients [J].Ecol. Model., 157:119～129.
[34] Swaine MD, Greig-Smith P. 1980. An application of principal components analysis to vegetation change in permanent plots[J].J.Ecol.,68:33～41.
[35] Ter Braak CJF, Prentice IC. 1988. A Theory of Gradient Analysis [J]. Adv. Ecol. Res., 18:271 ～317.
[36] Ter Braak CJF, Smilauer P. 2002. CANOCO Reference Manual and CanoDraw for Windows User's Guide: Software for Canonical Community Ordination (version 4. 5 ) [M]. Ithaca, NY USA: Microcomputer Power, 500.
[37] Ter Braak CJF. 1985. Correspondence Analysis of Incidence and Abundance Data: Properties in Terms of a Unimodal Response Modal [J]. Biometrics, 41: 859～ 873.
[38] Ter Braak CJF. 1986. Canonical correspondence analysis: A new eigenvector method for multivariate direct gradient analysis [J].Ecology, 67:1167～ 1179.
[39] Ter Braak CJF. 1987. The analysis of vegetation-environment relationships by canonical correspondence analysis [J]. Vegetation,69:69～77.
[40] Ter Braak CJF. 1994. Canonical community ordination. Part Ⅰ:Basic theory and linear methods [J]. Ecoscience, 1:127～140.
[41] Ter Braak CJF. 1995. Canonical correspondence analysis and related multivariate methods in aquatic ecology [J]. Aqua. Sci.,55:255 ～ 289.
[42] Yee TM, Mackenzie M. 2002. Vector generalized additive models in plant ecology [J]. Ecol. Model., 157:141 ～ 156.