基于RBF神经网络的城市森林碳储量遥感建模与动态变化分析——以南京市为例

林业资源管理 ›› 2013, Vol. 0 ›› Issue (4): 120-125.

基于RBF神经网络的城市森林碳储量遥感建模与动态变化分析——以南京市为例

凌子燕¹, 黄进²

1.广西基础地理信息中心,南宁 530023;
2.广西民族大学,南宁 530006

收稿日期:2013-02-18 修回日期:2013-05-08 出版日期:2013-08-28 发布日期:2020-11-23
作者简介:凌子燕(1985-),助工,主要从事林业遥感与GIS应用研究工作。Email:lingziyan@163.com
基金资助:
广西民族大学重点科研课题(MDZD044)

Study on Estimation Model and Dynamic Change of Urban Forest Carbon Storage Based on RBF and RS ——A Case Study in Nanjing

LING Ziyan¹, HUANG Jin²

1. Geomatics Center of Guangxi,Nanning 530023,Guangxi,China;
2. Guangxi University for Nationnalities,Nanning 530006,Guangxi,China

Received:2013-02-18 Revised:2013-05-08 Online:2013-08-28 Published:2020-11-23

摘要/Abstract

摘要： 研究将TM遥感数据的归一化植被指数(NDVI)和短波红外植被指数(SWVI)融入径向基神经网络中,构建了森林碳储量估算模型,并应用该模型反演了南京市不同时期的森林碳储量。结果显示:1)该模型的平均估算精度达到74.62%,比线性回归模型精度更高、拟合更稳定,是一种相对简便、易于操作和准确度高的森林碳储量估算方法;2)南京市的森林碳储量总体分布面积较大,但在区域上分布不太均匀,1988—2005年期间各辖区的森林碳储量均有不同程度的减少,减少率由小到大依次排列为:栖霞区<玄武区<下关区<雨花台区<鼓楼区<白下区<秦淮区<建邺区。

关键词: 森林碳储量, RBF神经网络, 遥感, 南京

Abstract: NDVI and SWVI derived from Landsat TM images are blent into Radial Basis Function to generate an estimation model of forest carbon storage in Nanjing,and this model is compared with Regression Model.The results show that:(1)The precision of Radial Basis Function Model is 74.62%,it’s more stable and has higher precision than Regression Model,so it’s a relatively simple,easy-to-operate and accurate estimation method for carbon storage of forest;(2)From spatial distribution pattern perspective,the forest carbon storage in Nanjing is distributed extensively,but not even regionally.From time dynamic change perspective,during the period of 1988 to 2005,the forest carbon storage of each district declined in different degrees,the descent rates were listed in a low-to-high order:Qixia,Xuanwu,Xiaguan,Yuhuatai,Gulou,Baixia,Qinhuai and Jianye.

Key words: carbon storage of forest, RBF network, RS, Nanjing

中图分类号:

S718.55
TP79

凌子燕, 黄进. 基于RBF神经网络的城市森林碳储量遥感建模与动态变化分析——以南京市为例[J]. 林业资源管理, 2013, 0(4): 120-125.

LING Ziyan, HUANG Jin. Study on Estimation Model and Dynamic Change of Urban Forest Carbon Storage Based on RBF and RS ——A Case Study in Nanjing[J]. FOREST RESOURCES WANAGEMENT, 2013, 0(4): 120-125.

参考文献

[1]Roger A.Sedjo.The Carbon Cycle and Global Forest Ecosystem[J].Water,Air,And Soil Pollution,1993,70:295-307.
[2]温家石.城市化对建成区植被碳吸收和碳储存的影响研究[D].杭州:浙江大学,2010.
[3]Piao S L,Fang J Y,Ciais P.The carbon balance of terrestrial ecosystems in China[J].Nature,2009,458:1009-1014.
[4]方精云,刘国华,徐篙龄.我国森林植被的生物量和净生产量[J].生态学报,1996,16(5):497-508.
[5]马钦彦,谢征鸣.中国油松林储碳量基本估计[J].北京林业大学学报,1996,18(3):31-34.
[6]阮宏华,姜志林,高苏铭.苏南丘陵主要森林类型碳循环研究——含量与分布规律[J].生态学杂志,1997,16(6):17-21.
[7]卢义山,梁珍海,吴仲祥,等.苏北海堤防护林主要造林树种林分生物量与生产力的研究[J].江苏林业科技,2000,27(2):12-15.
[8]张德全,桑卫国,李曰峰,等.山东省森林有机碳储量及其动态的研究[J].植物生态学报,2002,26(增刊):93-97.
[9]罗怀良,朱波,陈国阶.川中丘陵地区主要植被含碳率研究——以四川省盐亭县为例[J].西南农业学报,2004,17(增刊):113-116.
[10]Birdsey R A.Carbon storage and accumulation in United States forest ecosystems[R].United States Department of Agriculture Forest Service,General Technical Report,1992:50-59.
[11]韩爱惠,智长贵,王雪军,等.利用MODIS数据反演森林生物量的方法研究[J].林业资源管理,2011(4):40-47.
[12]Myeong S,Nowak D J,Duggin M J.A temporal analysis of urban forest carbon storage using remote sensing[J].Remote Sensing of Environment,2006,101:277-282.
[13]Lee N J,K Nakane.Forest vegetation classification and biomass estimation based on Landsat TM data in a mountainous region of west Japan[C]//Gholz H L,K Nakane,and H Himoda eds:The use of remote sensing in the modeling of forest productivity.Netherlands:Kluwer Academic Publishers,1997:159-171.
[14]李仁东,刘纪远.应用Landsat ETM 数据估算鄱阳湖湿生植被生物量[J].地理学报,2001,56(5):532-540.
[15]Myneni R B,Dong J,Tucker C J,et al.A large carbon sink in the woody biomass of Northern forests[J].PNAS,2001,98(26):14784-14789.
[16]Dong Jiarui,Kaufmann Robert K,Myneni Ranga B,et al.Remote sensing estimates of boreal and temperate forest woody biomass:carbon pools,sources,and sinks[J].Remote Sens.Environ.2003,84:393-410.
[17]Fraser R H,Li Z.Estimating Fire-related Parameters in Boreal Forest Using SPOT Vegetation[J].Remote Sensing of Environment,2002,82:95-110.
[18]郑光,田庆久,陈镜明,等.结合树龄信息的遥感森林生态系统生物量制图[J].遥感学报,2006,10(6):932-940.
[19]田雨波.混合神经网络技术[M].北京:科学出版社,2009,60-73.
[20]王雪军,孙玉军.基于遥感地学模型的辽宁省森林生物量和碳储量估测[J].林业资源管理,2011(1):100-105.