文献类型：期刊文献

中文题名：利用分布有/无数据预测物种空间分布的研究方法综述!

英文题名：The review of methods for mapping species spatial distribution using presence/absence data

作者：刘芳[1] 李晟[2] 李迪强[1]

第一作者：刘芳

机构：[1]中国林业科学研究院森林生态环境与保护研究所,国家林业局森林生态环境重点实验室;[2]Department of Forest&Wildlife Ecology,University of Wisconsin-Madison,209 Russell Labs,1630 Linden Drive University of Wisconsin-Madison Madison

年份：2013

卷号：33

期号：22

起止页码：7047-7057

中文期刊名：生态学报

外文期刊名：Acta Ecologica Sinica

收录：CSTPCD;;Scopus;北大核心:【北大核心2011】；CSCD:【CSCD2013_2014】；

基金：国家"十二五"科技支撑课题资助项目(2013BAC09B02)

语种：中文

中文关键词：有;无数据;物种分布模型;取样;模型评估

外文关键词：Presence-absence data ; species distribution model; sampling ; model assessment

分类号：Q959.837

摘要：详细的物种地理分布信息是生态学研究和制定保护策略的基础。相比较于直接估测种群数量,获取物种分布的有/无数据更为实用。因此,利用分布有/无数据并结合环境变量建立模型预测物种空间分布的方法在近年来得到了长足发展,并被广泛应用。利用分布有/无数据预测物种分布,关键的步骤包括:1)构建总体概念模型,2)收集物种分布有/无数据,并准备环境变量图层;3)选择合适的统计模型和算法,以及4)对模型进行评估。概念模型提出研究假设,并确定数据收集及模型方法。收集物种分布数据有系统调查及非系统调查方法。筛选并准备与物种分布相关的环境变量,利用GIS工具处理,使之成为符合模型条件的具有合适的空间尺度的数字化图层。利用环境变量和物种分布有/无的数据,选择合适的方法及软件建立模型,并对模型进行检验和评估。总结了用于构建物种分布模型的不同算法和软件。针对以上各个环节,阐述利用物种分布有/无数据进行研究所需要的技术细节,以期望为读者提供借鉴。
Detailed species geographical distribution is fundamental to ecology management and conservation strategies. Compared with the considerable effort needed for direct population estimates, obtaining presence/absence data are more practical and economically effective, so the methods on mapping species distribution with presence/absence data and related environmental variables have been well developed in recent decades with broad implications. The key steps for species distribution modeling using presence/absence data include： 1 ） develop the conceptual model, 2） collect presence/absence data and prepare environmental variable layers, 3） select statistic model and algorithm, and 4） run the model and conduct model evaluation. Conceptual models propose the hypothesis of the research and build outlines for data preparation and statistical methods. Systematic sample and non-systematic sampling are used to collect presence/absence data. Relatedenvironmental variables are selected and georeferenced by GIS tools to be used in species distribution models at a proper scale. We summarized the algorithms and software that have been used to develop species distribution models combining presence/absence data and environmental variables. Assessment of model performance is crucial for implicating the results generated by those models. This paper reinvented the details of techniques for the above steps of species distribution models. We expect that this paper will show readers some revelation on how to use species distribution models.