基于GEE平台的国家公园土地覆盖变化遥感检测方法构建

doi:10.12302/j.issn.1000-2006.202101013

国家林草科技领军期刊
中国精品科技期刊
中国高校百佳科技期刊
江苏省新闻出版政府奖期刊奖
RCCSE林学权威期刊（A+）
CSCD核心期刊
Scopus数据库收录期刊
中文核心期刊
SCD核心期刊

作者加群：102861116

微信公众号：南京林业大学学报

高级检索

南京林业大学学报（自然科学版） ›› 2022, Vol. 46 ›› Issue (2): 213-220.doi: 10.12302/j.issn.1000-2006.202101013

基于GEE平台的国家公园土地覆盖变化遥感检测方法构建

毛丽君¹^,²(), 李海涛³, 薛晓明², 李建伟⁴^,^*(), 李明诗¹^,⁵

1.南京林业大学林学院,江苏南京 210037
2.南京森林警察学院刑事科学技术学院,野生动植物物证技术国家林业和草原局重点实验室,江苏南京 210023
3.临沂市城市管理综合服务中心,山东临沂 276000
4.云南省林业调查规划院大理分院,云南大理 671000
5.南京林业大学,南方现代林业协同创新中心,江苏南京 210037

收稿日期:2021-01-08 接受日期:2021-04-18 出版日期:2022-03-30 发布日期:2022-04-08
通讯作者: 李建伟
基金资助:
中央高校基本科研业务费专项资金项目(LGYB201704);国家自然科学基金项目(31971577);江苏省自然科学基金面上项目(BK20181338);江苏高校优势学科建设工程资助项目(PAPD);江苏省高校优秀科技创新团队项目(2019-29)

Developing remote sensing based methods for land cover change detection in national parks from GEE platform: a case study from the Qianjiangyuan National Park pilot area

MAO Lijun¹^,²(), LI Haitao³, XUE Xiaoming², LI Jianwei⁴^,^*(), LI Mingshi¹^,⁵

1. College of Forestry, Nanjing Forestry University, Nanjing 210037, China
2. College of Criminal Science and Technology, Nanjing Forest Police College; Key Laboratory of State Forest and Grassland Administration on Wildlife Evidence Technology, Nanjing 210023, China
3. Linyi City Management Comprehensive Service Center, Linyi 276000, China
4. Dali Branch of Yunnan Forestry Survey and Planning Institute, Dali 671000, China
5. College of Forestry, Co-Innovation Center for the Sustainable Forestry in Southern China, Nanjing Forestry University, Nanjing 210037, China

Received:2021-01-08 Accepted:2021-04-18 Online:2022-03-30 Published:2022-04-08
Contact: LI Jianwei

摘要/Abstract

摘要：

【目的】高效、可靠地提取土地覆盖动态变化信息对于国家公园保护与管理方案制定具有基础性数据支撑作用。本研究主要目的是构建经济高效的遥感监测方法,提取钱江源国家公园体制试点区2001—2017年的土地覆盖变化信息,评估国家公园保护与管理活动的有效性。【方法】首先定义研究区的土地覆盖分类体系(耕地、森林、草地、水体、人造地表和裸地),并利用增强型目视解译方法选取历史训练样本点;其次综合Landsat多季节复合数据的光谱波段、光谱指数、纹理特征和地形特征,基于Google Earth Engine(GEE)云平台执行随机森林分类算法,生成土地覆盖数据集;最后建立土地覆盖转换规则从而识别主要的土地覆盖变化类型及其空间分布模式。【结果】独立样本集验证结果表明,钱江源国家公园2001年、2009年和2017年土地覆盖总体分类精度分别为83.12%、81.82%和87.35%。造林活动、耕地废弃和开发建设的变化信息体现出保护管理工作取得显著成效。【结论】增强型目视解译能够高效、可靠地选取历史训练样本点,从而支持历史土地覆盖数据集构建。采用基于云计算的变化检测技术可快速评估国家公园管理措施的有效性并发现生态脆弱地带,具有高效、经济及计算资源限制性少等优势,可广泛推广应用。

关键词: 钱江源国家公园, 土地覆盖变化, 遥感监测, Google Earth Engine(GEE), 增强型目视解译

Abstract:

【Objective】Efficient and reliable extraction of land cover change information plays an important role in formulating management and protection plans for national parks. The primary objective of the current work was to develop an efficient and economical remote sensing detection method to extract land cover change information in the Qianjiangyuan National Park pilot area from 2001 to 2017, and to assess the effectiveness of the adopted protection and management measures. 【Method】A land cover classification scheme, including cropland, forest, grassland, water, artificial surface, and bare land, was first defined, followed by the selection of the historical training dataset using an enhanced visual interpretation method. Next, the original spectral bands, spectral indices, and textural features derived from Landsat multi-season composite data and terrain features were incorporated as input variables for implementing the random forest classification algorithm in the Google Earth Engine (GEE)cloud platform to generate land cover datasets. Finally, the major land cover conversion types and their spatial distributions were mapped by establishing land cover conversion rules. 【Result】The validation results derived from an independent sample set indicated that the overall accuracies for the 2001, 2009 and 2017 classifications were 83.12%, 81.82% and 87.35%, respectively. Afforestation activities, cropland abandonment, and distributions of development and construction projects demonstrated the effectiveness of the protection and management measures adopted in this park. 【Conclusion】The enhanced visual interpretation process helps to identify historical sample points efficiently and reliably to support the creation of historical land cover datasets. With cloud computing-based change detection technology, it is possible to quickly assess the protection and management efficacy and locate ecologically vulnerable zones. Moreover, the method has the advantages of high efficiency, economy, and fewer limitations on computing resources; thus, the GEE-based land cover change detection method proposed in the current work is suitable for an application in other similar scenarios.

Key words: Qianjiangyuan National Park, land cover change, remote sensing monitoring, Google Earth Engine(GEE), enhanced visual interpretation

中图分类号:

TP79

毛丽君,李海涛,薛晓明,等. 基于GEE平台的国家公园土地覆盖变化遥感检测方法构建[J]. 南京林业大学学报（自然科学版）, 2022, 46(2): 213-220.

MAO Lijun, LI Haitao, XUE Xiaoming, LI Jianwei, LI Mingshi. Developing remote sensing based methods for land cover change detection in national parks from GEE platform: a case study from the Qianjiangyuan National Park pilot area[J].Journal of Nanjing Forestry University (Natural Science Edition), 2022, 46(2): 213-220.DOI: 10.12302/j.issn.1000-2006.202101013.

0
/ / 推荐

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://nldxb.njfu.edu.cn/CN/10.12302/j.issn.1000-2006.202101013

http://nldxb.njfu.edu.cn/CN/Y2022/V46/I2/213

图/表 8

图1

表1

表2

研究中使用的光谱指数"

光谱指数 spectral index	计算公式 calculation formula	参考文献 reference
NDVI	I_NDVI= $ρ N I R - ρ R e d ρ N I R + ρ R e d$	[23]
EVI	I_EVI=2.5× $ρ N I R - ρ R e d ρ N I R + 6 × ρ R e d - 7.5 × ρ B l u e + 1$	[24]
GNDVI	I_GNDVI= $ρ N I R - ρ G r e e n ρ N I R + ρ G r e e n$	[25]
GLI	I_GLI= $ρ G r e e n - ρ R e d + (ρ G r e e n - ρ B l u e) 2 × ρ G r e e n + ρ R e d + ρ B l u e$	[26]
VARI	I_VARI= $ρ G r e e n - ρ R e d ρ G r e e n + ρ R e d - ρ B l u e$	[27]
GARI	I_GARI= $ρ N I R - [ρ G r e e n - 1.7 × ρ B l u e - ρ R e d] ρ N I R + [ρ G r e e n - 1.7 × ρ B l u e - ρ R e d]$	[28]
MSAVI	I_MSAVI= $2 × ρ N I R + 1 - (2 × ρ N I R + 1) 2 - 8 × (ρ N I R - ρ R e d) 2$	[29]
MNDWI	I_MNDWI= $ρ G r e e n - ρ S W I R ρ G r e e n + ρ S W I R$	[30]

表2

表3

表4

图2

图3

表5

参考文献 35

[1]	HE S Y, SU Y, WANG L, et al. Taking an ecosystem services approach for a new national park system in China[J]. Resour Conserv Recycl, 2018, 137:136-144. DOI: 10.1016/j.resconrec.2018.04.030. doi: 10.1016/j.resconrec.2018.04.030
[2]	米湘成. 生物多样性监测与研究是国家公园保护的基础[J]. 生物多样性, 2019, 27(1):1-4. doi: 10.17520/biods.2019047
	MI X C. Biodiversity monitoring and research are basis of national park conservation[J]. Biodivers Sci, 2019, 27(1):1-4. DOI: 10.17520/biods.2019047. doi: 10.17520/biods.2019047
[3]	王子芝, 李玥, 华世明, 等. 基于生态保护加权的普达措国家公园功能分区研究[J]. 南京林业大学学报(自然科学版), 2021, 45(6):225-231.
	WANG Z Z, LI Y, HUA S M, et al. Functional zoning of Potatso National Park by ecological protection weighting[J]. J Nanjing For Univ (Nat Sci Ed), 2021, 45(6):225-231. DOI: 10.12302/j.issn.1000-2006.202012053. doi: 10.12302/j.issn.1000-2006.202012053
[4]	MUCOVA S A R, FILHO W L, AZEITEIRO U M, et al. Assessment of land use and land cover changes from 1979 to 2017 and biodiversity & land management approach in Quirimbas National Park,northern Mozambique,Africa[J]. Glob Ecol Conserv,2018, 16:e00447. DOI: 10.1016/j.gecco.2018.e00447. doi: 10.1016/j.gecco.2018.e00447
[5]	SCHARSICH V, MTATA K, HAUHS M, et al. Analysing land cover and land use change in the Matobo National Park and surroundings in Zimbabwe[J]. Remote Sens Environ, 2017, 194:278-286. DOI: 10.1016/j.rse.2017.03.037. doi: 10.1016/j.rse.2017.03.037
[6]	WANG Y Q, LU Z, SHENG Y W, et al. Remote sensing applications in monitoring of protected areas[J]. Remote Sens, 2020, 12(9):1370. DOI: 10.3390/rs12091370. doi: 10.3390/rs12091370
[7]	MAO L J, LI M S, SHEN W J. Remote sensing applications for monitoring terrestrial protected areas:progress in the last decade[J]. Sustainability, 2020, 12(12):5016. DOI: 10.3390/su12125016. doi: 10.3390/su12125016
[8]	SACCHIS LOPES M, KOZHIKKODAN VEETTIL B, SALDANHA D L. Assessment of small-scale ecosystem conservation in the Brazilian Atlantic forest:a study from Rio Canoas State Park,southern Brazil[J]. Sustainability, 2019, 11(10):2948. DOI: 10.3390/su11102948. doi: 10.3390/su11102948
[9]	RECANATESI F, GIULIANI C, RIPA M. Monitoring Mediterranean oak decline in a peri-urban protected area using the NDVI and Sentinel-2 images: the case study of Castelporziano State Natural Reserve[J]. Sustainability, 2018, 10(9):3308. DOI: 10.3390/su10093308. doi: 10.3390/su10093308
[10]	SCHARSICH V, OTIENO D O, BOGNER C. Climbing up the hills:expansion of agriculture around the Ruma National Park,Kenya[J]. Int J Remote Sens, 2019, 40(17):6720-6736. DOI: 10.1080/01431161.2019.1591647. doi: 10.1080/01431161.2019.1591647
[11]	RAPINEL S, MONY C, LECOQ L, et al. Evaluation of Sentinel-2 time-series for mapping floodplain grassland plant communities[J]. Remote Sens Environ, 2019, 223:115-129. DOI: 10.1016/j.rse.2019.01.018. doi: 10.1016/j.rse.2019.01.018
[12]	BEY A, SÁNCHEZ-PAUS DÍAZ A, MANIATIS D, et al. Collect earth:land use and land cover assessment through augmented visual interpretation[J]. Remote Sens, 2016, 8(10):807. DOI: 10.3390/rs8100807. doi: 10.3390/rs8100807
[13]	TALUKDAR S, SINGHA P, MAHATO S, et al. Land-use land-cover classification by machine learning classifiers for satellite observations: a review[J]. Remote Sens, 2020, 12(7):1135. DOI: 10.3390/rs12071135. doi: 10.3390/rs12071135
[14]	Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32. doi: 10.1023/A:1010933404324
[15]	TSAI Y H, STOW D, AN L, et al. Monitoring land-cover and land-use dynamics in Fanjingshan National Nature Reserve[J]. Appl Geogr, 2019, 111:102077. DOI: 10.1016/j.apgeog.2019.102077. doi: 10.1016/j.apgeog.2019.102077
[16]	SHORES C R, MIKLE N, GRAVES T A. Mapping a keystone shrub species,huckle berry (Vaccinium membranaceum),using seasonal colour change in the rocky mountains[J]. Int J Remote Sens, 2019, 40(15):5695-5715. DOI: 10.1080/01431161.2019.1580819. doi: 10.1080/01431161.2019.1580819
[17]	GORELICK N, HANCHER M, DIXON M, et al. Google Earth Engine:planetary-scale geospatial analysis for everyone[J]. Remote Sens Environ, 2017, 202:18-27. DOI: 10.1016/j.rse.2017.06.031. doi: 10.1016/j.rse.2017.06.031
[18]	余建平, 申云逸, 宋小友, 等. 钱江源国家公园体制试点区功能分区对黑麂保护的有效性评估[J]. 生物多样性, 2019, 27(1):5-12. doi: 10.17520/biods.2018246
	YU J P, SHEN Y Y, SONG X Y, et al. Evaluating the effectiveness of functional zones for black muntjac (Muntiacus crinifrons) protection in Qianjiangyuan National Park pilot site[J]. Biodivers Sci, 2019, 27(1):5-12. DOI: 10.17520/biods.2018246. doi: 10.17520/biods.2018246
[19]	李双, 孙孝平, 方彦君, 等. 钱江源国家公园体制试点区农户生计资本评价及其空间特征[J]. 生物多样性, 2019, 27(1):64-75. doi: 10.17520/biods.2018176
	LI S, SUN X P, FANG Y J, et al. Evaluation of the spatial characteristics of farmer livelihood assets in the Qianjiangyuan National Park pilot[J]. Biodivers Sci, 2019, 27(1):64-75. doi: 10.17520/biods.2018176
[20]	FARR T G, ROSEN P A, CARO E, et al. The shuttle radar topography mission[J]. Rev Geophys, 2007, 45(2):RG2004. DOI: 10.1029/2005RG000183. doi: 10.1029/2005RG000183
[21]	HAKLAY M, WEBER P. Open Street Map:user-generated street maps[J]. IEEE Pervasive Comput, 2008, 7(4):12-18. DOI: 10.1109/MPRV.2008.80. doi: 10.1109/MPRV.2008.80
[22]	杨晓涵, 李雨彤, 王宁, 等. 遥感影像融合方法比较研究[J]. 重庆工商大学学报(自然科学版), 2017, 34(2):90-94.
	YANG X H, LI Y T, WANG N, et al. Comparative research on remote sensing image fusion method[J]. J Chongqing Technol Bus Univ (Nat Sci Ed), 2017, 34(2):90-94. DOI: 10.16055/j.issn.1672-058X.2017.0002.018. doi: 10.16055/j.issn.1672-058X.2017.0002.018
[23]	ROUSE J H, HASS R H, SCHELL J, et al. Monitoring vegetation systems in the great plains with ERTS[C]// Proceedings of the Third Earth Resources Technology Satellite-1 Symposium. Greenbelt, 1974: 3010-3017.
[24]	HUETE A, DIDAN K, MIURA T, et al. Overview of the radiometric and biophysical performance of the MODIS vegetation indices[J]. Remote Sens Environ, 2002, 83(1/2):195-213. DOI: 10.1016/S0034-4257(02)00096-2. doi: 10.1016/S0034-4257(02)00096-2
[25]	GITELSON A A, MERZLYAK M N. Remote sensing of chlorophyll concentration in higher plant leaves[J]. Adv Space Res, 1998, 22(5):689-692. DOI: 10.1016/S0273-1177(97)01133-2. doi: 10.1016/S0273-1177(97)01133-2
[26]	LOUHAICHI M, BORMAN M M, JOHNSON D E. Spatially located platform and aerial photography for documentation of grazing impacts on wheat[J]. Geocarto Int, 2001, 16(1):65-70. DOI: 10.1080/10106040108542184. doi: 10.1080/10106040108542184
[27]	GITELSON A A, STARK R, GRITS U, et al. Vegetation and soil lines in visible spectral space:a concept and technique for remote estimation of vegetation fraction[J]. Int J Remote Sens, 2002, 23(13):2537-2562. DOI: 10.1080/01431160110107806. doi: 10.1080/01431160110107806
[28]	GITELSON A A, KAUFMAN Y J, MERZLYAK M N. Use of a green channel in remote sensing of global vegetation from EOS-MODIS[J]. Remote Sens Environ, 1996, 58(3):289-298. DOI: 10.1016/S0034-4257(96)00072-7. doi: 10.1016/S0034-4257(96)00072-7
[29]	QI J, CHEHBOUNI A, HUETE A R, et al. A modified soil adjusted vegetation index[J]. Remote Sens Environ, 1994, 48(2):119-126. DOI: 10.1016/0034-4257(94)90134-1. doi: 10.1016/0034-4257(94)90134-1
[30]	徐涵秋. 利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究[J]. 遥感学报, 2005, 9(5):589-595.
	XU H Q. A study on information extraction of water body with the modified normalized difference water index (MNDWI)[J]. J Remote Sens, 2005, 9(5):589-595.
[31]	SIRIN A, MEDVEDEVA M, MASLOV A, et al. Assessing the land and vegetation cover of abandoned fire hazardous and rewetted peatlands:comparing different multispectral satellite data[J]. Land, 2018, 7(2):71. DOI: 10.3390/land7020071. doi: 10.3390/land7020071
[32]	LEINONEN U, KOSKINEN J, MAKANDI H, et al. Open foris and google earth engine linking expert participation with natural resource mapping and remote sensing training in Tanzania[J]. Int Arch Photogramm Remote Sens Spatial Inf Sci, 2018, XLII-4/W8:117-122. DOI: 10.5194/isprs-archives-xlii-4-w8-117-2018. doi: 10.5194/isprs-archives-xlii-4-w8-117-2018
[33]	VOROVENCII I. Quantification of forest fragmentation in pre-and post-establishment periods,inside and around Apuseni Natural Park,Romania[J]. Environ Monit Assess, 2018, 190(6):367. DOI: 10.1007/s10661-018-6741-0. doi: 10.1007/s10661-018-6741-0
[34]	YANG Y, HOBBIE S E, HERNANDEZ R R, et al. Restoring abandoned farmland to mitigate climate change on a full earth[J]. One Earth, 2020, 3(2):176-186. DOI: 10.1016/j.oneear.2020.07.019. doi: 10.1016/j.oneear.2020.07.019
[35]	王宇飞, 苏红巧, 赵鑫蕊, 等. 基于保护地役权的自然保护地适应性管理方法探讨:以钱江源国家公园体制试点区为例[J]. 生物多样性, 2019, 27(1):88-96. doi: 10.17520/biods.2018203
	WANG Y F, SU H Q, ZHAO X R, et al. Conservation easement-inspired adaptive management methods for natural protected areas:a case study on Qianjiangyuan National Park pilot[J]. Biodivers Sci, 2019, 27(1):88-96. DOI: 10.17520/biods.2018203. doi: 10.17520/biods.2018203

年份 year	传感器 sensor	可用图像数量 number of available images
年份 year	传感器 sensor	春季 spring	秋季 autumn	总数 total
2001	Landsat-7 ETM+	4	6	10
2009	Landsat-5 TM	4	4	8
2017	Landsat-8 OLI	5	8	13

类型 type	2001年		2009年		2017年
类型 type	用户精度/% user accuracy	生产者精度/% producer accuracy	用户精度/% user accuracy	生产者精度/% producer accuracy	用户精度/% user accuracy	生产者精度/% producer accuracy
耕地cropland	76.27	94.74	80.00	76.71	77.08	88.10
森林forest	93.38	94.78	90.60	96.43	93.10	96.43
草地grassland	73.33	70.00	74.41	77.55	86.27	91.67
水体water	89.47	78.00	95.24	76.67	89.66	86.47
人造地表artificial surface	70.00	68.00	71.05	75.00	76.32	70.73
裸地bare land	50.00	50.00	75.00	63.08	92.31	73.16
总体精度/% overall accuracy	83.12		81.82		87.35

年份 year	类别 type	古田山国家级自然保护区 Gutianshan National Nature Reserve		钱江源省级风景名胜区 Qianjiangyuan Provincial Scenic Spot		连接地带 connection area		钱江源国家公园 Qianjiangyuan National Park
年份 year	类别 type	面积/hm² area	比例/% proportion	面积/hm² area	比例/% proportion	面积/hm² area	比例/% proportion	面积/hm² area	比例/% proportion
2001—2009	造林活动	609.69	31.42	306.37	15.79	1 024.49	52.79	1 940.55	100.00
2009—2017	造林活动	238.05	22.40	124.61	11.73	699.83	65.87	1 062.49	100.00
2001—2009	耕地废弃	477.84	26.54	273.58	15.20	1 048.95	58.26	1 800.36	100.00
2009—2017	耕地废弃	122.12	21.89	96.78	17.35	339.06	60.77	557.96	100.00

基于GEE平台的国家公园土地覆盖变化遥感检测方法构建

Developing remote sensing based methods for land cover change detection in national parks from GEE platform: a case study from the Qianjiangyuan National Park pilot area

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 35

相关文章 3

编辑推荐

Metrics

本文评价

作者

读者

审稿

[1]	杨乐, 黄晓君, 包玉海, 包刚, 佟斯琴, 苏都毕力格. 无人机航高对落叶松毛虫虫害遥感监测精度的影响[J]. 南京林业大学学报（自然科学版）, 2023, 47(4): 13-22.
[2]	周梅,李春干,代华兵. 红树林空间分布遥感监测精度的影响因素及应对措施Symbol[J]. 南京林业大学学报（自然科学版）, 2018, 61(01): 163-168.
[3]	李明诗,梅昭容. 土地覆盖变化检测中不同相对辐射归一化方法的评价[J]. 南京林业大学学报（自然科学版）, 2017, 60(04): 108-114.

年份 year	耕地 cropland	森林 forest	草地 grassland	水体 water	人造地表 artificial surface	裸地 bare land
2001	250	338	54	74	88	26
2009	166	339	124	71	92	38
2017	119	333	151	80	96	51