基于多源遥感数据橡胶林识别及其种植适宜区研究

doi:10.13466/j.cnki.lyzygl.2020.06.009

林业资源管理 ›› 2020, Vol. 0 ›› Issue (6): 54-60.doi: 10.13466/j.cnki.lyzygl.2020.06.009

基于多源遥感数据橡胶林识别及其种植适宜区研究

蒋永泉(), 史正涛(), 高书鹏

云南师范大学地理学部,昆明 650500

收稿日期:2020-09-30 修回日期:2020-10-26 出版日期:2020-12-28 发布日期:2021-01-26
通讯作者: 史正涛
作者简介:蒋永泉(1993-),女,云南保山人,在读硕士,主要从事热带地区地物识别。Email: jiangyongquan1108@163.com
基金资助:
云南省水利厅水利科技项目“云南主要人工经济林对区域水资源安全的影响调查研究”(2014003)

Research on Rubber Plantation Identification and Suitable Areas Based on Multi-source Remote Sensing Data

JIANG Yongquan(), SHI Zhengtao(), GAO Shupeng

Department of Geography,Yunnan Normal University,Kunming 650500,China

Received:2020-09-30 Revised:2020-10-26 Online:2020-12-28 Published:2021-01-26
Contact: SHI Zhengtao

摘要/Abstract

摘要：

以云南省德宏州瑞丽市为研究区,以Sentinel-2多光谱遥感影像和MOD13Q1为数据源,利用支持向量机分类方法识别该区的橡胶林,并对橡胶林空间分布特征及其种植适宜区进行分析。结果表明:1)联合Sentinel-2和MOD13Q1-NDVI数据识别橡胶林的结果明显优于仅用Sentinel-2多光谱影像或MOD13Q1-NDVI时序数据,实现了橡胶林高精度识别,总体精度为91.41%,Kappa系数为0.76。2)截至2019年底,瑞丽市橡胶林种植面积为7 693.83 hm²,主要种植区海拔范围为682~900 m,占总种植面积的83.61%;主要种植区坡度为5~15°,占总种植面积的84.61%。3)瑞丽市有6 433.11 hm²橡胶林种植在适宜区内,有1260.72 hm²种植在不适宜区内,分别占橡胶林种植总面积的83.61%和16.39%。

关键词: 多源遥感数据, 橡胶林识别, 空间分布特征, 种植适宜区, 瑞丽市

Abstract:

Taking Ruili City,Dehong Prefecture,Yunnan Province as the study area,this paper analyzes the spatial distribution characteristics of the rubber forest and its suitable planting areas.Sentinel-2 multi-spectral remote sensing images and MOD13Q1 were used as data sources to identify the rubber forest with the method of supporting vector machine classification.The results show that 1) the identification ability of the combined Sentinel-2 and MOD13Q1 NDVI data for rubber forest was significantly better than using only one kind of Sentinel-2 multi-spectral image or MOD13Q1-NDVI time series data,which achieved high precision identification of rubber forest.The overall identification accuracy is 91.41% and the Kappa coefficient is 0.76.2) By the end of 2019,the rubber planting area in Ruili was 7693.83 hm²,with the elevation range of the main planting area being 682-900 m,accounting for 83.61% of the total planting area;the slope of the main planting area was 5-15°,accounting for 84.61% of the total planting area.3) At present,6,433.11 hm² of rubber was planted in the suitable areas in Ruili while 1,260.72 hm² of rubber was not,accounting for 83.61% and 16.39%,respectively.

Key words: multi-source remote sensing data, rubber plantation identification, spatial distribution characteristics, suitable planting areas, Ruili City

中图分类号:

TP79
P237

蒋永泉, 史正涛, 高书鹏. 基于多源遥感数据橡胶林识别及其种植适宜区研究[J]. 林业资源管理, 2020, 0(6): 54-60.

JIANG Yongquan, SHI Zhengtao, GAO Shupeng. Research on Rubber Plantation Identification and Suitable Areas Based on Multi-source Remote Sensing Data[J]. FOREST RESOURCES WANAGEMENT, 2020, 0(6): 54-60.

图/表 8

图1

图2

表1

图3

表2

表3

表4

图4

参考文献 28

[1]	杜香玉. 从战略资源到生态系统遭受破坏——对20世纪以来橡胶引种的认知与反思[J]. 原生态民族文化学刊, 2020,12(1):10-20.
[2]	何康, 黄宗道. 热带北缘橡胶树栽培[M]. 广州: 广东科技出版社, 1987.
[3]	付如作. 瑞丽市橡胶产业发展现状及对策研究[J]. 农业与技术, 2017,37(5):145-147.
[4]	Li H, Aide T M, Ma Y, et al. Demand for rubber is causing the loss of high diversity rain forest in SW China[M]// Plant Conservation and Biodiversity.Dordrecht:Springer, 2006: 157-171.
[5]	陈汇林, 陈小敏, 陈珍丽, 等. 基于MODIS遥感数据提取海南橡胶信息初步研究[J]. 热带作物学报, 2010,31(7):1181-1185.
[6]	Senf C, Pflugmacher D, Van der Linden S, et al. Mapping rubber plantations and natural forests in Xishuangbanna(Southwest China) using multi-spectral phenological metrics from MODIS time series[J]. Remote Sensing, 2013,5(6):2795-2812.
[7]	高书鹏, 史正涛, 刘晓龙, 等. 基于高时空分辨率可见光遥感数据的热带山地橡胶林识别[J]. 遥感技术与应用, 2018,33(6):1122-1131.
[8]	Chen Bangqian, Li Xiangping, Xiao Xiangming, et al. Mapping tropical forests and deciduous rubber plantations in Hainan Island,China by integrating PALSAR 25-m and multi-temporal Landsat images[J]. International journal of applied earth observation and geoinformation, 2016,50:117-130.
[9]	Kou Weili, Xiao Xiangming, Dong Jinwei, et al. Mapping deciduous rubber plantation areas and stand ages with PALSAR and Landsat images[J]. Remote Sensing, 2015,7(1):1048-1073.
[10]	Pohl C Van Genderen J L. Review article multisensor image fusion in remote sensing:concepts,methods and applications[J]. International journal of remote sensing, 1998,19(5):823-854. doi: 10.1080/014311698215748
[11]	黄艳, 廖玲. 云研77-2和77-4橡胶树新品种区域性栽培试验初报[J]. 中国热带农业, 2009(2):38-40.
[12]	江爱良. 云南南部、西南部生态气候和橡胶树的引种[J]. 中国农业气象, 1995(5):26-31.
[13]	吴俊. 云南橡胶树气候生态适应性分析[J]. 现代农业科技, 2011(19):308-309.
[14]	德宏州农垦分局. 德宏垦区1982/83年冬期橡胶树寒害调查报告[J]. 云南热作科技, 1984(2):4-8.
[15]	付如作. 瑞丽市橡胶产业发展现状及对策研究[J]. 农业与技术, 2017,37(5):145-147.
[16]	祁栋灵, 王秀全, 左雪冬. 宣贯橡胶树栽培和割胶技术规程提升植胶业科技水平[J]. 热带农业科学, 2008,28(5):50-54.
[17]	康靖. 浅论瑞丽市热带天然林停伐后的保护问题[J]. 农村实用技术, 2020(5):123-124.
[18]	董丽芳, 杨军, 杨涛, 等. 瑞丽市橡胶产业发展现状及对策[J]. 云南农业, 2018(8):42-45.
[19]	陈柳嫱, 刘杰, 杨雯雯. 森林培育对瑞丽市生态环境建设的影响[J]. 绿色科技, 2019(8):128-129.
[20]	尹啟令, 李仕娇, 雷荣华, 等. 瑞丽地区橡胶幼林套种百香果初探[J]. 热带农业科技, 2019,42(4):29-31.
[21]	Main-Knorn M, Pflug B, Louis J, et al. Calibration and validation plan for the L2A processor and products of the Sentinel-2 mission[C]// Proceedings of International Symposium on Remote Sensing of Environment.36th International Symposium on Remote Sensing of Environment(ISRSE), 2015.
[22]	Emde C, Buras-Schnell R, Kylling A, et al. The libRadtran software package for radiative transfer calculations(version 2.0.1)[J]. Geoscientific Model Development, 2016,9(5):1647-1672.
[23]	罗媛, 谢堂民, 龙显静, 等. 基于红光和近红外反射光谱特征参数反演草地地上生物量[J]. 草原与草坪, 2015,35(5):65-69.
[24]	Tucker C J. Red and photographic infrared linear combinations for monitoring vegetation[J]. Remote sensing of Environment, 1979,8(2):127-150.
[25]	Haralick R M, Shanmugam K, Dinstein I H. Textural features for image classification[J]. IEEE Trans.SMC, 1973(6):610-621.
[26]	陈强, 田杰, 黄海宁, 等. 基于统计和纹理特征的SAS图像SVM分割研究[J]. 仪器仪表学报, 2013,34(6):214-221.
[27]	SL 190-2007,土壤侵蚀分类分级标准[S].
[28]	NY/T221-2016,橡胶树栽培技术规程[S].

特征描述
光谱特征	光谱反射率	植被红边波段1,B5 植被红边波段4,B8A	1 1	—
植被指数		NDVI指数	1	23
		均值	4
纹理特征	灰度共生矩阵	同质性相关性熵对比度	4 4 4 4	—

数据源	橡胶林/%	其他/%	PA/UA	OA/%	Kappa系数
MOD13Q1-NDVI数据	71.48	88.89	PA	85.31	0.57
MOD13Q1-NDVI数据	62.46	92.34	UA	85.31	0.57
Sentinel-2数据	73.00	92.13	PA	88.20	0.64
Sentinel-2数据	70.59	92.96	UA	88.20	0.64
多源数据	89.73	91.84	PA	91.41	0.76
多源数据	73.98	97.19	UA	91.41	0.76

海拔/m	面积/hm²	所占比例/%
682~ 900	6433.11	83.61
900~1100	1174.23	15.26
1000~1200	86.49	1.12
1200~1300	0	0
1300~1400	0	0
≥1400	0	0
合计	7693.83	100.00

坡度/(°)	面积/hm²	所占比例/%
0~ 5	439.11	5.71
5~ 8	3818.61	49.63
8~15	2691.63	34.98
15~25	660.33	8.58
25~35	84.15	1.09
≥35	0	0
合计	7693.83	100.00

基于多源遥感数据橡胶林识别及其种植适宜区研究

Research on Rubber Plantation Identification and Suitable Areas Based on Multi-source Remote Sensing Data

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 28

相关文章 2

编辑推荐

Metrics

本文评价

[1]	王政, 杨霞. 森林康养空间分布特征及其影响因素研究——以四川森林康养基地为例[J]. 林业资源管理, 2020, 0(2): 146-153.
[2]	张一, 张超. 云南省华山松林空间分布影响因子分析[J]. 林业资源管理, 2018, 0(6): 68-75.