资源一号02D高光谱数据红树林地上生物量反演

doi:10.3964/j.issn.1000-0593(2023)12-3906-10

摘要
参考文献
相关文章 (15)

全文: PDF (18469 KB)
输出: BibTeX | EndNote (RIS)

摘要：红树林生态系统是地球上生产力最高的生态系统之一，它也是海岸带“蓝碳”生态系统的重要组成部分。地上生物量作为红树林蓝碳的重要组成部分，如何准确快速地获取红树林地上生物量已成为红树林生态系统研究的热门问题。分析北部湾茅尾海红树林地上生物量(AGB)空间分布格局及其量级，可为该区域红树林生态环境保护及“南红北柳”生态修复提供科学依据。资源一号数据作为我国自主研发的民用国产高光谱卫星，其高光谱数据为红树林地上生物量的研究提供了新的机遇。机器学习算法因其高性能、高效率的优势被越来越多的应用于红树林相关研究，目前已经成为获取红树林参数信息的重要手段。高光谱数据在红树林地上生物量的反演精度如何? 国产高光谱卫星数据和机器学习算法在红树林地上生物量的估算中能否应用? 这些问题仍需进一步验证。基于国产资源一号02D高光谱数据，采用极端梯度提升(XGBoost)、随机森林回归(RFR)以及K近邻回归(KNNR)三种不同的机器学习算法对茅尾海的红树林地上生物量进行估算，在此基础上对比了不同的机器学习算法的性能。结果显示：(1)无瓣海桑红树林地上生物量的平均值最高（90.93 Mg·ha^-1），桐花树次之（52.63 Mg·ha^-1），而秋茄最小（20.27 Mg·ha^-1）。(2)采用XGBoost、RF以及KNN三种机器学习算法进行红树林地上生物量和红树林光谱变量建模后发现，基于对数倒数1阶变换的XGBoost模型精度最高，为最佳的机器学习模型。其模型在测试阶段R²=0.751 5，RMSE=27.494 8 Mg·ha^-2。(3)基于资源一号02D高光谱数据，采用XGBoost算法反演茅尾海的红树林地上生物量介于4.58~208.35 Mg·ha^-2之间，平均值为88.98 Mg·hm^-2，地上生物量在空间上呈现出中部低，两边高的空间分布格局。总之，该研究论证了国产高光谱卫星数据和XGBoost机器学习算法的组合在红树林生物量的估算方面具有良好的应用前景，可为茅尾海红树林的生态修复和保护提供科学依据和技术支撑。

关键词：高光谱数据；资源一号02D；机器学习；红树林地上生物量；遥感反演；北部湾茅尾海

Abstract：Mangrove ecosystem is one of the most productive ecosystems on the earth, and it is one of the important components of the coastal "blue carbon" ecosystem. As an important part of mangrove blue carbon, obtaining mangrove aboveground biomass accurately and quickly has become one of the hot issues in mangrove ecosystem research. Analyzing spatial distribution pattern and magnitude of Aboveground biomass (AGB) of mangroves in the Maowei Sea of Beibu Gulf can provide the scientific basis for the protection of the mangrove ecological environment and the ecological restoration of “South Red and North Willow” in this area. As a domestic civil hyperspectral satellite independently developed by China, the hyperspectral data of ZY-1-02D provides a new opportunity to research mangrove aboveground biomass. Because of its high performance and efficiency, machine learning algorithms are increasingly used in mangrove-related research. It has become an important means to obtain mangrove parameter information. How accurate is the retrieval of hyperspectral data in mangrove aboveground biomass, whether the domestic hyperspectral satellite data and machine learning algorithm can be applied to the estimation of mangrove aboveground biomass needs further verification. Based on ZY-1-02D Satellite hyperspectral data, three different machine learning algorithms, eXtreme Gradient Boosting (XGBoost), Random Forest Regression (RFR) and k-nearest neighbor regression (KNNR), were used to estimate the biomass of mangrove forests in the Maowei Sea. On this basis, the performance of different machine learning algorithms was compared. The results showed that: (1) The average aboveground biomass of Sonneratia apetala mangrove was the highest (90.93 Mg·ha^-1), followed by Aegiceras corniculatum mangrove(52.63 Mg·ha^-1), and Kandelia candel mangrove was the lowest (20.27 Mg·ha^-1). (2) XGBoost, RF and KNN machine learning algorithms are used to model mangrove aboveground biomass and mangrove spectral variables. The XGBoost model based on log reciprocal first-order transformation has the highest accuracy and is the best machine learning model. In the testing phase, R²=0.751 5, RMSE=27.494 8 Mg·ha^-2. (3) Based on the ZY-1-02D Satellite hyperspectral data, the XGBoost algorithm is used to retrieve the aboveground biomass of mangroves in Maowei Sea, which is between 4.58 and 208.35 Mg·ha^-2, with an average value of 51.92 Mg·ha^-2. The aboveground biomass shows a spatial distribution pattern of low in the middle and high on both sides. In a word, this paper demonstrates that the combination of domestic hyperspectral data and XGBoost machine learning algorithm has a good application prospect in the estimation of mangrove biomass, which can provide scientific basis and technical support for the ecological restoration and protection of Maowei Sea mangroves.

Key words：Domestic hyperspectral data; ZY-1-02D Satellite; Machine learning; Mangrove aboveground biomass; Remote sensing inversion; Maowei Sea in Beibu Gulf

收稿日期: 2022-10-07 修订日期: 2023-04-17

中图分类号:

TP79

基金资助: 国家自然科学基金项目（42261024），广西高校人文社会科学重点研究基地“北部湾海洋发展研究中心”项目（JDZD202214, BHZKY2202），北部湾大学海洋科学一流学科项目(DRB003)，广西林业科技推广示范项目（桂林科研[2022]第4号），广西基地和人才项目(2019AC20088)资助

通讯作者: 田义超 E-mail: tianyichao1314@yeah.net

作者简介: 黄友菊，1980年生，广西壮族自治区自然资源遥感院高级工程师 e-mail: 17017131@qq.com

引用本文:

黄友菊，田义超，张强，陶进，张亚丽，杨永伟，林俊良. 资源一号02D高光谱数据红树林地上生物量反演[J]. 光谱学与光谱分析, 2023, 43(12): 3906-3915.
HUANG You-ju, TIAN Yi-chao, ZHANG Qiang, TAO Jin, ZHANG Ya-li, YANG Yong-wei, LIN Jun-liang. Estimation of Aboveground Biomass of Mangroves in Maowei Sea of Beibu Gulf Based on ZY-1-02D Satellite Hyperspectral Data. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(12): 3906-3915.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2023)12-3906-10 或 https://www.gpxygpfx.com/CN/Y2023/V43/I12/3906

[1] Murdiyarso D, Purbopuspito J, Kauffman J B, et al. Nature Climate Change, 2015, 5(12): 1089.
[2] LI Shan-shan, TIAN Qing-jiu(李姍姍，田庆久). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2013, 33(1): 136.
[3] FAN Hang-qing, LIANG Shi-chu(范航清，梁士楚). Mangrove Research and Management in China(中国红树林研究与管理). Beijing: Science Press(北京: 科学出版社), 1995. 33.
[4] Osland M J, Hughes A R, Armitage A R, et al. Global Change Biology, 2022, 28(10): 3163.
[5] Afonso F, Félix P M, Chainho P, et al. Regional Studies in Marine Science, 2022, 49: 102114.
[6] TIAN Yi-chao, ZHANG Qiang, TAO Jin, et al(田义超, 张强, 陶进, 等). Research and Exploration in Laboratory(实验室研究与探索), 2022，(12): 207.
[7] Tian Y C, Zhang Q, Huang H, et al，Ecological Indicators, 2022, 136: 108694.
[8] Gao T, Ding D, Guan W, et al. Polish Journal of Environmental Studies, 2018, 27(3): 1061.
[9] DENG Jing-wen, TIAN Yi-chao, ZHANG Qiang, et al(邓静雯, 田义超, 张强, 等). Remote Sensing for Natural Resources(自然资源遥感), 2022, 34(3): 129.
[10] LI Chun-gan, TAN Bi-zeng(李春干, 谭必增). Journal of Natural Resources(自然资源学报), 2003, 18(2): 215.
[11] Twilley R R, Chen R H, Hargis T. Water, Air & Soil Pollution，1992, 64：265.
[12] Hutchison J, Manica A, Swetnam, R, et al. Conservation Letters, 2014, 7：233.
[13] Rovai A S, Riul P, Twilley R R, et al. Global Ecology and Biogeography, 2016, 25：286.
[14] Castillo J A A, Apan A A, Maraseni T N, et al. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 134：70.
[15] Giri C. Remote Sensing, 2021, 13(4): 563.
[16] Avtar R, Kumar P, Oono A, et al. Geocarto International, 2017, 32：874.
[17] Zheng D L, Rademacher J, Chen J Q, et al. Remote Sensing of Environment, 2004, 93(3): 402.
[18] CAO Qing-xian, XU Da-ping, JU Hong-bo(曹庆先, 徐大平, 鞠洪波). Forest Resources Management(林业资源管理), 2010, (6): 102.
[19] Zhu Y H, Liu K, Liu L, et al. Remote Sensing, 2015, 7(9): 12192.
[20] Hirata Y, Tabuchi R, Patanaponpaiboon P, et al. Journal of Forest Research, 2014, 19(1): 34.
[21] Sibanda M, Mutanga O, Rouget M. GIScience & Remote Sensing，2016, 53, 614.
[22] Salum R B, Souza-Filho P, Simard M, et al. Estuarine, Coastal and Shelf Science, 2020, 236: 106585.
[23] Wang D, Wan B, Qiu P, et al. Remote Sensing, 2019, 11(18): 2156.
[24] Pandit S, Tsuyuki S, Dube T, Remote Sensing, 2018, 10(11)：1848.
[25] Zhao Q, Yu S, Zhao F, et al. Forest Ecology and Management, 2019, 434: 224.
[26] Pham L T H, Brabyn L. ISPRS Journal of Photogrammetry and Remote Sensing. 2017, 128：86.
[27] YANG Ming-xing, XU Tian-shu, NIU Xiao-hua, et al(杨明星, 徐天蜀, 牛晓花, 等). Journal of West China Forestry Science(西部林业科学), 2019, 48(2): 52.
[28] Chen T Q, Guestrin C. XGBoost: A Scalable Tree Boosting System, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016: 785.
[29] TIAN Yi-chao, HUANG Yuan-lin, TAO Jin, et al(田义超, 黄远林, 陶进, 等). Tropical Geography(热带地理), 2019, 39(4): 583.
[30] HU Yi-kai, XU Yao-wen, XUE Chun-quan, et al(胡懿凯, 徐耀文, 薛春泉, 等). Journal of South China Agricultural University(华南农业大学学报), 2019, 40(6): 95.
[31] Tam N F Y, Wong Y S, Lan C Y, et al. Hydrobiologia, 1995, 295(1-3): 193.
[32] Breiman L. Machine Learning, 2001, 45(1): 5.
[33] Huang J, Perry M. International Journal of Forecasting, 2016, 32(3): 1081.
[34] YANG Fu-qin, FENG Hai-kuan, LI Zhen-hai, et al(杨福芹, 冯海宽, 李振海, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2016, 32(3): 163.
[35] Alsumaiti T S. The Arab World Geographer, 2014, 17(2): 166.
[36] Pham T D, Yokoya N, Xia J, et al. Remote Sensing, 2020, 12：1334.
[37] LIU Xue-lian, OU Shao-long, LU Shuang-fei, et al(刘雪莲, 欧绍龙, 陆双飞, 等). Journal of West China Forestry Science(西部林业科学), 2020, 49(6): 128.
[38] Wang D, Wan B, Liu J, et al. International Journal of Applied Earth Observations and Geoinformation，2020, 85: 101986.
[39] PENG Cong-jiao, QIAN Jia-wei, GUO Xu-dong, et al(彭聪姣, 钱家炜, 郭旭东, 等). Chinese Journal of Applied Ecology(应用生态学报), 2016, 27(7): 2059.
[40] WANG Ren, LI Xiao-jing, CAI Jin-biao, et al(王韧, 李晓景, 蔡金标, 等). Journal of Southwest Forestry University(Natural Sciences)[西南林学院学报(自然科学版)], 2010, 30(1): 16.