Random forest prediction model for the soil organic matter with optimized spectral inputs

ZHANG Xiaohan; MENG Xiangtian; TANG Haitao; LIU Huanjun; ZHANG Xinle; LIU Qiong

doi:10.11975/j.issn.1002-6819.202207035

Volume 39 Issue 2

Jan. 2023

Turn off MathJax

Article Contents

Abstract

References

Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE) > 2023 > 39(2): 90-99. > DOI: 10.11975/j.issn.1002-6819.202207035

ZHANG Xiaohan, MENG Xiangtian, TANG Haitao, LIU Huanjun, ZHANG Xinle, LIU Qiong. Random forest prediction model for the soil organic matter with optimized spectral inputs[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2023, 39(2): 90-99. DOI: 10.11975/j.issn.1002-6819.202207035

Citation:

PDF (2589 KB)

Random forest prediction model for the soil organic matter with optimized spectral inputs

1.
School of Public Administration and Law, Northeast Agricultural University, Harbin 150030, China
2.
Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun 130012, China
3.
College of Information Technology Jilin Agricultural University, Changchun 130118, China

More Information

Received Date: July 03, 2022
Revised Date: October 09, 2022
Published Date: January 30, 2023

Graphical Abstract

Abstract

Abstract

Abstract: Soil organic matter (SOM) is one of the most important parts of the soil carbon pool. The carbon-containing organic matter in soil mainly includes animal and plant residues, microorganisms, and various organic matter decomposed or synthesized in agriculture. Among them, the SOM content is one of the important indicators to measure the soil fertility level. Accurate measurement of SOM is of great significance for soil fertility evaluation, environmental protection, agricultural and forestry development. Therefore, accurate prediction of SOM content is extremely important so far. Previous research on the SOM prediction of random forest (RF) usually only uses one spectral input without considering the complementarity between different spectral inputs. Therefore, it is necessary to select an appropriate method for the noise reduction of the reflection spectrum, in order to reduce the influence of spectral noise. Among them, discrete wavelet transform can be used to reduce high spectral noise, while preserving the effective information for the SOM prediction. In this study, the combination of different spectral inputs and discrete wavelet transform was used to predict the SOM with the optimized spectral input using RF. The stochastic forest model was also used to predict the SOM. Firstly, the original spectral reflectance of 204 soil samples from Baoqing County was analyzed using discrete wavelet transform. Secondly, the spectral characteristic parameters and principal components were extracted from the decomposed characteristic spectral curves, in order to construct the spectral indices. Finally, the three spectral inputs were substituted into the RF model to explore the optimal combination of spectral inputs for the SOM prediction. Meanwhile, the variation trend of different spectral inputs was obtained under different wavelet decomposition scales, in order to provide a new idea for the selection of spectral inputs for the SOM hyperspectral prediction. The RF model was better to predict the SOM in this case. The optimal combination of different spectral information was obtained to predict the organic matter and the optimal decomposition scale of the discrete wavelet transform. Finally, the combination with the highest accuracy was obtained among all the inputs at all decomposition scales. The results show that: 1) The accuracy of SOM prediction under different spectral inputs was higher than that of direct spectral reflectance modeling. The highest verification accuracy of the principal component in the single spectral index, similar to the combination of spectral characteristic parameters and principal component, was higher than that of the principal component modeling alone, indicating that combining different spectral inputs improved the prediction accuracy. However, simply stacking spectral inputs was not enough to improve the prediction accuracy. 2) There was also a different variation trend of prediction accuracy of different spectral inputs, with the increase of decomposition scale. The variation trend of prediction accuracy of different spectral input combinations was changed with the different spectral inputs in the combination, indicating the variation characteristics of spectral inputs. 3) The highest verification accuracy was found in the combination of the spectral characteristic parameters and principal components with the decomposition scale of 6, R2 reaching 0.78, and RMSE reaching 1.32%, indicating an excellent prediction ability. Anyway, it is feasible to predict the organic matter using the spectral input combined with discrete wavelet transform modeling. The finding can provide a reliable idea and theoretical support for the dynamic monitoring of SOM under temporal and spatial changes.
- remote sensing,
- soils,
- organic matter,
- random forest,
- spectral input,
- discrete wavelet transform

FullText(HTML)

References (36)

References

[1]	刘焕军，赵春江，王纪华，等. 黑土典型区土壤有机质遥感预测[J]. 农业工程学报，2011，27(8)：211-215.LIU Huanjun, ZHAO Chunjiang, WANG Jihua, et al. Remote sensing of soil organic matter in typical black soil region[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2011, 27(8): 211-215. (in Chinese with English abstract)
[2]	GU X H, WANG Y C, SUN Q, et al. Hyperspectral inversion of soil organic matter content in cultivated land based on wavelet transform[J]. Computers and Electronics in Agriculture, 2019, 167: 105053.
[3]	NOWKANDEH S M,NOROOZI A A,HOMAEE M, Estimating soil organic matter content from Hyperion reflectance images using PLSR, PCR, MinR and SWR models in semi-arid regions of Iran[J].Environmental Development,2018,2523: 32.
[4]	玉米提·买明，王雪梅. 连续小波变换的土壤有机质含量高光谱估测[J]. 光谱学与光谱分析，2022，42(4)：1278-1284.YUMITI Maiming, WANG Xuemei. Hyperspectral estimation of soil organic matter content by continuous wavelet transform[J]. Spectroscopy and Spectral Analysis, 2022, 42(4): 1278-1284. (in Chinese with English abstract)
[5]	张子鹏，丁建丽，王敬哲，等. 利用三维光谱指数定量估算土壤有机质含量：以新疆艾比湖流域为例[J]. 光谱学与光谱分析，2020，40(5)：1514-1522.ZHANG Zipeng, DING Jianli, WANG Jingzhe, et al. Quantitative estimation of soil organic matter content using three-dimensional spectral index: A case study of the Ebinur Lake Basin in Xinjiang[J]. Spectroscopy and Spectral Analysis, 2020, 40(5): 1514-1522. (in Chinese with English abstract)
[6]	陆龙妹，张平，卢宏亮，等. 淮北平原土壤高光谱特征及有机质含量预测[J]. 土壤，2019，51(2)：374-380.LU Longmei, ZHANG ping, LU Hongliang, et al. Hyperspectral characteristics of soils in Huaibei Plain and estimation of SOM content[J]. Soils, 2019, 51(2): 374-380. (in Chinese with English abstract)
[7]	陈玮，徐占军，郭琦. 煤炭矿区耕地土壤有机质无人机高光谱遥感估测[J]. 农业工程学报，2022，38(8)：98-106.CHEN Wei, XU Zhanjun, GUO Qi. Estimation of soil organic matter by UAV hyperspectral remote sensing in coal mining areas[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2022, 38(8): 98-106. (in Chinese with English abstract)
[8]	张娟娟，席磊，杨向阳，等. 砂姜黑土有机质含量高光谱估测模型构建[J]. 农业工程学报，2020，36(17)：135-141.ZHANG Juanjuan, XI Lei, YANG Xiangyang, et al. Construction of hyperspectral estimation model for organic matter content in Shajiang black soil[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(17): 135-141. (in Chinese with English abstract)
[9]	朱传梅，王宏卫，谢霞，等. 基于光谱指数和机器学习的土壤有机质含量反演[J]. 江苏农业科学，2020，48(22)：233-241.ZHU Chuanmei, WANG Hongwei, XIE Xia, et al. Retrieval of soil organic matter content based on spectral index and machine learning[J]. Jiangsu Agricultural Sciences, 2020, 48(22): 233-241. (in Chinese with English abstract)
[10]	唐海涛，孟祥添，苏循新，等. 基于CARS算法的不同类型土壤有机质高光谱预测[J]. 农业工程学报，2021，37(2)：105-113.TANG Haitao, MENG Xiangtian, SU Xunxin, et al. Hyperspectral prediction of soil organic matter based on CARS algorithm [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(2): 105-113. (in Chinese with English abstract)
[11]	金慧凝，张新乐，刘焕军，等. 基于光谱吸收特征的土壤含水量预测模型研究[J]. 土壤学报，2016，53(3)：627-635.JIN Huining, ZHANG Xinle, LIU Huanjun, et al. Prediction model of soil water content based on spectral absorption characteristics [J]. Acta Pedologica Sinica, 2016, 53(3): 627-635. (in Chinese with English abstract)
[12]	叶勤，姜雪芹，李西灿，等. 基于高光谱数据的土壤有机质含量反演模型比较[J]. 农业机械学报，2017，48(3)：164-172.YE Qin, JIANG Xueqin, LI Xican, et al. Comparison of inversion models of soil organic matter content based on hyperspectral data [J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(3): 164-172. (in Chinese with English abstract)
[13]	李粉玲，王力，刘京，等. 基于高分一号卫星数据的冬小麦叶片 SPAD 值遥感估算[J]. 农业机械学报，2015，46(9)：278-286.LI Fenling, WANG Li, LIU Jing, et al. Remote sensing estimation of SPAD value for wheat leaf based on GF-1 data[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(9): 278-286. (in Chinese with English abstract)
[14]	BAO Y L, MENG X T, Ustin S, et al. Vis-SWIR spectral prediction model for soil organic matter with different grouping strategies[J]. Catena, 2020, 195: 104703.
[15]	郑淼，王翔，李思佳，等. 黑土区土壤有机质和全氮含量遥感反演研究[J]. 地理科学，2022，42(8)：1336-1347.ZHENG Miao, WANG Xiang, LI Sijia, et al. Remote sensing inversion of soil organic matter and total nitrogen content in black soil area [J]. Scientia Geographica Sinica, 2022, 42(8): 1336-1347. (in Chinese with English abstract)
[16]	郭静，龙慧灵，何津，等. 基于Google Earth Engine和机器学习的耕地土壤有机质含量预测[J]. 农业工程学报，2022，38(18)：130-137.GUO Jing, LONG Huiling, HE Jin, et al. Prediction of soil organic matter content based on Google Earth Engine and machine Learning [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2022, 38(18): 130-137. (in Chinese with English abstract)
[17]	刘焕军，鲍依临，孟祥添，等. 不同降噪方式下基于高分五号影像的土壤有机质反演[J]. 农业工程学报，2020，36(12)：90-98.LIU Huanjun, BAO Yilin, MENG Xiangtian, et al. Inversion of soil organic matter with different noise reduction methods based on Gaofen-5 image [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(12): 90-98. (in Chinese with English abstract).
[18]	中国土壤学会. 土壤农业化学分析方法[M]. 北京：中国农业科技出版社，2000.
[19]	ZHANG X K, LIU H J, ZHANG X L, et al. Allocate soil individuals to soil classes with topsoil spectral characteristics and decision trees[J]. Geoderma, 2018, 320: 12-22.
[20]	史舟，王乾龙，彭杰，等. 中国主要土壤高光谱反射特性分类与有机质光谱预测模型[J]. 中国科学：地球科学，2014，44(5)：978-988.SHI Zhou, WANG Ganlong, PENG Jie, et al. Classification of soil hyperspectral reflectance characteristics and spectral prediction model of organic matter in China [J]. Science China Earth Sciences, 2014, 44(5): 978-988. (in Chinese with English abstract)
[21]	蔡亮红，丁建丽. 小波变换耦合CARS算法提高土壤水分含量高光谱反演精度[J]. 农业工程学报，2017，33(16)：144-151.CAI Lianghong, DING Jianli. Wavelet transform coupled with CARS algorithm to improve the accuracy of soil water content hyperspectral inversion[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(16): 144-151. (in Chinese with English abstract).
[22]	蔡振宇，盖增喜. 人工智能在拾取地震P波初至中的应用：以汶川地震余震序列为例[J]. 北京大学学报(自然科学版)，2019，55(3)：451-460.CAI Zhenyu, GAI Zengxi. Application of Artificial Intelligence in picking up initial P-wave arrival of earthquake: A case study of Wenchuan Earthquake aftershock sequence[J]. Journal of Peking University (Natural Science Edition), 2019, 55(3): 451-460. (in Chinese with English abstract)
[23]	MENG X T, BAO Y L, LIU J G, et al. Regional soil organic carbon prediction model based on a discrete wavelet analysis of hyperspectral satellite data[J]. International Journal of Applied Earth Observation and Geoinformation, 2020, 89: 102111.
[24]	XU X B, LI M Z, TANG Z Y, et al. Echo signal extraction method of laser radar based on improved singular value decomposition and wavelet threshold denoising[J]. Infrared Physics & Technology, 2018, 92: 327-335.
[25]	HONG Y S, CHEN S C, ZHANG Y, et al. Rapid identification of soil organic matter level via visible and near-infrared spectroscopy: Effects of two-dimensional correlation coefficient and extreme learning machine[J]. Science of the Total Environment, 2018, 644: 1232-1243.
[26]	IHUOMA S O,MADRAMOOTOO C A. Narrow-band reflectance indices for mapping the combined effects of water and nitrogen stress in field grown tomato crops[J]. Biosystems Engineering, 2020, 192: 133-143.
[27]	郑曼迪，熊黑钢，乔娟峰，等. 基于宽波段与窄波段综合光谱指数的土壤有机质遥感预测[J]. 激光与光电子学进展，2018，55(7)：457-465.ZHEN Mandi, XIONG Heigang, QIAO Juanfeng, et al. Remote sensing inversion of soil organic matter based on broad band and narrow band comprehensive spectral index[J]. Laser & Optoelectronics Progress, 2018, 55(7): 457-465. (in Chinese with English abstract).
[28]	JIN X L, SONG K S, DU J, et al. Comparison of different satellite bands and vegetation indices for estimation of soil organic matter based on simulated spectral configuration[J]. Agricultural and Forest Meteorology, 2017, 244/245: 57-71.
[29]	丰明博，牛铮，孙刚. 多波段激光雷达植被光谱分析[J]. 光谱学与光谱分析，2017，37(6)：1809-1813.FENG Mingbo, NIU Zheng, SUN Gang. Spectral analysis of multi-band lidar vegetation[J]. Spectroscopy and spectral analysis, 2017, 37(6): 1809-1813. (in Chinese with English abstract)
[30]	于仙毅，巫江虹，高云辉. 基于主成分分析与支持向量机的热泵系统制冷剂泄漏识别研究[J]. 化工学报，2020，71(7)：3151-3164.YU Xianyi, WU Jianghong, GAO Yunhui. Research on refrigerant leakage identification of heat pump system based on principal component analysis and support vector machine[J]. Journal of Chemical Industry and Engineering, 2020, 71(7): 3151-3164. (in Chinese with English abstract)
[31]	CLARK R N, ROUSH T L. Reflectance spectroscopy: Quantitative analysis techniques for remote sensing application[J]. Journal of Geophysical Research, 1984, 89: 6329-6340.
[32]	刘焕军，孟祥添，王翔，等. 反射光谱特征的土壤分类模型[J]. 光谱学与光谱分析，2019，39(8)：2481-2485.LIU Huanjun, MENG Xiangtian, WANG Xiang, et al. Soil Classification Model of Reflective Spectroscopic Characteristics[J]. Spectroscopy and Spectral Analysis, 2019, 39(8): 2481-2485. (in Chinese with English abstract)
[33]	包青岭，丁建丽，王敬哲，等. 基于随机森林算法的土壤有机质含量高光谱检测[J]. 干旱区地理，2019，42(6)：1404-1414.BAO Qingling, DING Jianli, WANG Jingzhe, et al. Hyperspectral detection of soil organic matter content based on random forest algorithm[J]. Arid Land Geography, 2019, 42(6): 1404-1414. (in Chinese with English abstract)
[34]	姜赛平，张怀志，张认连，等. 基于三种空间预测模型的海南岛土壤有机质空间分布研究[J]. 土壤学报，2018，55(4)：1007-1017.JIANG Saiping, ZHANG Huaizhi, ZHANG Renlian, et al. Research on spatial distribution of soil organic matter in Hainan Island based on three spatial prediction models[J]. Acta Pedologica Sinica, 2018, 55(4): 1007-1017. (in Chinese with English abstract)
[35]	徐夕博，吕建树，吴泉源，等. 基于PCA-MLR和PCA-BPN的莱州湾南岸滨海平原土壤有机质高光谱预测研究[J]. 光谱学与光谱分析，2018，38(8)：2556-2562.XU Xibo, LV Jianshu, WU Quanyuan, , et al. Prediction of soil organic matter hyperspectral based on PCA-MLR and PCA-BPN in the coastal plain south of Laizhou Bay[J]. Spectroscopy and Spectral Analysis, 2018, 38(8): 2556-2562. (in Chinese with English abstract)
[36]	王永敏，田林亚，李西灿，等. 基于小波与包络线的土壤有机质高光谱估测[J]. 地理信息世界，2018，25(4)：36-41.WANG Yongmin, TIAN Linya, LI Xican, et al. Hyperspectral estimation of soil organic matter based on wavelet and envelope[J]. Geographical Information World, 2018, 25(4): 36-41. (in Chinese with English abstract)

Cited By

Cited by

Periodical cited type(6)

1.	李艾雯，李文丹，宋靓颖，冉敏，陈丹，成金礼，齐浩然，郭聪慧，李启权. 四川盆地耕地表层土壤容重缺失数据填补方法. 土壤学报. 2025(01): 40-53 .
2.	丁启东，王怡婧，张俊华，贾科利，黄华雨. 利用CARS算法联合协变量估算盐碱农田土壤水分和有机质含量. 应用生态学报. 2024(05): 1321-1330 .
3.	张博，罗维平. 基于机器学习的奶牛饲料消耗状态预测模型. 农业工程学报. 2024(09): 164-172 . 本站查看
4.	王世芳，宋海燕. 土壤有机质可见-近红外反射光谱特性研究. 中国农业科技导报. 2024(07): 183-188 .
5.	刘雅璇，于慧，罗勇. 基于辅助变量的紫色土耕地土壤有机质空间预测. 土壤. 2024(04): 857-865 .
6.	江振蓝，陈付勋，沙晋明，罗双飞，罗烨琴. 基于组合光谱输入量的土壤全氮反演模型. 闽江学院学报. 2024(05): 117-128 .