干旱区生态环境知识资源中心(Arid): 基于人工神经网络的青藏公路铁路沿线生态系统风险研究

Arid

	基于人工神经网络的青藏公路铁路沿线生态系统风险研究
其他题名	Ecological Risk Assessment of Regions Alongside Qinghai-Xizang Highway and Railway Based on Artificial Neural Network
	陈辉 1; 李双成 2; 郑度 1
来源期刊	北京大学学报. 自然科学版
ISSN	0479-8023
出版年	2005
卷号	41 期号:4 页码:586-593
中文摘要	根据青藏公路铁路沿线(50 km缓冲区)生态系统特征,选取雪灾、旱灾、崩塌滑坡等7项指标,依托人工神经网络MLP(Multilayer Percetron)模型,构建青藏公路铁路沿线生态风险评价模型.评价结果显示:青藏公路铁路沿线生态系统所跨越的6个自然区的平均生态风险值居前3位的是:柴达木山地荒漠区(4.258 5),果洛那曲高寒灌丛草甸区(2.764 0)、青东祁连山地草原区(2.733 5);沿线10种植被生态系统平均生态风险值居前3位的是:针叶林生态系统(4.309 6)、荒漠生态系统(4.117 4)和无植被地段(3.618 2).在影响各区、各植被生态系统风险值大小的因素中,自然因素为主要控制因素,人为因素影响相对较弱.依据评价结果,将青藏公路铁路沿线生态系统划分为4个区:柴达木盆地高风险区、西大滩至当雄中度风险区、青东祁连和青南2个轻度风险区.
英文摘要	Concept model of regional ecological risk is built for the characteristics of ecosystems alongside Qinghai-Xizang highway and railway based on MLP (Multilayer percetron) model. Seven indices such as snow hazard, drought hazard and landslide are selected to evaluate the integrated ecological risk of the ecosystems alongside Qinghai-Xizang highway and railway. Results show that Qaidam mountain desert region has the highest average ecological risk value (4.258 5), followed by Guoluonaqu alpine scrub meadow region (2.764 0) and Qinghai and Qilian mountain steppe region (2.733 5) among these ecosystems alongside Qinghai-Xizang highway and railway through six natural regions. As far as land cover types are concerned, the top three ecological risk values appear in the needle-leaved forest (4.309 6), desert ecosystem (4.117 4) and hind without vegetation (3.618 2) respectively, which are higher than those in the other seven land cover types in the study site. Although the risk/’values are influenced by natural factors and human activities, they are more strongly controlled by natural factors. According to the ecological risk characteristics, these ecosystems alongside Qinghai-Xizang highway and railway are subdivided to four subregions, including Chaidam basin high risk region, Xidatan to Dangxiong moderate risk region, and east Qinghai, Qilian and south Qinghai slight risk regions.
中文关键词	生态风险评价 ; 人工神经网络 ; MLP模型 ; 自然因素 ; 人为因素
英文关键词	ecological risk assessment artificial neural network MLP(Multilayer percetron model natural factors, artificial factors
语种	中文
国家	中国
收录类别	CSCD
WOS类目	ENVIRONMENTAL SCIENCES
WOS研究方向	Environmental Sciences & Ecology
CSCD记录号	CSCD:2186722
来源机构	中国科学院地理科学与资源研究所 ; 北京大学
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/205901
作者单位	1.中国科学院地理科学与资源研究所, 北京 100101, 中国; 2.北京大学环境学院, 北京 100871, 中国
推荐引用方式 GB/T 7714	陈辉,李双成,郑度. 基于人工神经网络的青藏公路铁路沿线生态系统风险研究[J]. 中国科学院地理科学与资源研究所, 北京大学,2005,41(4):586-593.
APA	陈辉,李双成,&郑度.(2005).基于人工神经网络的青藏公路铁路沿线生态系统风险研究.北京大学学报. 自然科学版,41(4),586-593.
MLA	陈辉,et al."基于人工神经网络的青藏公路铁路沿线生态系统风险研究".北京大学学报. 自然科学版 41.4(2005):586-593.