Ecology and Environment ›› 2023, Vol. 32 ›› Issue (6): 1007-1015.DOI: 10.16258/j.cnki.1674-5906.2023.06.001
• Research Articles • Next Articles
WANG Xuemei1,2(), YANG Xuefeng1,2, ZHAO Feng1,2, AN Baisong1, HUANG Xiaoyu1
Received:
2023-01-28
Online:
2023-06-18
Published:
2023-09-01
王雪梅1,2(), 杨雪峰1,2, 赵枫1,2, 安柏耸1, 黄晓宇1
作者简介:
王雪梅(1976年生),女,教授,博士,硕士研究生导师,研究方向为干旱区资源环境遥感应用研究。E-mail: wangxm_1225@sina.com
基金资助:
CLC Number:
WANG Xuemei, YANG Xuefeng, ZHAO Feng, AN Baisong, HUANG Xiaoyu. Estimation of Aboveground Biomass in the Arid Oasis Based on the Machine Learning Algorithm[J]. Ecology and Environment, 2023, 32(6): 1007-1015.
王雪梅, 杨雪峰, 赵枫, 安柏耸, 黄晓宇. 基于机器学习算法的干旱区绿洲地上生物量估算[J]. 生态环境学报, 2023, 32(6): 1007-1015.
Add to citation manager EndNote|Ris|BibTeX
URL: https://www.jeesci.com/EN/10.16258/j.cnki.1674-5906.2023.06.001
样本类型 | 样本数 | 地上生物量/(g·m-2) | 变异系数/ % | |||
---|---|---|---|---|---|---|
最大值 | 最小值 | 平均值 | 标准差 | |||
总体样本 | 94 | 1448.5 | 7.4 | 387.9 | 319.4 | 82.3 |
训练样本 | 64 | 1448.5 | 7.4 | 402.9 | 329.1 | 81.7 |
验证样本 | 30 | 1387.8 | 12.8 | 355.8 | 300.6 | 84.5 |
Table 1 Basic statistical characteristics of aboveground biomass in various sites
样本类型 | 样本数 | 地上生物量/(g·m-2) | 变异系数/ % | |||
---|---|---|---|---|---|---|
最大值 | 最小值 | 平均值 | 标准差 | |||
总体样本 | 94 | 1448.5 | 7.4 | 387.9 | 319.4 | 82.3 |
训练样本 | 64 | 1448.5 | 7.4 | 402.9 | 329.1 | 81.7 |
验证样本 | 30 | 1387.8 | 12.8 | 355.8 | 300.6 | 84.5 |
反演模型 | 变量组合 | 训练集 | 验证集 | |||||||
---|---|---|---|---|---|---|---|---|---|---|
R2 | σMAE/(g·m-2) | σRMSE/(g·m-2) | σRPD | R2 | σMAE/(g·m-2) | σRMSE/(g·m-2) | σRPD | |||
SVM | TV | 0.693 | 113.5 | 195.7 | 1.78 | 0.630 | 89.2 | 151.7 | 1.67 | |
IV | 0.683 | 125.3 | 198.1 | 1.75 | 0.616 | 90.3 | 156.4 | 1.62 | ||
BV | 0.695 | 117.7 | 194.4 | 1.79 | 0.642 | 88.1 | 149.1 | 1.70 | ||
PV | 0.691 | 125.7 | 197.1 | 1.76 | 0.623 | 89.7 | 153.7 | 1.65 | ||
BPNN | TV | 0.844 | 101.2 | 139.5 | 2.49 | 0.571 | 99.1 | 163.5 | 1.55 | |
IV | 0.802 | 111.9 | 156.6 | 2.22 | 0.541 | 110.5 | 171.7 | 1.48 | ||
BV | 0.836 | 102.1 | 142.3 | 2.44 | 0.542 | 100.4 | 168.8 | 1.50 | ||
PV | 0.841 | 96.3 | 139.7 | 2.48 | 0.617 | 98.3 | 155.4 | 1.63 | ||
XGBoost | TV | 0.731 | 118.4 | 178.8 | 1.94 | 0.665 | 116.1 | 151.6 | 1.67 | |
IV | 0.730 | 129.9 | 186.0 | 1.87 | 0.648 | 93.3 | 152.1 | 1.67 | ||
BV | 0.754 | 134.2 | 171.0 | 2.03 | 0.670 | 101.9 | 146.0 | 1.74 | ||
PV | 0.742 | 137.2 | 177.2 | 1.96 | 0.719 | 100.0 | 133.0 | 1.91 | ||
RF | TV | 0.885 | 84.3 | 118.7 | 2.93 | 0.716 | 83.7 | 135.2 | 1.87 | |
IV | 0.888 | 83..2 | 119.8 | 2.90 | 0.711 | 92.7 | 139.2 | 1.82 | ||
BV | 0.898 | 82.1 | 110.8 | 3.14 | 0.742 | 79.2 | 132.1 | 1.92 | ||
PV | 0.894 | 79.7 | 113.1 | 3.07 | 0.775 | 92.6 | 133.4 | 1.90 |
Table 2 Estimation accuracy of different inversion models
反演模型 | 变量组合 | 训练集 | 验证集 | |||||||
---|---|---|---|---|---|---|---|---|---|---|
R2 | σMAE/(g·m-2) | σRMSE/(g·m-2) | σRPD | R2 | σMAE/(g·m-2) | σRMSE/(g·m-2) | σRPD | |||
SVM | TV | 0.693 | 113.5 | 195.7 | 1.78 | 0.630 | 89.2 | 151.7 | 1.67 | |
IV | 0.683 | 125.3 | 198.1 | 1.75 | 0.616 | 90.3 | 156.4 | 1.62 | ||
BV | 0.695 | 117.7 | 194.4 | 1.79 | 0.642 | 88.1 | 149.1 | 1.70 | ||
PV | 0.691 | 125.7 | 197.1 | 1.76 | 0.623 | 89.7 | 153.7 | 1.65 | ||
BPNN | TV | 0.844 | 101.2 | 139.5 | 2.49 | 0.571 | 99.1 | 163.5 | 1.55 | |
IV | 0.802 | 111.9 | 156.6 | 2.22 | 0.541 | 110.5 | 171.7 | 1.48 | ||
BV | 0.836 | 102.1 | 142.3 | 2.44 | 0.542 | 100.4 | 168.8 | 1.50 | ||
PV | 0.841 | 96.3 | 139.7 | 2.48 | 0.617 | 98.3 | 155.4 | 1.63 | ||
XGBoost | TV | 0.731 | 118.4 | 178.8 | 1.94 | 0.665 | 116.1 | 151.6 | 1.67 | |
IV | 0.730 | 129.9 | 186.0 | 1.87 | 0.648 | 93.3 | 152.1 | 1.67 | ||
BV | 0.754 | 134.2 | 171.0 | 2.03 | 0.670 | 101.9 | 146.0 | 1.74 | ||
PV | 0.742 | 137.2 | 177.2 | 1.96 | 0.719 | 100.0 | 133.0 | 1.91 | ||
RF | TV | 0.885 | 84.3 | 118.7 | 2.93 | 0.716 | 83.7 | 135.2 | 1.87 | |
IV | 0.888 | 83..2 | 119.8 | 2.90 | 0.711 | 92.7 | 139.2 | 1.82 | ||
BV | 0.898 | 82.1 | 110.8 | 3.14 | 0.742 | 79.2 | 132.1 | 1.92 | ||
PV | 0.894 | 79.7 | 113.1 | 3.07 | 0.775 | 92.6 | 133.4 | 1.90 |
[1] |
BYRNE K M, LAUENROTH W K, ADLER P B, et al., 2011. Estimating aboveground net primary production in grasslands: a comparison of nondestructive methods[J]. Rangeland Ecology & Management, 64(5): 498-505.
DOI URL |
[2] |
CHING P M L, ZOU X, WU D, et al., 2022. Development of a wide-range soft sensor for predicting wastewater BOD5 using an eXtreme gradient boosting (XGBoost) machine[J]. Environmental Research, 210(4-5): 112953.
DOI URL |
[3] | COHEN A R, CHEN G, BERGER E M, et al., 2021. Dynamically controlled environment agriculture: integrating machine learning and mechanistic and physiological models for sustainable food cultivation[J]. ACS ES&T Engineering, 2(1): 3-19. |
[4] |
GHOSH A K, DAS B S, REDDY N, 2020. Application of VIS-NIR spectroscopy for estimation of soil organic carbon using different spectral preprocessing techniques and multivariate methods in the middle Indo-Gangetic plains of India[J]. Geoderma Regional, 23(3-4): e00349.
DOI URL |
[5] |
GITELSON A A, KAUFMAN Y J, MERZLYAK M N, 1996. Use of a green channel in remote sensing of global vegetation from EOS-MODIS[J]. Remote Sensing of Environment, 58(3): 289-298.
DOI URL |
[6] |
HUANG C D, YE X Y, DENG C B, et al., 2016. Mapping above-ground biomass by integrating optical and SAR imagery: A case study of Xixi National Wetland Park, China[J]. Remote Sensing, 8(8): 647.
DOI URL |
[7] |
KAUFMAN Y J, TANRE D, 1992. Atmospherically resistant vegetation index(ARVI) for EOS-MODIS[J]. IEEE Transaction on Geoscience and Remote Sensing, 30(2): 261-270.
DOI URL |
[8] | KHAN R W A, SHAHEEN H, 2022. Biomass carbon stock estimation in lesser Himalayan subtropical broadleaf forests of Kashmir[J]. Taiwania, 67(1): 47-54. |
[9] |
LI C H, ZHOU L Z, XU W B, 2021. Estimating aboveground biomass using Sentinel-2 MSI data and ensemble algorithms for grassland in the Shengjin Lake Wetland, China[J]. Remote Sensing, 13(8): 1595.
DOI URL |
[10] |
MORAIS T G, TEIXEIRA R F M, FIGUEIREDO M, et al., 2021. The use of machine learning methods to estimate aboveground biomass of grasslands: A review[J]. Ecological Indicators, 130(3): 108081.
DOI URL |
[11] |
NESHA M K, HUSSIN Y A, VAN LEEUWEN L M, et al., 2020. Modeling and mapping aboveground biomass of the restored mangroves using ALOS-2 PALSAR-2 in East Kalimantan, Indonesia[J]. International Journal of Applied Earth Observation and Geoinformation, 91: 102158.
DOI URL |
[12] |
QI J, CHEHBOUNI A L, HUETE A R, et al., 1994. A modified soil adjusted vegetation index (MSAVI)[J]. Remote Sensing of Environment, 48(2): 119-126.
DOI URL |
[13] |
SHARMA P, LEIGH L, CHANG J, et al., 2022. Above-ground biomass estimation in Oats using UAV remote sensing and machine learning[J]. Sensors, 22(2): 601.
DOI URL |
[14] | SI M X, DU K, 2020. Development of a predictive emissions model using a gradient boosting machine learning method[J]. Environmental Technology & Innovation, 20: 101028. |
[15] | SUN L N, WANG M B, FAN X H, 2020. Spatial pattern and driving factors of biomass carbon density for natural and planted coniferous forests in mountainous terrain, eastern Loess Plateau of China[J]. Forest Ecosystems, 7(1): 104-116. |
[16] |
SUN W L, LIU X H, 2020. Review on carbon storage estimation of forest ecosystem and applications in China[J]. Forest Ecosystems, 7(1): 1-14.
DOI |
[17] | TAPPAYUTHPIJARN K, VINDEVOGEL B S, 2022. High-accuracy Machine Learning Models to Estimate above Ground Biomass over Tropical Closed Evergreen Forest Areas from Satellite Data[C]// IOP Conference Series: Earth and Environmental Science. IOP Publishing, 1006(1): 012001. |
[18] |
WANG X Y, LIU C X, LÜ G T, et al., 2022. Integrating Multi-Source Remote Sensing to Assess Forest Aboveground Biomass in the Khingan Mountains of North-Eastern China Using Machine-Learning Algorithms[J]. Remote Sensing, 14(4): 1039.
DOI URL |
[19] |
WONGCHAI W, ONSREE T, SUKKAM N, et al., 2022. Machine learning models for estimating above ground biomass of fast growing trees[J]. Expert Systems with Applications, 199(7): 117186.
DOI URL |
[20] |
YUAN Z Q, FANG C, ZHANG R, et al., 2019. Topographic influences on soil properties and aboveground biomass in lucerne-rich vegetation in a semi-arid environment[J]. Geoderma, 344: 137-143.
DOI URL |
[21] |
ZHOU Y J, LIU T X, BATELAAN O, et al., 2023. Spatiotemporal fusion of multi-source remote sensing data for estimating aboveground biomass of grassland[J]. Ecological Indicators, 146: 109892.
DOI URL |
[22] |
党晓宏, 高永, 蒙仲举, 等, 2017. 西鄂尔多斯地区5种荒漠优势灌丛生物量分配格局及预测模型[J]. 中国沙漠, 37(1): 100-108.
DOI |
DANG X H, GAO Y, MENG Z J, et al., 2017. Biomass allocation pattern sand estimation model of five desert shrub species in west Ordos region[J]. Journal of Desert Research, 37(1): 100-108. | |
[23] | 丁家祺, 黄文丽, 刘迎春, 等, 2021. 基于机器学习和多源数据的湘西北森林地上生物量估测[J]. 林业科学, 57(10): 36-48. |
DING J Q, HUANG W L, LIU Y C, et al., 2021. Estimation of forest aboveground biomass in northwest Hunan province based on machine learning and multi-source data[J]. Scientia Silvae Sinicae, 57(10): 36-48. | |
[24] | 方精云, 2000. 北半球中高纬度的森林碳库可能远小于目前的估算[J]. 植物生态学报, 24(5): 635-638. |
FANG J Y, 2000. Forest biomass carbon pool of middle and high latitudes in the north hemisphere is probably much smaller than present estimates[J]. Chinese Journal of Plant Ecology, 24(5): 635-638. | |
[25] | 李妙宇, 上官周平, 邓蕾, 2021. 黄土高原地区生态系统碳储量空间分布及其影响因素[J]. 生态学报, 41(17): 6786-6799. |
LI M Y, SHANGGUAN Z P, DENG L, 2021. Spatial distribution of carbon storages in the terrestrial ecosystems and its influencing factors on the Loess Plateau[J]. Acta Ecologica Sinica, 41(17): 6786-6799. | |
[26] |
陶冶, 张元明, 2013. 荒漠灌木生物量多尺度估测——以梭梭为例[J]. 草业学报, 22(6): 1-10.
DOI |
TAO Y, ZHANG Y M, 2013. Multi-scale biomass estimation of desert shrubs: A case study of Haloxylon ammodendron[J]. Acta Prataculturae Sinica, 22(6): 1-10.
DOI |
|
[27] |
童新风, 杨红玲, 宁志英, 等, 2018. 科尔沁沙地优势固沙灌木的生物量预测模型[J]. 中国沙漠, 38(3): 553-559.
DOI |
TONG X F, YANG H L, NING Z Y, et al., 2018. Biomass estimation models for dominant sand-fixing shrubs in Horqin sand land[J]. Journal of Desert Research, 38(3): 553-559.
DOI |
|
[28] |
邢晓语, 杨秀春, 徐斌, 等, 2021. 基于随机森林算法的草原地上生物量遥感估算方法研究[J]. 地球信息科学学报, 23(7): 1312-1324.
DOI |
XING X Y, YANG X C, XU B, et al., 2021. Remote sensing estimation of grassland aboveground biomass based on random forest[J]. Journal of Geo-information Science, 23(7): 1312-1324. | |
[29] | 薛海连, 田相林, 曹田健, 2021. 利用经验-过程混合建模方法优化华山松过程模型的参数[J]. 林业科学, 57(9): 21-33. |
XUE H L, TIAN X L, CAO T J, 2021. Optimizing parameters of a process based model for Pinus armandii: A compromise between empirical and process based modelling approaches[J]. Scientia Silvae Sinicae, 57(9): 21-33. | |
[30] |
杨宪龙, 魏孝荣, 邵明安, 2016. 黄土高原北部典型灌丛枝条生物量估算模型[J]. 应用生态学报, 27(10): 3164-3172.
DOI |
YANG X L, WEI X R, SHAO M A, 2016. Stem biomass estimation models for dominant shrubs on the northern Loess Plateau of China[J]. Chinese Journal of Applied Ecology, 27(10): 3164-3172. | |
[31] | 姚雪玲, 姜丽娜, 李龙, 等, 2019. 浑善达克沙地6种灌木生物量模拟[J]. 生态学报, 39(3): 905-912. |
YAO X L, JIANG L N, LI L, et al., 2019. Biomass simulation of six shrub species in Otindag sandy land[J]. Acta Ecologica Sinica, 39(3): 905-912. | |
[32] |
张殿岱, 王雪梅, 昝梅, 2021. 基于Landsat 8 OLI影像的渭-库绿洲植被地上生物量估算[J]. 草业学报, 30(11): 1-12.
DOI |
ZHANG D D, WANG X M, ZAN M, 2021. Estimation of vegetation aboveground biomass in the Wei-Ku Oasis based on Landsat 8 OLI images[J]. Acta Prataculturae Sinica, 30(11):1-12. | |
[33] | 张华, 张玉红, 张改改, 2020. 民勤绿洲青土湖植被优势种地上生物量估算[J]. 干旱区地理, 43(1): 201-210. |
ZHANG H, ZHANG Y H, ZHANG G G, 2020. Aboveground biomass estimation of the dominant species of vegetation in the Qingtu Lake at Minqin Qasis[J]. Arid Land Geography, 43(1): 201-210. | |
[34] |
张亦然, 刘廷玺, 童新, 等, 2021. 基于XGBoost算法的草甸地上生物量的高光谱遥感反演[J]. 草业学报, 30(4): 1-12.
DOI |
ZHANGY R, LIU T X, TONG X, et al., 2021. Hyperspectral remote sensing inversion of meadow aboveground biomass based on an XGBoost algorithm[J]. Acta Prataculturae Sinica, 30(4): 1-12. |
[1] | RUAN Huihua, XU Jianhui, ZHANG Feifei. Spatiotemporal Changes of Vegetation and Land Surface Temperature during 2001 and 2020 in the Guangdong-Hong Kong-Macao Greater Bay Area of China [J]. Ecology and Environment, 2022, 31(8): 1510-1520. |
[2] | CHEN Wenyu, XIA Lihua, XU Guoliang, YU Shiqin, CHEN Hang, CHEN Jinfeng. Dynamic Variation of NDVI and Its Influencing Factors in the Pearl River Basin from 2000 to 2020 [J]. Ecology and Environment, 2022, 31(7): 1306-1316. |
[3] | NIE Tong, DONG Guotao, JIANG Xiaohui, GUO Xinwei, DANG Suzhen, ZHENG Jiahao, LI Lichan, WANG Jiang. Spatio-temporal Variations and Influencing Factors of Vegetation in Yulin [J]. Ecology and Environment, 2022, 31(1): 26-36. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||
Copyright © 2021 Editorial Office of ACTA PETROLEI SINICA
Address:No. 6 Liupukang Street, Xicheng District, Beijing, P.R.China, 510650
Tel: 86-010-62067128, 86-010-62067137, 86-010-62067139
Fax: 86-10-62067130
Email: syxb@cnpc.com.cn
Support byBeijing Magtech Co.ltd, E-mail:support@magtech.com.cn